Problem/Motivation

The support experience on Drupal.org has not received focused development attention in many years. Users find it difficult to find high-quality support on Drupal.org, since there is a large divide between where most developers work (issue queues, working groups on g.d.o, #drupal-contribute on IRC) and where end users seek support (forums, the support/development mailing lists, #drupal/#drupal-support on IRC).

In response to this, several community members rallied to get Drupal Stack Exchange (Drupal.SE) created a few months back. Stack Exchange is a free SaaS platform to create technology-specific clones of Stack Overflow. The Stack Exchange platform provides many nice features around support, including focused Q&A style postings, rather than rambling discussions, an easy way to mark a support request as answered and the answers themselves as more or less valuable than others, and a robust gamification/karma system that encourages participation. Having just launched as a full-fledged Stack Exchange site site, the uptake of Drupal.SE during its beta has been popular. However, as a third-party service, it lacks integration with the Drupal.org website.

The question posed is whether to improve the support tools here on Drupal.org, or to officially recommend http://drupal.stackexchange.com/ as our support outlet.

Proposed resolution

There are two proposed resolutions:

  1. Endorse http://drupal.stackexchange.com/ as the official venue for Drupal support.
  2. Develop a home-grown alternative at http://support.drupal.org which would keep user support on site.

Regarding Drupal SE

  • Kill d.o forums and point every general support request to Drupal SE
  • Not endorsing Drupal SE
  • Endorsing Drupal SE and keep forums too

In any case we also need to solve the trademark issues.

Regarding improving support on D.O.

  • Organise a volunteer-driven effort to develop an improved support experience.
  • Approach the Drupal Association and ask them to fund the initiative.
  • Some combination of the above

Pros and cons

Option Pros Cons
Drupal.StackExchange.com
  • Very good Q&A platform, good experience for support seekers as well as support givers
  • We don't maintain it
  • Available now
  • Free
  • We gain visibility from other sites of the stackexchange network (?)
  • Not in drupal.org : risk of losing contribution
  • means killing drupal.org forums (when they're still used)
  • Trademark issues
  • Corporate, we don't know how it will turn in the future
  • No integration with drupal.org
  • Data will not belong to d.o, though it is CC licensed and available
  • No 1:1 feature match with the current forums
  • No single sign on, although OpenID is an option
  • Possible future migration concerns if we do ever decide to bring it back "in house", although we likely wont.
  • Division of user accesses looking for support. Some would go to issue queues, some would go to mailing archives, some would go to SE.
Improved support on Drupal.org
  • Tighter integration of support data with other D.O. systems
  • Access to Q&A for a project right from the project page?
  • Ability to use other metrics we have access to to influence a person's reputation
  • Single sign on for support
  • The benefits of improving the code infrastructure comes back to the community.
  • We could potentially migrate current forum posts into it
  • Possibly allow Issues to 'become' questions, or vice versa
  • We can tag questions / answers with more specific / structured meta data
  • We're seen eating our own dog food.
  • It can serve as an example of what the dog food can do.
  • Unification of support resources. If "John Smith" wants answers. support.drupal.org is the first thing that pops into his head. Otherwise, he would have to look on SE + Mailing Lists + Issue Queues and possible more places.
  • Someone has to build it (us), and possibly pay for it.
  • Someone has to support it (us), and possibly pay for that support
  • It will take time to build, especially if it's a volunteer effort
  • Not endorsing Drupal SE may give a bad signal to the outer world (the Not Invented Here Syndrom)

Remaining tasks

// depends on the solution, so don't know what to say here.

User interface changes

// Same as remaining tasks.

API changes

API changes may happen if we revamp the core forum module.

Original report by chx

Let's slap a support.drupal.org CNAME over drupal.stackexchange.com . I know DrupalCon Chicago have raised the idea of creating a similar support.drupal.org site. It won't take more than a few years to debate the finer points of it and another year or two to get it up and running. I know it's not ideal. The user database is not integrated (however SE has openid!). It's not Drupal. However it's there, it works, it's maintained and enhanced daily. It's good enough. Perfect is the enemy of good enough.

Comments

cweagans’s picture

I would LOVE to see this happen. The support channels on d.o are old and crusty and SE is a great way to address that.

rfay’s picture

I completely agree. I think we should endorse it in multiple ways.

drupal.stackexchange has become a de-facto "best of class" for Drupal support, far better than anything we offer on-site. It's not what we wish, and we don't have any control of it, but let's just help out our users by admitting where the good stuff is.

cweagans’s picture

Also, since SE supports OpenID, perhaps we could enable the OpenID Provider module on d.o to support logging in with d.o creds on SE? Just an idea.

The SE people are pretty cool. Maybe they'd even come up with a way to integrate with bakery if we asked them really nicely.

killes@www.drop.org’s picture

Project: Drupal.org webmasters » Drupal.org infrastructure

I personally don't think it is a good idea. However, it's Dries' domain and if you can get him to approve this, it will be done.

dddave’s picture

Project: Drupal.org infrastructure » Drupal.org webmasters

As someone who is doing a lot of triageesque work in the D6 and D7 core queues I would really appreciate it if I could "officially" refer support requests to another source than the /support page.

edit: If we don't want to send people "away" (this would effectively be the result) we could make it very clear that support requests don't belong into the core queue's so that we can close such issues faster and justified. Additionally the endorsement of stackexchange could happen on the /support page.

lisarex’s picture

I am in favor of this too, if people here thinking that "fixing" the forums and and maliling lists seems like an impossible amount of work even for our dyanamo volunteer community. There do seem to be other d.o. issues people feel more strongly about fixing.

Having the good content in one place (free of old and crusty content) seems like a huge win. And we don't have to stick with SE forever. If one day there's a movement towards streamlining the support channels on d.o., we can always bring it back?

dddave, the /support page is crusty for sure (see #314325-30: Redesign the support page but text can be fixed... would love your comments on that issue :))

tim.plunkett’s picture

Project: Drupal.org webmasters » Drupal.org infrastructure

Fixing cross-post.
I'm a mod on Drupal Answers, so obviously I support this. Is there anything specifically we'd need from their end wrt Open ID?

greggles’s picture

Status: Active » Postponed

Let's wait for the site to be fully accepted and out of sketchy mode before we do anything like this http://drupal.stackexchange.com/faq#sketchy

cweagans’s picture

Status: Postponed » Active

SE already supports authentication with OpenID. If we want to allow people to log in with d.o creds, we'd have to set up the OpenID provider module on d.o. Nothing required from SE.

I think there are much better things to be spending time on than fixing the forums on d.o so that people can ask questions about Drupal. SE already has this problem solved and we don't have to maintain it. IMO, it's a win for both parties.

ksenzee’s picture

Also worth pointing out that SE content is CC, so if we did eventually build something in-house we could syndicate their existing content.

cweagans’s picture

Status: Active » Postponed

Whoops, xpost.

If we endorse it as the official support site, it'd get out of sketchy mode very quickly.

chx’s picture

Status: Postponed » Active

Yeah. No reason to postpone this. Also, dont forget that HedgeMage has been hired by SE so we have good contacts to SE already. (I do not want to ask any favorites through Susan it's just easier to make initial contact through her)

tim.plunkett’s picture

While I don't think that endorsing the site will directly influence its move out of beta, the influx of users will certainly help with that.

See this chart for our current stats, and this response from a moderator.

moshe weitzman’s picture

I'm a huge fan of this as well. Lets focus our precious drupal.org resources on other things.

killes@www.drop.org’s picture

Status: Active » Postponed (maintainer needs more info)

As explained: Dries needs to approve this, so somebody needs to get him to comment to that effect here. Until then this is postponed.

killes@www.drop.org’s picture

Ah, and there's a security implication too: we'd all send our bakery cookies to stackoverflow. I am not comfortable with that.

moshe weitzman’s picture

yeah, we use a different domain. no big deal. more important is that we start linking to it from all over drupal.org.

chx’s picture

Then make a redirect and not a CNAME.

cweagans’s picture

/me would foot the bill for drupalsupport.org

Mark Trapp’s picture

+1: Stack Exchange is a really great engine for Q&A, and this is one place we don't need to eat our own dogfood.

When Ubuntu decided to partner with Stack Exchange, they used a separate domain: Ask Ubuntu. I think we could work something like that out here, too.

tim.plunkett’s picture

And to contradict myself from #13, the Ubuntu stackexchange was moved out of beta early because they were endorsed.

cweagans’s picture

It appears that drupalsupport is taken, but I have registered askdrupal.org for two years for this purpose. I will continue to renew it for as long as it's used or gladly xfer it to the DA.

Let's just make this happen!

cweagans’s picture

Ticket opened for OpenID provider support on Drupal.org: #1236354: Deploy openid_provider on Drupal.org

rfay’s picture

@killes, could you please explain why this has to be approved by Dries? It was my understanding that d.o is entirely a community entity.

sreynen’s picture

chx’s picture

@rfay if we are to redirect support.drupal.org to SE then it is Dries' domain (drupal.org that is). If it's just askdrupal.org then it's his trademark :P

webchick’s picture

Assigned: Unassigned » Dries

rfay: Dries owns the drupal.org domain name, and the Drupal trademark. He's also the project lead, and a proposal to "outsource" such a fundamental part of our community should really be signed off on by him.

Adding to the list of issues for Dries to review.

catch’s picture

So if we do this, we also need to do the following:

* Deprecate everything under http://drupal.org/forum/18

* Fix #231728: Users still using deprecated forums

Those two seem pretty urgent to me, two more things to consider:

There are a lot of support questions being posted to groups.drupal.org, there should be an easy way to redirect those to stackexchange.

What happens with 'support request' issues in the queue? It would make tonnes of sense to disable it for Drupal core since that is full of tumbleweed, maybe not so much for contrib projects.

My only concern here would be if there were people actively working on the support.drupal.org site, but that is clearly not the case. Note that rfay opened #1047632: Proposal: Deploy a "support.drupal.org" site which would provide quality support-answers, has been very involved in those discussions, and has posted on this issue in favour. HedgeMage is the other main person I associate with that effort but if she is also our closest contact to stackexchange then great.

David_Rothstein’s picture

Despite its flaws, the current system has produced many great Drupal contributors (who started off as newbies asking for support in the forums)...

Is there any evidence yet that Stack Exchange will have a similar effect?

What's the procedure for a Stack Exchange issue that starts off as a support request to actually turn into a bug report, patch, or new documentation (if that's what it more properly is) that benefits Drupal and gets people involved in the community?

rfay’s picture

The place where the support site initiative has been going on is http://groups.drupal.org/support-infrastructure; Great people there, but it doesn't look like substantial progress at this point. I posted a note there to point people to this issue.

Thanks for explaining why Dries has to be involved. I certainly would never exclude his involvement, but there is obviously a limit to what he can be directly involved in.

I see this really as two things:

1. Deciding that for now we're outsourcing our support website. I'd love to have Dries support on that. Might or might not be absolutely necessary.
2. Updating our documentation and pointers to make that happen.

The use of a CNAME is actually extraneous, but useful. I think the big idea is "Endorse drupal.stackexchange.com", not "Make a CNAME".

@catch, #28: I don't think we should deprecate or remove the creation of support issues in the issue queue. They are the most direct and obvious way to get something in front of project maintainers. Triage can point some of them to stackexchange where warranted. Some of them even turn out to be bugs.

rfay’s picture

@David, #29: We all agree that we'd love to have our own functional support system. What we have, though, is completely nonfunctional, and we don't really have a path forward in the foreseeable future. Back in the day, support forum posts might have gotten answered. Now, the forums are a wasteland. There are probably 10 times as many posts that never get a useful answer as there are that do. It's not fair to people to suggest that people post there when in reality none of us would ever think of doing so.

There are definitely issues about how we get people on stackexchange to come back and become involved in d.o. The only answer I know of is to encourage all participants to point out explicitly how to file an issue when that's required, how to search the issues when that's required, etc. I know that stackexchange apparently doesn't let us put a block on the page with drupalish information, but perhaps we can approach them again about that sort of thing. @hedgemage?

catch’s picture

@David, that used to happen in 2005-7 when core and contrib developers actually posted in the forums on a regular basis. I have not seen any signs of it happening recently, and if it does it is likely despite the forums rather than because of them.

What's the procedure for a Stack Exchange issue that starts off as a support request to actually turn into a bug report, patch, or new documentation (if that's what it more properly is) that benefits Drupal and gets people involved in the community?

We have the same problem for forum topics now - there is no way to promote one to a bug report, and there is no cross-fertilization between the forums and the issue queue. If we wanted to put things like that in place, it would require concerted Drupal.org development resources to make it work, and these are not being applied to this problem currently.

@rfay that's fine, if there's a widely advertised support site that actually works then it might lead to less of the needless support requests being posted in issue queues (thinking of things like Views issues asking for CSS help etc.).

drumm’s picture

(I have no position here- I don't frequent the forums or StackExchange, and have no plans to start either. I do like the idea of improving the forums on Drupal.org, and happy to help where I can on infrastructure to support volunteer contributions, http://drupal.org/make-drupalorg-awesome. StackExchange thinks about their site full time, and they will always have some advantages, but nothing Drupal-specific. One thing we should not do is outright clone them.)

Many of the people commenting here don't look at the forums at all. And people active in the forums might not be aware of this. Who here talked to the target audience, or is in the target audience?

rfay’s picture

@drumm: Please do go over to http://drupal.stackexchange.com and take a look at the questions being asked and the quality of the answers being given. It's so far above anything we have in the forums it's not even comparable.

I would say that the audience for a support offering is everybody from newbies on up. I have posted on stackexchange and got decent answers for things I didn't know how to find out otherwise. If you look at the questions and answers you'll see a wide range.

Again, I would much prefer that we do our own support offering, and we are not giving up any future options here, as pointed out above. And we may even be able to bring in stackexchange content in the future, per #10.

The reality is: Those in the know are already using stackexchange. I think we only hurt newbies by not endorsing it.

chx’s picture

Yes, when we were still using the forums, the audience was still the community exploring web technologies. From http://www.garfieldtech.com/blog/drupal-priorities it pretty much follows that most users are now using Drupal dayjob-like and couldn't care less about contributing back. These days the problem is people wandering into irc or the core queue and not knowing where to start. We now have two answers to that, the office hours (which i needed to google to find, damnit!) http://groups.drupal.org/node/164629 and the core@drupal.org mailing list at #1236280: We need a central place for core contributors broadcast their business

joshk’s picture

I'm happy for this! :)

We should ask the SE folks and see if our endorsement will help get it out of "sketchy" mode.

cweagans’s picture

@all, killes has ruled out the CNAME anyways, as it's a security issue.

The best we can hope for is an official endorsement for stackexchange as a support venue. As I said before, I have a different domain (askdrupal.org) that I registered specifically for this task, so that would mitigate the security issue.

There is nothing saying that we cannot move the support channel back to Drupal.org at some point in the future. It would involve a bit of work to move things back to Drupal.org, but I think the change right now is worth it.

I'd even go so far to say that once this change is made, we could disable the forums! *braces for flames*

catch’s picture

You might possibly get some flames, but there was a lot of support for killing the forums during the d.o redesign, see the discussion at http://www.disambiguity.com/drupalorg-what-makes-up-drupalorg-community/ for example.

DjebbZ’s picture

Not a bad idea. Drupal SE is more robust than the forums, and is already frequented by some Drupal contributors. The good things : content is CC, and it could be used for a real world testing test of some of the ideas of the Prairie initiative wrt the redesign of the issue queue (cross-posted in the Prairie Initiative).
Oh, and I like the "kill the forums" idea.

bojanz’s picture

Definitely smart. I've seen d.o forum replies saying "did you post to SE as well?" so users are using the StackExchange site whether we like it or not.
It is here now. If we have a fancy support.drupal.org in a year, we can always migrate the content and endorse that, but let's have something good in the meantime.

davidhernandez’s picture

@all, as probably the most active member in the Support Infrastructure group (http://groups.drupal.org/support-infrastructure) I can say it has indeed not progressed much. I've still been mostly testing modules and doing research.

The main reasons for avoiding Stack Exchange were concerns over the ownership/control of the content and tighter integration with drupal.org (e.g. creating issues, adding credit to user profiles.) If that is not a concern, then Stack Exchange is ready made. I personally don't have a problem stopping what I'm doing, and us using another tool, if that is what everyone prefers.

Just one caveat - if this becomes the official support site, I don't see it migrating back. People will become entrenched there over the coming years, and it would seem like a lot of unnecessary effort to force everyone back to drupal.org. Someone working on drupal.org infrastructure would have to go through the effort of doing so, and why would they if drupal.stackexchange.com is working and their time can be better spent on another project.

You also have to think carefully about sending your site visitors to another site for a particular service. This will eventually lead to the habit of visiting the other site before drupal.org, and eventually not visiting drupal.org at all. (I don't mean in general, just for that service. A certain percentage of visits will be lost.) This might mean less people (especially newbies, who are most likely to be visiting drupal.org for support) will see the news updates, visit their dashboard, etc. It's not doom and gloom. I just want to point out that there will be some side affects and long-term implications, so those should be considered.

DjebbZ’s picture

Here's how I see a proper usage of Stack Exchange (SE) :
- usual help needed related to a module still belongs to the issue queue as a support request. It may lead to writing a patch, a documentation page, or better in code comments, or better README.txt, etc.
- broader questions would have a better home than the *cough* forums or mailing list *cough*. By broader questions, I mean questions that may concern several modules, or "meta-questions". For instance, a question like "What's the best way to create a quizz for my site ?" does not belong to any project specifically. So a system like SE may be appreciated : more user-friendly, possibility to tag cleanly questions, the voting and threaded discussion model is proven to be working well. That may help newbies helping get better support, in a more effective way. The existence of decent analytics in SE, of the possibility to follow a tag and the content licensed under Creative Commons may help think and provide an integration (even manual) with d.o. This will lead (to some extent at least) to this situation : d.o issue queues for contributors of any kind, Drupal SE for general help.

@davidhernandez You're rising pertinent questions. It sure needs careful thinking, as sending people out of drupal.org is not a consequence-less move. Here is what I think about long-term consequences. What may happen is that a module maintainer may be tempted to redirect a question to Drupal SE. And there, both the user and the maintainer are left into the wild. Drupal SE is a jungle that we haven't explored yet. And this very maintainer may lack some valuable feedback that exist only in Drupal SE. But we as community fail to provide broad support, we only (kinda) manage to provide per project support. And there's this site out there that has proven to be successful in providing a good experience related to general Q&A. I don't see it as a threat, but as an opportunity. It will leave forums and issue queues more peaceful, so developers will be able to concentrate on providing better core and modules. And one big difference between forums and Drupal SE is that Drupal SE is already frequented by some Drupal contributors, is close to leaving beta (see link in #13). Endorsing it officially will just make it official (no more sketchy mode), bring a lot more traffic - "questionners" as well as "supporters", and I'm pretty sure that current d.o "supporters" will navigate to Drupal SE to give help, as well as keeping the spirit of redirecting people to the proper place in d.o when necessary.

webchick’s picture

Thanks, davidhernandez, for voicing some of my own concerns. I fully expect to be steamrolled in this thread, but just so I can say I had my piece... ;)

My concern around this proposal is far less around the technology—which is unquestionably fantastic—and more about the parallels with the Git migration.

I had to fight very, *very* hard to keep Git on Drupal.org and not just close up shop and outsource everything to Github. Replace "SE" with "Github" in the OP and you have exactly the same argument. Github has unquestionably better technology, and that's their primary business focus, so we'll never out-do them on contributor experience. Would've been way, way cheaper for the DA, too. However, our developers are a fundamental part of our community. Out-sourcing them to a third party would mean obliterating a huge chunk of our community's culture, eliminating collaboration opportunities, etc.

The results of the Git migration speak for themselves: http://buytaert.net/drupal-contributor-statistics-2011 It has served to basically double development activity on Drupal.org. It took a year, granted, but it seems to have been worth it.

People who need support, also, are a fundamental part of our community. Most of the time, this is their first point of contact with our community. This is how they learn the Drupal community's culture, how they get pointed to places to contribute, how they integrate into Drupal.org.

Outsourcing support to a third-party is arguably even more dangerous than outsourcing our development community. Because people who need support now are the new contributors of tomorrow. If you impart upon newcomers today the idea that Drupal.org is merely a place to consume (get modules/themes, read docs) rather than a place to participate (ask/answer questions, post bug reports, edit docs, test patches) you go down a very perilous road.

What's difficult in this situation though, is that unlike the Git migration, there isn't the collective will of developers to see support.drupal.org come to fruition (beyond the general undertone here of wanting a place to shove newbies so the issue queue can be focused on development). The people who want to see it done do not have the skills to make it happen. The people who have the skills to make it happen are completely swamped with other duties that are in our project's primary focus. In the meantime, SE exists and it works.

However, David is completely spot-on that there's absolutely no coming back from this if we choose to officially endorse SE, and those who think we could someday in the future pull back to support.drupal.org are completely deluding themselves. It would require a 1:1 feature parity with SE, which is all-but-impossible, since they're dedicated to this mission vs. also trying to create an open source CMS. ;) It would require large-scale data migration.

So we basically need to make a decision in this issue, and the decision reached here is going to draw a line in the concrete for all Drupal generations to come. Choose wisely.

killes@www.drop.org’s picture

Thanks webchick for summarizing this so eloquently.

I'd be essentially a sell-out.

I also like to point out (as somebody else already did IIRC) that the people discussing this here are mostly not the ones usually using the support channels in either way (as support seekers or givers). Myself included of course.

I think the people who discuss this here are the ones who should be building that support site here but are too shy to step forward. Or too lazy. :p

rfay’s picture

@webchick, you're right in everything you say.

Could we gather the consensus to build our own support? To invest as we did in git?

In the git migration, developers were investing in developers. With support, it means the community significantly investing in n00bs and those not in the center of the community with a major effort.

I have long supported us doing our own support, as you know. But if we're not going to invest enough to do it, let's let the support at least get done somehow.

If we're going to build our own support infra, let's make a commitment to it in the same way we've committed to D8 initiatives or the git migration.

moshe weitzman’s picture

This askdrupal.org proposal improves the quality of support a ton. I have not heard any dispute of that.

What is the counter proposal? The DA spends hundreds of thousands to build a SE clone and begs the security team to review all the modules? And we have huge code sprints every n years to upgrade the site? We'd need an endowment to maintain it. Or is the counter-proposal to keep using drupal.org/forum for another 10 years. I think we have added icons in the past 10 years and that's about it. We added a couple forums and deprecated some others. Wow, progress. Maybe one day we'll be blessed with user pictures.

Folks are very good at pointing out risks but there are always risks. Strong decision makers can balance all the risks and make difficult choices.

Fidelix’s picture

@davidhernandez and @webchick couldn't have put this in a better way.

Please, lets not give third-parties the responsibility of dealing with one of the main duties of drupal.org, finding and contributing drupal-related community resources (help and information are resources too).

chx’s picture

People who need support, also, are a fundamental part of our community.

This is where we disagree. Once this was true but no longer.

webchick’s picture

Moshe: Yeah, no argument at all about SE providing a better support experience than Drupal.org now. But the point is, no one would've argued that Github provided a better developer experience than Drupal.org back in January 2010, either. The question is whether it's worth the cost / benefit trade-off to keep things in-house, and even whether that's actually doable. Your position on this is pretty clear, as is others'.

However, I'm not sure a SE clone would be hundreds of thousands of dollars, since it's mostly about deploying new functionality to d.o (vote up/down + activity + user points, i guess?), rather than an enormous data migration + new hardware deployments + shell scripts + new custom module code + new contributed module code + etc. that the Git migration required. The modules required to get SE functionality are deployed extensively "in the wild" so chances are they'd be ported for us by the time we upgrade drupal.org, and if not we'd be doing the community a nice service. It's not the same situation as Project* and VersionControl* modules, at all.

And although in the past (the past being before DrupalCon Chicago or so) I would agree with the general feeling of helplessness around making change on d.o, that honestly is no longer the case. http://drupal.org/make-drupalorg-awesome lays out the steps one needs to go through to to get changes made to Drupal.org, and API change nodes and issue summaries are proof that a group of people who want to see change happen on drupal.org *can* band together and make it happen these days. It's great. And since Drumm already pointed out he was in favour of deploying the tools we want, and killes is clearly opposed to moving support off-site, I think infra team buy-in won't be an issue.

Does anyone who was working on this at DrupalCon Chicago have something approaching a "spec" to understand how much work we'd be talking about? If we chose to try and seek funding from the DA for this, how much funding would we be talking about? And is there anyone here who has the bandwidth to head up such an initiative? These are all important factors to weigh in this decision.

killes@www.drop.org’s picture

I'd welcome it very much if a to-be-build support site with a lot (or some) nice features was it's own subdomain and Drupal site.

If this makes sense, is of course to be discussed.

Oh, and in principle I totally agree with chx. But if we move the support infra elsewhere it is unlikely that at least some of these people will _become_ part of the community.

chx’s picture

Also read http://meta.drupal.stackexchange.com/questions/377/making-sure-the-drupa... it's not like I have not anticipated this... the tl;dr version is what I said in the previous comment but also there's channeling back already when there's a bug but that's rare.

chx’s picture

Status: Postponed (maintainer needs more info) » Closed (won't fix)

I am closing this down and I will reopen on my next birthday (ie 2012 july 31) when we can revisit on how much drupal.org progressed in this.

jerdavis’s picture

Status: Closed (won't fix) » Needs work

I agree with webchick that I don't think this needs to be as great of an expense as you're suggesting Moshe, but much like rfay and webchick are suggesting, I do think the effort warrants a financial commitment from the Association.

We have a couple of options right now.

1) We move to SE and outsource support and accept the cost/benefit of that decision which webchick and davidhernandez have outlined, plus whatever consequences we haven't thought of yet.

2) We continue to use the tools we have and hope the community solves the support.drupal.org problem through volunteer efforts. This has the beneift of being "free" but the cost of taking a really long time, further putting weight behind choosing option #1 or having our audience choose option #1 for us (as they're already doing now).

3) The D.A. takes this on as a critical community issue and provides financial support, dedicating resources to push through a support.drupal.org initative. Obviously this costs the Association some money, but the benefit would be we get a solution much, much faster - and we own it and the data resulting from it (for better or worse). Could the Git migration have happened without the support it was given?

It's plain that we need a solution. It's also evident that folks are adopting SE, whether it's officially endorsed or not. To me, option #1 or option #3 seem the best choices. If the D.A. isn't willing to or is unable to take on option #3, then we should accept the cost of outsourcing this support channel and endorse the use of SE.

Also, opening this back up as this has stirred a good discussion that may result in some positive action - closing is a bit premature IMO

Mark Trapp’s picture

I understand the desire to control all aspects of the Drupal community by keeping it centralized, but it's not sustainable.

Like WordPress, people will ignore the strong-arming, use the tools and websites that best suit their needs, and create an ecosystem around the product that doesn't all exist at the site where the software is downloaded. We should be welcoming that: why wouldn't we welcome it? It cements Drupal as more than just an open-source project, but recognizes the larger culture of people using Drupal who want to contribute, but just simply don't use Drupal.org. And Stack Exchange has proven to be capable at handling support questions; why not support other people's efforts to improve the Drupal ecosystem? Why do we always have to reinvent the wheel?

Drupal.SE is on track to graduate in the next couple of months, it will continue to exist, and people are going to keep going to it and increasingly ignore Drupal.org's support offerings: that's to say nothing, like the others here have pointed out, people are using Drupal.SE despite it still being in beta. And the fears about someone else owning the data is unwarranted: all content is licensed under CC-BY-SA 3.0, and content authors own their data, not Stack Exchange.

It works, people are using it, the content license is liberal, and there is no viable alternative: leave Drupal.org forums to rot? Hope that someone (whoever that is, I'm glad people are volunteering unnamed people's time for a project) manages to come up with something we can launch several months from now? What guarantee is that ever going to happen?

Even if it does, are we really satisfied with having something that's not as good as the alternative (and that seems to be already conceded) merely because we invented it? If we are, are we really going to believe people are going to use something that's worse than a site that's already established?

People aren't going to use Drupal.org's Q&A just because you want them to: they're going to use the place that gets them answers. And more and more, that's Drupal.SE: it has a 6 month head start, and it's backed by a company that has 3 years of experience building Q&A communities. Believe me, functional Q&A is a lot more than enabling a voting module.

If you love something, set it free: let's endorse someone else to grow part of the Drupal ecosystem.

rfay’s picture

Mark Trapp is right. Either we do something amazing immediately or SE becomes the de facto support arm of Drupal anyway. This may already have happened. I guess it's probably already happened and per chx, we might as well just acknowledge it. This is way different from the git situation, where some projects were starting to go to github. In this case, nearly all the useful support (all except the support issues in the issue queue) has already gone to SE.

davidhernandez’s picture

@webchick, nothing approaching a spec sheet. The most is what we have in Susan's original sandbox, which we used to dump the features we wanted. http://drupal.org/sandbox/HedgeMage/1088372 Also, the gdo group.

Having done some local development on this, I agree that it shouldn't be that difficult; however, I'm not responsible for deploying sites to drupal.org, so feel free to disregard my opinion. The real customization would be for ideas we had, like pushing support requests from issue queues to support.do, and other cross-site type integration.

Stealing from Randy, "If we're going to build our own support infra, let's make a commitment to it..." I feel this is the main hurdle. To date, whenever I mention this project to people I get a lot of "That's awesome", and then chirping crickets. If there isn't buy-in, the discussion is moot. Maybe that is ok, and it isn't worth doing. I think it is, because some really nice things could be done with integrating it with other parts of drupal.org, but that all requires work. I think saying, "this is a problem, but we don't have the time or energy to fix it ourselves" is an honest, acceptable answer. If the outcome of that is not acceptable, well, I think we all know wishing upon a star isn't going to do anything about it.

R.J. Steinert’s picture

I'm interested to find out how Stack Exchange would feel about us scraping Drupal.SE in case they went terminal and lost all of the valuable data people will be placing in Drupal.SE or if we decided that Drupal.SE wasn't a good fit and decided to roll our own.

Damien Tournoud’s picture

I'm totally in support of moving our support tools to stack exchange. They are doing a great job, are already endorsed by preeminent opensource projects and handling support requests right is both critical for the future of Drupal and currently eating us alive.

Of course, we want to move the support-hungry folks to become contributors one day. This is *already happening* on Drupal Answers (beta). By handling support right, we can point people to the proper channel to report bugs and contribute patches.

For whoever would wonder: Stackexchange content is licensed in Creative Commons-wiki with attribution required. Full data dumps are available on a monthly basis.

In a way, the community has already chosen. Either we properly follow the lead (and get the chance to closely integrate SE with Drupal.org) or Drupal.org will just become this dead place.

dddave’s picture

I feel where webchick is coming from and I do feel a slight unease about "sending the kids away". However I strongly agree with Mark Trapp. The support system on drupal.org is very lacking and for various reasons no major improvements are imminent. SE is going to rock the support world anyways and we as a community should embrace it and see the benefits. We do ease the work in the major queues, we ally with a "good" partner and most importantly folks with problems do get support. We as a community want to help newbies and we can easily by embracing SE.
Is there a danger of bleeding out because of the lack of new contributors? I really doubt it. Those who get the Drupal bug will find their way. Perhaps we could even use the popularity of SE to get attention via ads/announcements or something like this?

(I think I qualify as someone who answers a lot of support requests...for what this info is worth)

jerdavis’s picture

@Mark Trap

Your points are completely valid, and the argument is sound. There's still questions though. Is SE the only thing we need out of a support environment? It certainly provides a solid foundation and maybe it's enough.

There's also merit to the arguments of wanting to keep people's attention on D.O. - and I think before a final decision is made it would be good to really step back and take a long term look at what the cost benefit looks like. There's a lot to be said for a tight integration of the support system to other parts of D.O.. It's worth considering these things.

Endorsing SE gives us a ton of short and long term gain. Particularly because it already exists and it's maintenance is someone else's problem. Both wins. It would also be a nice example of breaking free from the "not invented here" stigma. Also a nice win.

I'd suggest a couple of next steps for whoever has the authority. In no particularly sane order...

1) Let's find out if the Association is interested in supporting development of support.drupal.org. If the answer to that is no, then perhaps the rest is moot.
2) Why don't we engage the Ubuntu community and find out more about askubuntu.com? Now that I mention that though, I'm not actually seeing how Ubuntu is fully endorsing it? http://www.ubuntu.com/support.
3) Why don't we document the pros / cons somewhere that they can be collectively updated, so that the proper folks can make an informed decision. This could directly influence #1. Perhaps an Etherpad doc? (here's one http://piratepad.net/drupal-support-eval)
4) Identify the folks responsible for actually making this decision. Assuming Dries is a given, but what about the D.A. board?

Mark Trapp’s picture

@jerdavis

I can't immediately speak to your other questions, but can to the relationship between Canonical and Ask Ubuntu, as it was conveniently answered on Ask Ubuntu's meta-discussion site: :)

Canonical gave permission for [Stack Exchange] to use the Ubuntu trademark and logo for Ask Ubuntu. A Canonical employee also assisted in the site's new design to help meet the Ubuntu Brand Guidlines [sic].

Beyond that Canonical has made no public announcements relating to Ask Ubuntu. (So we must assume that the site is not officially endorsed by or related to Canonical).

(Note: There are Canonical employees that are members of the Ask Ubuntu community.)

The Canonical design team did blog about it when Ask Ubuntu was launched.

While I'm in the "let's just fully endorse it" camp for all the reasons above, that is another option: grant Drupal.SE the use of the trademark, help with some integration/design guidance, but stop short of fully endorsing it.

jherencia’s picture

Subscribing.

bojanz’s picture

While I'm in the "let's just fully endorse it" camp for all the reasons above, that is another option: grant Drupal.SE the use of the trademark, help with some integration/design guidance, but stop short of fully endorsing it.

No. Either we do it fully, or we don't do it at all. We desperately need to kill our crappy forums, and emptying the Drupal issue queue of support requests is a welcome bonus (though contribs can choose themselves, I'm fine with support requests in VBO's queue.)

sreynen’s picture

it's mostly about deploying new functionality

I worry we're drastically underestimating the effort involved in making something that would actually draw people away from drupal.stackexchange.com. There's a huge difference in effort between a functional Q&A site and a truly useful resource, and it's not really worth building just a functional Q&A site. Here's a good article about all the easy-to-overlook details that make StackExchange work well:

http://bitquabit.com/post/one-which-i-call-out-hacker-news/

webchick’s picture

"I worry we're drastically underestimating the effort involved in making something that would actually draw people away from drupal.stackexchange.com. There's a huge difference in effort between a functional Q&A site and a truly useful resource, and it's not really worth building just a functional Q&A site."

I actually disagree with this, on two points:

1. There is definitely value in improving the support offerings here on drupal.org, regardless of drupal.se.com's existence (unless of course we decide to outsource it entirely). Doubly so if we're using tried and true modules and solutions that other people building Drupal sites use. Then any maintenance work required to build out this functionality goes to benefit lots of other people. That's the often under-reported plus side of eating our own dog food. :)

2. I also don't think anyone here is proposing a 1:1 feature parity clone of SE. We're proposing is attacking some the worst problems that are currently driving people away from the forums as a support venue. And some of those are actually fairly easy to do in Drupal: Q&A format vs. forum format; a flag to delineate a question as "answered"; the ability to rate and sort by quality of responses; incentives for people to help out with support. And so on. These won't bring people back from SE, but that's not really the point; the point is to make asking/answering support questions here on Drupal.org a lot more rewarding so people actually do it, and users searching for support have multiple fruitful avenues from which to achieve it.

To jerdavis's point in #60, I feel like we would need to summarize the pros/cons (by updating the issue summary; I took a limited crack at it in the etherpad document), and ideally outline some kind of an idea of a scope, before approaching the DA about whether or not this is something they're inclined to fund. The discussion so far though seems pretty clearly in favour of outsourcing support (and the website traffic, talent, and mindshare that goes with it) to a third-party commercial entity rather than investing in our own tools.

David_Rothstein’s picture

I'm going to have to disagree with the prevailing wisdom on this thread, which says that the d.o. support forums are a dysfunctional wasteland. I just did a little research, and that's not what I saw at all.

The last time I was an active participant in the forums was probably 2007. I remember them being pretty good (and based on the comments above apparently that was the "golden age"), but I eventually moved on to other areas of contribution. Going back to look at them now, I expected them to be much worse than before. But instead, frankly, they seem to be the same quality I remember.

When I looked at the forums just now, I saw a huge number of posts, a huge number of posts getting replies, and a lot of respected community members who know a lot about Drupal participating in them (@WorldFallz, @nedjo, and @Itangalo are just a few of the names I noticed). For example, you can look at a view like this (or a similar one for other forums) to see how recently-posted support questions are doing:
http://drupal.org/forum/20?sort=desc&order=Created

Does this really look broken and dysfunctional to everyone?

I don't have anything against Stack Exchange; they obviously have some great tools, and the recent growth of the Drupal support channel there has been significant. But at the same time, I don't see what the evidence is that suggests we should be dropping focus on our own site and telling everyone to move over there. That idea seems very premature.

davidhernandez’s picture

Just for completeness, here is the original gdo post from Chicago and Susan's original rant:

http://groups.drupal.org/node/133494
http://binaryredneck.net/node/175

What is gained by officially endorsing, if it is decided that a CNAME is out of the question, instead of say a link somewhere to askdrupal.com? Is it to finally kill off the support forum? Is that an immediate need? Is it to remove the support request option from issue queues? Isn't that really a per maintainer decision? Some maintainers do actually use their issue queues to support users. Are we going to force an end to that process, and declare that support should be handled only at Stack Exchange? If not, do we not just muddy the waters by saying go here...or here...or here...or depending on the module here...or maybe...? These are honest questions I'm asking. I do so without judgement.

To reiterate, one of my main concerns is long-term implications. I hope everyone takes a breather and really considers the various aspects of this. One of the charges for a decision maker for a community is to consider the impact of decisions beyond your own term. That goes for both sides of this. There is an affect from action, as well as inaction.

I do, however, find it interesting that this has become a problem that all of a sudden needs to be dealt with ASAP, less the Drupal-sphere explode to bits. Unless I'm not in tune with the goings on, which is always likely. Maybe it's simply woken a sleeping giant.

Mark Theunissen’s picture

@David_Rothstein I agree, it's not a total wasteland, however SE is doing a better job of providing incentive for people to help out.

Let's not shut down the forums, but rather just officially acknowledge that SE is a good place to get support, maybe by listing it on /support.

DjebbZ’s picture

I understand webchick concerns regarding "sending the kids away". They're perfectly valid, and it's completely legitimate to answer these questions before making any move. Here are answers I can give :

Giving support is first a matter of seeking support, which means finding the answer to a specific question. This is what the Internet is good for. Thus it's the "supportee experience" that matters. In ANY projects out there, open source or not, sources of support are the following : #1 Google, #2 the official site(s), #3 the others unofficial sites (let's not forget #4 Wikipedia). The #3 tends to outgrow the #2 in the web in its whole. I need an answer about a video game ? Sure, I could visit the official website, but chances are that I will find my answer in an unofficial fansite. An answer about a Wordpress plugin ? I'm lucky, there's this post on this blog by X about how to use it, and I couldn't find a proper doc or answer in the official website (developer just released code in a zip file and voilà).

Drupal is not different. Our blog sphere out there is pretty cool to get informed, to watch instructive screencasts and sometimes even to discuss relevant points regarding Drupal the software/community. But (almost) nothing out there about concrete support. On drupal.org what do we have when one clics on "Community and Support" on the drupal.org website header ?

IRC
Ok for a small number of people (already using it), but can't scale, so let's forget it for this discussion (I'm not AT ALL saying deleting them, they're super useful).
Mailing lists
Hidden behind mail clients/providers and a daunting interface, and core mailing list already polluted by support requests. Doesn't scale either, can't search good answers, duplicates etc. Both IRC and mailing lists are not really "web" experience. We want support happening natively in the web browser.
Forums
Hum. I'm really visiting each of the sections before writing here. I see several problems in forum. #1, I've been visiting forum/22 "Post installation", and David_Rosthein just pointed forum/20 "Before you start". Furthermore, it's written above the list of forum topics in the page /forum that we should go in the issue queue first. Where is the proper place to get an answer ? #2 I would love some statistics about quality of answers given in d.o, but fear that we cannot give them. In the links chx give in comments #13 Drupal SE explains (their) criteria of what makes a good Q&A platform (number of per day, numbers of visits per day, percentage of questions answered, the answer ratio, number of users and of avid users. Votes could be added here to get a complete (IMO) overview of the performance of a Q&A system. Chances are, if we provide these stats for the d.o forums, we won't perform as well as them. And from what I see in the forum, we don't even match 90% questions answered which is sufficient to say we don't compete with Drupal SE in terms of perfomant and qualitative Q&A system. And even if we had these analytics regarding the forums, they won't include all the stale support requests in the issue queue.

So let's admit that our actual Q&A system s**ks. Let's admit how the web works, with several unofficial communities appearing for each successful project. Let's endorse one of these community that has already proven to be effective. From the "supportee" point of view, it's a win. What about the supporter and contributor point of view ?

Building our own system is still one of these Drupalisms, the famous syndrom of "not invented here". It *may* reveal functionnal and efficient, but when ? Going this route means relaying either on community efforts or the DA, which will be slow, and expensive (in terms of money or efforts). Going the Drupal SE way doesn't prevents the current "supporters" in the forums to create an account on SE and continue to help other people. It's specifically the point of SE, so they'll find a good place, with good tools and a good mood. It means less burden in our issue queue (super bonus as bojanz said). It means navigating from one site to another, which is maybe the only bad point about it. But it doesn't prevent current and future supporters of SE to convey the Drupal spirit in SE. So we won't lose that much people in the way, and maybe the opposite could happen. Contributors to the Drupal projects may even decide to visit Drupal SE, and because of the tools they provide, it could be easier for a module maintainer to provide support there : you can follow a tag ! I can follow Views related question for instance (on http://drupal.stackexchange.com/questions/tagged/views), have RSS feed of this very page, and even a detail of the tag itself here, with synonyms etc. (http://drupal.stackexchange.com/tags/views/info). As I'm a new member of the Views Bug Squad, I promise it's like a jewel compared to skimming the daunting drupal issue queue for support request. Our current tool just doesn't fit support general request.

Other points in favor of going to Drupal SE : less expensive, faster, so the energy not used/wasted on developing our own solution could be used to provide better integration between d.o and SE AND do other useful stuff for Drupal with our brain. Also, as Moshe said, in several years we didn't do anything serious about our forums (maybe because we never did something serious about our core Forum module as well ?). It's a proof that we're bad at designing forums. They were good when Drupal wasn't that mainstream, but now they're not anymore. SE is a good Q&A platform, it scales, we don't maintain it, the content is Creative Commons, the support is already happening... Let's face reality, change a few links on d.o to point on Drupal SE and focus on making Drupal 8 the best framework/CMS/CMF ever (and I really think it could be).

(Sorry for the long comment and for English mistakes if any, it's not my primary language)

DjebbZ’s picture

Webchick, I forgot to say this also : by endorsing Drupal SE, we're not sending the kids away, quite the opposite. We will tell kids to play *there*. Not to some other undefined, random and dangerous places. So we'll know where to look were we to search for them and get some news. It's better than leaving part of them messing at home, and part of them on the "Internet" (read : on their own).

tim.plunkett’s picture

#70 brings up a good point, people are already looking to external sites for support, and many are much lower quality than Drupal Answers. For example, I just yesterday realized that people are using Reddit for support.

webchick’s picture

I feel like I'm being misrepresented (and patronized a bit, too) here, which is a little frustrating.

To be clear, my position is none of these:

  • Drupal.org is the only website and the rest of the Internet doesn't exist. Lalalalala.
  • drupal.SE.com isn't a good avenue for support and SE doesn't what they do very well.
  • People who need support are "kids" who must somehow be "protected" from being "sent away."
  • We on Drupal.org must somehow strive to "control" our users and the flow of Drupal information.

Those are ludicrous positions.

My position is, however, some of these:

  • It takes two things to make a good support site: people asking good questions and people providing good answers. Drupal.org gets some 70% of its incoming traffic from search engines. If there are infrastructure improvements we can make to encourage more of the people providing good answers to do so on Drupal.org, that has a very direct impact on growing our community. The inverse is true for encouraging people to provide good answers somewhere off-site.
  • The primary reason not to do this seems to be that we're a bunch of web developers who don't know how to make changes to our own bleeping website. If so, let's fix that problem. Please, please read http://drupal.org/make-drupalorg-awesome. It is possible to make change on Drupal.org. Really! We are in a completely different situation now than we were before DrupalCon Chicago, as evidenced by API change nodes and issue summaries, among other incremental improvements rolled out in the past few months by people *not* on the infrastructure team.
  • When Drupal.org uses the same modules as other people in the community do, "eating our own dogfood" is not a detriment: it's an investment in our own platform. A lot of our users would benefit if modules like Vote Up/Down, Flag, Activity, and User Points were given attention by very smart people to clean them up enough for deployment on Drupal.org. Similarly, the support/maintenance burden for these modules, unlike something like Project or Version Control API which is where this attitude comes from I think, is shared among tens of thousands of other users. Where we run into trouble with NiH syndrome is when we're doing one-off custom things to Drupal.org, which the infra team takes great pains to minimize. This is not the same situation, at all.
  • There is no "undo" button once you shut down the forums and redirect them to drupal.SE.com. At that point we would need 1:1 feature parity of what SE does to pull people back, and we'll never achieve it for reasons outlined in #64 and elsewhere.

To be clear, I am NOT telling us not to make this decision. I am, however, asking us to consider the full picture of the ramifications of such a decision, rather than just throwing up our hands and giving up since we see an "easy" way out.

WorldFallz’s picture

As someone who has spent the majority of my contribution time in the forums over the last four years (sometimes to the tune of several dozen posts a day), I feel obliged to respond.

I'm going to try and avoid needless emotion-- but I have to admit, 'stumbling' across this thread felt like a punch in the face. While I've always known the forums are the redheaded step-child of drupal.org, I would have thought, at the very least, such a discussion would have warranted a post in the forums themselves-- where those of us who actually do spend our time might even see it. That this discussion is already on comment #70 and not one person has even suggested asking anyone who actually does contribute to the forums what they think is incredibly discouraging (have I really just wasted the last 4 years of my life?).

Ok, first, I contribute in the forums because so far it's the only type of contribution I can regularly fit in with my life at this time (it can be done in dribs and drabs through out the day). I feel strongly about contributing back-- and though I maintain some modules, I have been unsuccessful in my efforts to advance myself enough to contribute core. Also, it was helping out others in the forums that taught me just about EVERYTHING I know about drupal-- enough to create and maintain my own modules (i'm not a developer). Not to mention the fact that it is insanely rewarding in a way that slogging through an issue that goes nowhere for years can never be, lol. Totally selfish I know. ;-)

Second, the most striking, and irritating, thing about this thread is that not one name in the list of commenters / judgers are regular contributors to the forums. Specifically, Randy-- you contribute regularly to neither area (drupal.org forums or SE), never have (so it's not like you gave up out of frustration), and yet you seem to have a stringent criticism and strong opinion about what the decision should be. If we bless SE-- are you going to start contributing dozens of posts per day there? Is anyone else on this thread going to become a prolific contributor if we make the move? Shouldn't the people that actually contribute in this space now count for something?

That's not a criticism-- you're an awesome community member and prolific contributor-- just not in this space. It strikes me like Toyota trying to tell GM how it make better Chevys. If you can't or don't want to participate in the forums that's fine-- but is this decision seriously going to be made by those that don't, and likely won't, contribute to this area without even any input from those that do?

How do all you core developers respond when non core people try and tell you how best to do core development? How is this any different?

As for my 'vote' as it were-- I think webchick summarized the risks in #43 perfectly. Obviously my vote is to improve our own dog food (no, not duplicate SE) much like we did with the git project (we did not clone github). And to that end I'm willing to put even more skin in the game. No I'm not 'lazy'-- but I've been completely unsuccessful thus far in getting any improvements actually brought to fruition (after all, we're still left without even so much as a forum search on the new drupal.org even after all this time). I don't blame any one for that but myself-- I often have to choose between spending my time helping out newbies and spending time fighting battles in the issue queues and always cave to the former. I know intellectually, that the latter will be more beneficial in the long run, but with limited time I feel more personal satisfaction from helping out a newbie than trying to make a case in an issue somewhere.

Well, that's my $5 worth... let the flames begin ;-)

dddave’s picture

@webchick
I apologize for bringing the term "sending the kids away" into the discussion. It was never meant to patronize,just a sloppy formulation by a non-native speaker. Let us forget this formulation please.

@Worldfallz
I started my community contribution as a support seeker in the queue and transformed into a support giver there. However I noticed a while ago that all I do nowadays happens in various queues but almost never in the forums. I am not really sure why that was the case but I don't feel the urge to get back there. The SE system however has a huge incentive to participate and I am not sure that it wears down as fast as the incentive to do the forums did for me.
From a support seeker position I recall really not liking the forums. ;)

That said I understand both points raised by webchick and Worldfallz. If we can improve the system here I am all for doing it. However I think that we should embrace SE anyways because it frickin' works. Should that mean we nuke the forums? Not necessarily.

I think as a first improvement we should strongly communicate that support requests don't belong into the core queues and that support can be get in the forums or at SE. I think that would calm down the people working on core and we could think about the whole situation more carefully.

(Btw: How many bugs in the upgrade path to D7 were discovered due to support requests?)

kiamlaluno’s picture

To what already said, I would add that Drupal Answer is not the only used channel for support, and what we have on Drupal.org can co-exists with a Stack Exchange site. Before Drupal Answers there was already Stack Overflow with questions about Drupal.
To make an example, every Stack Exchange has a chat room, but the one associated with Drupal Answer has been already frozen once, simply because it's not used; if it is still active is because I wondered why it was active, and a moderator re-enabled it. This means that users prefer using the existing IRC channels we already had, and nobody used that chat room for more than 14 days.

About the comparison with Git, I would say that the situation is different, as the update code running on Drupal checks for updates on drupal.org; it would be a problem outsourcing the Project repositories, when the external repositories could stop working in any moment.
Having another support channel is different, as more support channels should be welcome, especially when the external support channel is redirecting the users to the appropriate issue queues, when necessary; that is not something I see happening on Stack Overflow, where I saw many issues with a module being answered, instead of suggesting to the OP to report the problem in the project queue.

Creating a support site with Drupal would be a showcase of what Drupal can do, in the same way the other sub-sites we had do. It would be if such site would be created.

As for what SE sites have that catch the interests of many experts is probably the fact privileges depends on participation. A user can have privileges closer to a moderator just gaining reputation, which is then gained with the participation.

DjebbZ’s picture

Thank you webchick and WorldFallz for answering.

Webchick : I've read http://drupal.org/make-drupalorg-awesome, and I only found technical and process details, when I thought I would see a battle plan to make it awesome, I'm a bit disapointed :/ Can you point me/us to a post with explanations regarding how the API change nodes and issue summaries changes happened ?

Anyway, I want to quote you :

Outsourcing support to a third-party is arguably even more dangerous than outsourcing our development community. Because people who need support now are the new contributors of tomorrow. If you impart upon newcomers today the idea that Drupal.org is merely a place to consume (get modules/themes, read docs) rather than a place to participate (ask/answer questions, post bug reports, edit docs, test patches) you go down a very perilous road.

I disagree with you here. The point of outsourcing to Drupal SE is just a part of people asking and answering questions. The whole point is to free the place where we want to contribute (a.k.a, the core mailing list, the core development team in general, focused module development, issue queues in general) of unrelated support requests, not to remove all kind of support request from d.o (I'm NOT in favor of removing the "support request" category of issues) neither to kill all kind of contribution, quite the opposite. Moving to Drupal SE is just a matter of cleaning up the place so d.o remains a clean place for contribution, and Drupal SE a good place for general support request. They're not mutually exclusive.

But if we were to go to the "make drupal.org awesome" route (good thing !), I will quote WorldFallz himself. You're regretting that this discussion is not happening in the forums. It's precisely because no interesting community discussion happens in the forum that we should not use the forum anymore. I would even go the opposite route, to the risk of contradicting myself : since most of the work happen in the issue queues, the support should go there too. Issue queues have tags, status, category, components, etc. By adding voting, flagging, maybe refactoring a bit the overall issue queues + creating a dummy project to contain the non project-specific support requests, maybe we could "kill the forums" without moving them away but making "in-place support" more efficient.

Even if I believe that building a good Q&A system inside d.o is more than adding a bunch of code as Moshe explained...

(To respond to WorldFallz flame mode, I'm giving support in Views issue queue as I freshly joined the Views Bug Squad, and also in IRC on #drupal-fr as I'm French)

DjebbZ’s picture

ddave, kiamlaluno thank you too for your words. I was going to call for your participation kiamlaluno as I noticed you're active already in Drupal SE. ddave your summed up it all :

I started my community contribution as a support seeker in the queue and transformed into a support giver there. However I noticed a while ago that all I do nowadays happens in various queues but almost never in the forums. I am not really sure why that was the case but I don't feel the urge to get back there. The SE system however has a huge incentive to participate and I am not sure that it wears down as fast as the incentive to do the forums did for me.
From a support seeker position I recall really not liking the forums. ;)

As I said it's about the support seeker experience : Drupal SE is just far better than the forums, I didn't like the forums too as most of the contributor community and don't travel there too. I'm quoting ddave again, I so agree with you :

I think as a first improvement we should strongly communicate that support requests don't belong into the core queues and that support can be get in the forums or at SE. I think that would calm down the people working on core and we could think about the whole situation more carefully.

To sum up (if I may) moving to Drupal SE is about giving better support and freeing core queues i.e enhancing contributor experience.

webchick’s picture

Assigned: Dries » Unassigned
Issue tags: +Needs issue summary update

Actually. Removing this from Dries's list until we get a decent issue summary.

DjebbZ’s picture

What's the relation between this issue and the issue summary issue ?

bojanz’s picture

What webchick means is that someone should create an issue summary for this issue, and summarize the discussion so far, as well as the options that we have.

Michelle’s picture

@DjebbZ: "herself" :)

@all:
I've just lurked in this debate since, while I'm against moving our support offsite, I also recognize that I'm not doing anything productive to keep it here. But Webchick made a very good point that I want to underscore: eating our own dogfood has a side benefit of improving Drupal itself.

I've spent most of my Drupal life in the social networking space and I can tell you I'm repeatedly frustrated by the state of the social networking modules. It's not just the forum that needs work but many of the modules that go along with it to make a normal forum / Q&A site experience. By putting focus on getting the social networking on d.o up to par, that would help a ton in getting the whole Drupal social networking site building experience where it should be.

And before anyone says we're talking support, not SN, really we are. Support is the subject but the tools used, ratings, karma, profiles of answerers, etc, that's all SN. This is a space that people really want to use Drupal for and time and again end up frustrated because it's not really ready for prime time. Well, here's a perfect excuse to make it happen if we can just get enough people willing to jump in and get it going.

Michelle

webchick’s picture

Assigned: Unassigned » Dries

Ok, just explained the issue to Dries, so putting it back on his list. ;)

Mark Trapp’s picture

Assigned: Unassigned » Dries

The primary reason not to do this seems to be that we're a bunch of web developers who don't know how to make changes to our own bleeping website.

That's not the issue at all: everyone seems to agree that at some point, Drupal.org can come up with something that will attempt to be better than what's available now through the support forums. The issue is that this already exists off-site, and the argument is we shouldn't be reinventing the wheel just because it's not on Drupal.org. Something out there works and is already doing a good job at it. Let's just recognize that instead of pretending it doesn't exist until a proposal, which hasn't had any traction in months, gets enough support.

There is no "undo" button once you shut down the forums and redirect them to drupal.SE.com. At that point we would need 1:1 feature parity of what SE does to pull people back, and we'll never achieve it for reasons outlined in #64 and elsewhere.

Isn't that the point? If we're conceding now that Drupal.org isn't ever going to offer something as good as Stack Exchange is today, why would we go through the motions of denying that a better offering exists while creating something that can't compete with it anyway? People are already finding Drupal.SE questions in Google already: people are going to go there, and increasing numbers. We have an opportunity to help people go to the best place available for support: that should be our primary focus, not use support as a test bed for an attempt to get modules more love.

And to those that think it's just a technical problem to overcome, it's not. Q&A is more than just the toolset, but a set of community standards, consensus, and guidelines about what makes a good question and answer pair and how to get people to answer questions. Even if you had a 1:1 feature parity with the Stack Exchange engine, Drupal.org would still be starting from scratch on those. There's a reason why people answer questions on Drupal.SE and Stack Overflow instead of even bothering to try on the forums, and why Stack Exchange is compared favorably to Yahoo! Answers, Answers.com, ChaCha, and Experts Exchange: they actually took the time to figure out all the various problems a Q&A free-for-all entails because that's all they do. It's going to take a long time for Drupal.org's Q&A system to reinvent those policies too, if people just don't give up beforehand.

So while I can appreciate wanting to keep it in house, I think it's important we recognize the genie is already out of the bottle, so to speak: people are branching out more and more from Drupal.org to talk about Drupal, there are places that are just way better at doing certain things than the Drupal community is, and we should be embracing that larger ecosystem.

And just as an aside to something WorldFallz said:

Is anyone else on this thread going to become a prolific contributor if we make the move? Shouldn't the people that actually contribute in this space now count for something?

I contribute in this space: I'm the 7th highest rep user on Drupal.SE and have answered over 130 Drupal questions on Stack Overflow. I don't do that in the Drupal.org support forums because they're a nightmare to use.

And I don't mean that to be an attack on you or others who give help in the support forums, because that's not what this is about. It's about identifying and supporting the best support channels we have available today. And the argument is that it isn't the support forums.

lisarex’s picture

Assigned: Dries » Unassigned

@DjebbZ's idea of using the issue queue for support is interesting too, since it's possible to isolate/list issues by component. And then, all the planned improvements for issue queues can benefit the support channel too. And hey, nothing new to build!
But we'd need a separate 'project' for support issues that don't fit neatly into module or core.

@DjebbZ, I clarified the link to the drupal.org initiatives on http://drupal.org/make-drupalorg-awesome. Let me know if it is still not clear enough (or someone else can go ahead and make those changes if you have the Documentation mainainer role) . I also added a bit about non-code contributions, such as revising communication and docs nodes, since those are important too :)

webchick’s picture

Restoring assigned-to flag that was lost in a cross-post.

(Sorry, I don't have time atm to respond to the responses since my last response atm; will try to do so tomorrow.)

davidhernandez’s picture

I don't think anyone is trying to deny the existence of Stack Exchange. I don't think anyone that is on the side of not endorsing it (and I don't want to claim a side) is telling anyone to not use it. The more resources, the better. To anyone that currently uses Stack Exchange, is a lack of an official endorsement keeping anyone from using it? When I have problems, I google them, and find help in lots of places.

It is easy enough to place a link and blurb somewhere pointing to it, but people are also talking about some very real changes to drupal.org, like deprecating forums and removing support from issue queues. That is a rather significant change that should be thought about carefully.

I completely understand Michelle's view on forums. Every time I discuss it with someone they say, "I don't want to help make them better because I don't use them." Why don't you use them? "Because Drupal forums suck." Well, that's a lovely, vicious little cycle, and I see that here, as well. I'm not aiming that at anyone particular, but it is a common problem. I'm guilty of it as much as anyone. But, I would like to actually do something about it, instead of just giving up. (which would be the usual)

VM’s picture

Had I known that the effort and time I put into supporting the project by way of the d.o. forums wasn't going to be considered worthwhile for this many people 5 years later. (Especially by people I look up to and respect.) I'd never have started the journey. I've answered near 10,000 questions in Drupal support forums over my near 5 years here. I did so without a need for a status symbol in the form of reputation, points or any of the other recognition that seems to now be a requirement for a successful support system/team. In particular for the people who know more than I do. It was enough for me to feel part of the community, to learn from those who knew more than I, and pass on that knowledge to others who were asking for it in their questions.

I don't understand why support for drupal can't live in multiple places, including drupal.org forums. I never have when this type of discussion comes up. Our forums may not be the best. Other systems may indeed have better tools to deal with many things. However what we have are drupal's forums and they are therefore ours. (whether they are used by everyone in this thread or not.) It's quite easy to make a decision to throw something away when its not a tool used by those making that decision.

Perhaps I'm better at navigating nightmares than some. : off to update my resume to that effect :

DjebbZ’s picture

@lisarex Good to see my idea of centralizing everything in the issue queue pleases you. It is to me the only viable alternative to the forums, which I think we should really kill in a way or another. Drupal.org communication is mostly around issue queues, so instead of forcing d.o users into forums, with whatever enhancements we're talking about, let's just work with d.o users (a.k.a. ourselves) and put our brain into making the issue queue a better place. You know, out there in the web design community, it's called "user-centered" design : let's follow our user's feedback and adapt our model to them, and not the opposite. The web is full of story where it has proven to be successful. Regarding our support requests problem, it means either endorsing a place that already works but that is external (Drupal SE) or enhancing our home where people are concentrating (the issue queue).

@lisarex, thank you for fixing the page. Now the page "making drupal.org awesome" deserves its title :)

@davidhernandez I'm gonna repeat it : it's about deprecating the forums but not removing support from the issue queue, but just removing support for issues not related to the issue queues where they're posted, and focusing support in issue queue when it's relevant. There's a real difference, because module maintainers still want feedback in d.o about their modules, but tell me in which issue queue should I post "How to to build an image gallery (possibly with Views, CCK and Views slideshow) ?".

bojanz’s picture

I'm sad to see the most frugal support contributors so put off by this discussion. VM and Worldfallz are the life of drupal.org forums.
I think we can all agree that the point of this discussion is to make Drupal even better, not dimish anyone's work. Still, the fact that the support channels could use more contributors still stands. And making that experience more pleasant is vital in order for that to happen. You can't force anyone to volunteer their time to something that they feel is unpleasant.

I am sure that drupal.org could be improved to the point of not sucking, but that would require an investment by the Drupal Association, just like the git migration did. History and experience has shown us that volunteer efforts can't make it happen fast enough.

webchick’s picture

Well, it's hard to not take it personally when your peers are saying things like "We desperately need to kill our crappy forums" and "The support channels on d.o are old and crusty" and basically pissing all over your work.

Once again, I'm happy to forward a request for funding off to the DA once this issue node contains a balanced summary of the pros/cons, and a basic outline of what improvements we'd be funding.

bojanz’s picture

Well, the fact that our support channels are old and crusty shouldn't be a point against the people who use it to help others (like VM and Worldfallz). After all, they didn't write the forum code that makes their work possible (if not a bit hard).

VM’s picture

Just to be clear I'm not advocating the forcing of anyone to work in any one specific space. However, that's seemingly what is being forced on those who passionately work/worked in the d.o. forums. Those who've heeded calls to help when and where we can in an effort to help make drupal awesome. The support forums have always needed more people to aid and is exactly why I got involved in them 5 years ago. It was and essentially still is, the main way I contribute back to the project.

Fact is that, I (answering) and most users on the forums asking questions don't require bells and whistles with the passing on of knowledge. I/we/they, merely require the knowledge to be passed on.

When I'm in musician mode I try to surround myself with musicians who are better than I am. I do this so that I can get better. Learning what I've learned of Drupal wasn't much different 5 years ago. Much of the knowledge I gained about Drupal was from the upper echelon of developers who at one time did use and frequent the d.o. forums. Over the years, for reasons as has been mentioned in this thread their support contributions are now completed elsewhere. IMO, this aided in the slow down of knowledge trickling down through the levels of users oin the d.o. forums and hurt them more than the tools desired not being there but that's a whole nother dicsussion.

Edit:
For the record, forcing all support into the issue queues will not work, IMO. It would essentially create the same issue we haev now. Many of the queues are as undermaintained as the forums are. There would be little way for those who provide support to navigate that idea. Example, Shall I watch every issue queue for modules that I know something about? Focus my support activities on only a few? Based on the current toolset, I don't see this being as easy as presented.

Mark Trapp’s picture

To be clear, this is not about shutting down all support channels by which people can contribute back to the greater Drupal community. That we can only properly contribute back to Drupal on Drupal.org and using Drupal software is what's being argued is unfounded.

Right now, we have two support offerings: the support forums and Drupal.SE. The former is sanctioned by the Drupal community by default and the latter is unsanctioned, but is quickly becoming popular. The motivation behind endorsing Drupal.SE is to consolidate community support to prevent fragmentation, which makes support harder: questions get duplicated, answers get duplicated, and people's time gets split (or the expertise of the community gets split).

So if we take that as being the goal here, the argument is that we have a better support channel ready to go now through Stack Exchange, and the alternative presented is to wait for some improvements that may come in the future (or otherwise leave the forums alone).

The former would not prevent anyone from contributing to support on Stack Exchange, and doesn't invalidate anyone's contribution to the community. It would mean that support would be run on a platform not owned by the Drupal Association and not developed with Drupal.

The latter would mean a few different things:

  • Whether Stack Exchange has the license to use the Drupal name becomes unclear
  • Support continues to be fragmented as people aren't going to stop going to Stack Exchange for answers
  • Support stays in the control of the Drupal Association and continues to be run on Drupal

And that's the main issue: by not endorsing Stack Exchange and continuing to point people to the support forums as the only official way to get support for Drupal, we're purposely ignoring a useful support channel that we should be welcoming.

There may be a compromise in avoiding the fragmentation issue and supporting both channels by "blessing" Drupal.SE as an alternative to the support forums through granting it a trademark license and linking to it on d.o/support, but that was shot down in #63.

So if we're going to continue to say the support forums are the only way forward for Drupal support, we're effectively saying Drupal.SE is a rogue support site and people shouldn't go there. That's just crazy, and I would hope nobody here thinks that's a reasonable option.

Which brings us back to the original focus of this issue: endorsing Drupal.SE. Helping people find support should be the first goal, and Drupal.SE is a valuable resource that's only going to grow. We can either throw our support efforts behind it, or if we're going take wholesale replacing the forums with Stack Exchange off the table—which I believe to be a mistake—we should at least be sanctioning it by providing links on drupal.org/support to it and granting it the use of the Drupal name.

Edit

It looks like Drupal Answers was added to the support page a couple of days ago. Is that an official sanction and endorsement of Drupal.SE? If so, this all seems to be a moot point, and a discussion about improving the support forums is better handled in another issue.

VM’s picture

The former would not prevent anyone from contributing to support on Stack Exchange, and doesn't invalidate anyone's contribution to the community.

The former does prohibit support from those who prefer to work here on d.o. forums. It's essentially forcing them to choose between contributing at SE or not contributing at all.

Adding to drupal support channels is not something I'm arguing. Support away. Forcing me to contribute to SE isn't something I'm willing to do, especially if the reason is popularity.

killes@www.drop.org’s picture

Sorry, Mark, but you better get your flameproof suit right now...

That we can only properly contribute back to Drupal on Drupal.org and using Drupal software is what's being argued is unfounded.

To eat its own dogfood is precisely the attitude that has brought the Drupal project forward in the past ten years. And I for one am not interested in ditching this just because some people see shinier toys elsewhere.

Please do not further pollute the infra issue queue. This issue is waiting for Dries to review it and for nothing else.

Mark Trapp’s picture

Please do not further pollute the infra issue queue. This issue is waiting for Dries to review it and for nothing else.

My mistake, I didn't realize the only way to discuss an issue with Drupal is to either have it approved by Dries or to agree with you wholeheartedly. This type of dismissal of others' viewpoints is a great way to grow the community, and I applaud your efforts in doing it.

killes@www.drop.org’s picture

Project: Drupal.org infrastructure » Drupal.org webmasters

moving to webmasters. Since we aren't going to use a cname for whatever we end up doing, this is not in the infra domain.

VM’s picture

@Mark

Edit
It looks like Drupal Answers was added to the support page a couple of days ago. Is that an official sanction and endorsement of Drupal.SE?

Unfortunately the editor didn't mention here the motive behind adding that to the support page. Though they are included in this discussion. I don't see a log entry as to why it was added, though that could be a perms issue. I can't speak for whether or not that's any official endorsement for SE.

kiamlaluno’s picture

Comparing Drupal forums with an SE site is like comparing an apple pie with a cake.

  • A Q&A site is not a forum; what you can post as an answer must be an answer, and not a comment saying "I have too this problem."
  • An SE site is more restrictive about which questions can be asked; every question that doesn't follow those criteria is closed.
  • If in an SE site there are questions that diminish the value of the site, Stack Exchange staff will give suggestions, and take any necessary action.
  • In an SE site there are some questions that are sponsored, which means (if I understand well) that some employees of a company answer to those questions; you understand which questions are sponsored by looking at the used tags: If one of the tags contains an icon, then the question is one of those.
  • SE sites are created by a company, Stack Exchange Inc., whose purpose is to create successful Q&A sites.

There is much difference between any SE site and Drupal.org because they have completely different purposes.

If the problem were the inadequate forums, then that is a point pro who has been able to help users all these years, and who keeps helping them. Forum should make better, and we could be more restrictive about the questions to ask, but the main topic is simply one: a Q&A site is not a forum, and there are questions asked on Drupal Answers that will be closed, leaving the OP without an answer. There is still the need of a forum, where anybody can ask a question without seeing it closed as not constructive. The difference is that in a forum, you guide the users to a solution, without putting any limits to the kind of questions being asked (well, with some obvious exceptions); in an SE site, you also need to check if the question follows the set criteria.

TL;DR Forums and SE sites are two different things that are for different audiences; they can both be used, and we are not forced to choose one instead of the other. Saying "this is better because […]" doesn't take to any place; using SE terminology, the question is not constructive ;-).

WorldFallz’s picture

What keeps getting lost in this thread by those that clearly hate the forums, don't contribute to them, and need badges, karma, post counts, or other pats on the back in order to reward their contributions is that much of what goes on in drupal.org drives drupal itself. If we do ths we are essentially abandoning the core forum module.

And yes, that's a bit snipey-- I've purposely tried to tone down rhetoric, but it's getting really hard and once again webchick captures a point perfectly:

Well, it's hard to not take it personally when your peers are saying things like "We desperately need to kill our crappy forums" and "The support channels on d.o are old and crusty" and basically pissing all over your work.

The only thing I would add to that is the this is most infuriating when those saying it don't contribute to what they're complaining about.

It's said all the time that drupal is a meritocracy. I've always believed that to be true and always used it as a guiding principal for my interactions on drupal.org. I spend a LOT of time in the core issues queue-- and generally keep my opinions to myself because I've not earned the merit to speak up. Even when I see important issues languish for years as people bikeshed over this or that minutia-- even when i want scream at the top of my lungs out of pure frustration.

Sorry folks, it may not be what you want to hear, but most of the most vocal people trying to shut down the forums in this thread simply haven't earned the merit to control the conversation. Why is it drupal is a meritocracy when someone criticizes core or views but not the forums? Has the hypocrisy infecting the world at large finally infected drupal as well? Is this really a place where those that yell the loudest get to control the discourse regardless of whether or not they've earned it?

Now can we please move on from the 'drupal support forums suck' undercurrent and keep to what to do about it?

I don't know if anyone has seen http://drupalize.me/videos/fantasy-site-stack-overflow-diwd-09 or http://drupal.org/project/arrayshift-- and no I'm not sucking up to webchick or the bots, I just drool over the functionality there, lol. And that was done in a weekend iirc.

And no, i'm not advocating recreating SE, but it demonstrates that we can have better drupal support, with existing tools, and in a decent amount of time-- providing we don't bikeshed over whether or not this link in that block should be first or second for 5 months. And that those that need shiny doodads can have them.

Most of us that do spend time in the forums know exactly what needs to be done-- from our own experience as well as from interacting with the forum users who tell us how they think it should work (which again, those that don't participate in the forums will not be exposed to). With the new ability to spin up drupal.org dev sites we can probably put something together demoing what a great drupal forum and support forum could be pretty easily. That would then give the infra folks something concrete to review-- as opposed to having to spend weeks and months in individual issues discussing what the performance implications might be.

So, it's definitely doable. And not in 10 years. The question then becomes whether or not we want to do it and that type of philosophical decision is up to Dries. IMO the drupal.org forums and support mechanism should be a showcase for drupal, but that's me.

WorldFallz’s picture

Didn't see kiam's post before my previous post. He makes a very good point. There's actually 3 issues here-- one being improving the forums, two whether or not support.drupal.org should use the Q&A format, and finally, if yes, whether we want to implement it here or give it over to SE.

EDIT:
by 'implementing here' I'm still not advocating recreating SE-- SE has a great model, but it's not the only way to Q&A and doing Q&A here will probably have some elements of that, but need not be a complete recreation of that.

DjebbZ’s picture

Thank you kiamlaluno for clarifying the way (Drupal) SE works. You made a good point in differentiating a Q&A a and a forum.

Thank you WorldFallz for shouting your feelings, I understand them. I really agree with you when you say "d.o should be a showcase for drupal". But hey, you know what ? I remember webchick herself saying that someday she will do something about core forum module. Because apart from storing nodes in different topics and subtopics and allowing comments, it can't compare with other forums out there. Technically and UX-ly speaking, yes, our core forum module sucks. A lot. This is why some people have pointed to Drupal SE. Or why I have suggested moving support forums to issue queues, because there we have tools to interact with them (status, tags, category etc.). And the coming improvements to the issue queue (=contribution) will benefit the support for free.

But you're suggesting we should improve the forum. As you're one of the most prolific contributors to the forums, if I were Dries I would designate you head of the "Better Forum Initiative" (no joke, I'm serious). This would help the core forum module, which seriously needs some love (or a complete rewrite).

juan_g’s picture

Mark Trapp wrote (#93):

a discussion about improving the support forums is better handled in another issue.

A detailed proposal is here: #1239326: Update the support forums structure according to the current needs, and naturally others can help with ideas, experience, etc.

I've helped a little on the forums these years and in my opinion, in addition to needed improvements to Drupal.org support forums, third-party technical Q&A like the one discussed have also a role in helping support the growing number of Drupal users. When we add links to third-party services, we should consider only the best interest of our users.

You know, some clever people say this is the #1 reason of their success:

1. Focus on the user and all else will follow.

Google: Ten things we know to be true

rfay’s picture

I would like it if we could focus this issue on

How can we successfully provide support to Drupal users, new and old, over the long term?

There are lots of other things being brought up, but IMO that's the issue to focus on. We're not doing well enough at it now, despite the heroic effort of several great people.

All of the other issues are peripheral, IMO. Eating our own dogfood? That's a completely different goal, it has to do with improving the builders and maintainers of a new site or subsite. Keeping control of our content and its presentation? That's a valid concern for the community, but again a very different goal. These are valid parts of any resolution. But *resolution* is the thing. Make support work.

Can we make that the primary issue here?

How can we successfully provide support to Drupal users, new and old, over the long term?

DjebbZ’s picture

Proposal for an issue summary for Dries to review (my 1st ever) :

Problem/Motivation

Endorsing drupal.stackexchange.com aka Drupal SE aka Drupal Answers. The site provides a good Q&A platform and experience, is well frequented and is already used by some Drupal contributors. Comparison has been made with d.o forums, where Drupal SE is somehow preferred. The motivation is to provide a good "support seeker" experience, reconciliate the contributor with the forums, and clean the issue queue and the mailing list from unrelated support request. Problem : do we officially endorse this site as one of our (or our only, meaning killing d.o forums) support platform ? What about trademark and policy issues, since they're using the word "drupal" in their domain name ?

Proposed resolution

Several solutions has been proposed, because we have parallel problems (Drupal SE and d.o forums) :

Regarding Drupal SE

  • Kill d.o forums and point every general support request to Drupal SE
  • Not endorsing Drupal SE
  • Endorsing Drupal SE and keep forums too

In any case we also need to solve the trademark issues.

Regarding d.o forum

  • Organise a volunteer-driven enhancement to the d.o forum and the core forum module
  • Fund with the DA these enhancements.

Pros and cons

Support place to keep Pros Cons
Drupal SE
  • More and more frequented
  • Well referenced in search engines
  • Very good Q&A platform, good experience for support seekers as well as support givers
  • We don't maintain it
  • Available now
  • Free
  • Content is licensed under Creative Commons-wiki with attribution required
  • Not in drupal.org : risk of losing contribution
  • means killing drupal.org forums (when they're still used)
  • Trademark issues
  • Corporate, we don't know how it will turn in the future
  • No integration with drupal.org
  • No 1:1 feature match with the current forums
d.o forums
  • Already here
  • Belongs to community
  • Drupal code behind : improving it means improving drupal the software as well
  • No risk to lose part of the community
  • Improving may be long
  • The current experience sucks
  • The future experience may not match a good forum / Q&A site
  • Not endorsing Drupal SE may give a bad signal to the outter world (the Not Invented Here Syndrom)
Both
  • Surely the best : a internal forum and an external (already developed) Q&A
  • Means embracing the larger Drupal community as well as keeping
  • All the other pros
  • The longest
  • Confusion between the two ?
  • All the other cons

Remaining tasks

// depends on the solution, so don't know what to say here.

User interface changes

// Same as remaining tasks.

API changes

API changes may happen if we revamp the core forum module.

jerdavis’s picture

@DjebbZ

Some folks have already begun collaborating on this here:http://piratepad.net/drupal-support-eval

Michelle’s picture

One other thought to keep in mind... Regardless of which way we go, if we decide we want to keep an actual forum either for the main support or in addition to, there will be a better native Drupal solution by next summer. It's contrib, so d.o will either need to get over its contrib-phobia or toss it to a sub domain, but it will be an option. The project is on hold until September while I finish something else but, soon as that's done, I plan on busting my butt to make Artesian Forum a reality. I know I haven't had the best track record for speedy coding in the past, but my son will be in school 3 hours a day, 4 days a week, and I can dedicate a good chunk of that time to the project. Plus, I plan to suck up my dislike of pre-planning and make an actual roadmap and scheduled releases, which should move things along faster.

A year may be too long to wait to fix this problem and I understand that. But people are talking like Drupal's forum is hopelessly unusable and it's really not. The bigger problem is d.o sticks with core. Even Advanced Forum gives more of a traditional forum experience than core forum does. We have tools available to make a better forum; we're just not using them.

Michelle

DjebbZ’s picture

@jerdavis, some pinged me in the IRC and suggested me to make it the actual summary. Didn't see your comment in the meantime (you're too fast ! :)

DjebbZ’s picture

Issue summary: View changes

Issue summary requested by webhick

jerdavis’s picture

Issue summary: View changes

Adding pros/cons from piratepad http://piratepad.net/drupal-support-eval

jerdavis’s picture

Issue summary: View changes

Updating summary based on webchick's work on the etherpad http://piratepad.net/drupal-support-eval

Fidelix’s picture

Issue summary: View changes

I moved one DSE "Pro" to "Con". The data ownership part. I also added "strong" in some parts.

Fidelix’s picture

Issue summary: View changes

-More and more frequented and -Well referenced in search engines. Honestly, are these pros? Can't we affirm the same thing for d.o ?

juan_g’s picture

@Michelle (#107), thank you for creating Advanced Forum and the new project Artesian Forum. According to the plans (new entities forum, thread, and post, with all posts equal), it looks like Artesian is going to be an architectural change to have the same flexibility and power of the most popular standard forums (vBulletin, phpBB, SMF, Invision Power...). That's great news, let's hope Artesian can also be used at drupal.org.

Since Artesian is a new, long term project and only for the technical side, on the other side of user support a first step for now would be the already mentioned issue #1239326.

One of the current problems is the large and growing number of new users, in different stages of the learning curve. A quote from the proposal in the above issue, about the roles of both drupal.org's support forums and third-party Q&A services like the discussed drupal.stackexchange.com:

...

Note: For cooperation in local and topic-specific groups, see Groups.Drupal.org. For all related to a specific module, theme, or distribution (bugs, tasks, features, support) see the issue queue. For core and other development, see the core issues and the mailing lists. Also available are chat, and third-party technical q&a.

For general discussion and support, see the Drupal forums:

...

(From #1239326: Update the support forums structure according to the current needs)

@DjebbZ: About the options SE, or drupal.org, or both, the summary in your comment #105 is indeed interesting.

jerdavis’s picture

I'd suggest that in the issue summary we "keep our options open" with regard to how we go about building a better support experience here on D.O. @DjebbZ's summary specifically sited improving the core forum module as the option. While this is an option for an improved support experience, it's not the only option and not even the option that's receiving the most support on this or other threads.

Just a minor nit pick.

laura s’s picture

It seems to me that if drupal.stackexchange.com is proving useful to people, we should at least link to it from drupal.org has a useful resource. I don't know that this requires DNS-level endorsement.

As for in-house efforts, stackexchange can prove to be a model for what works and what doesn't work, so that when we have a better native support system people choose to use it because it's better. I don't like the idea of enforcing a solution. Let people decide what they find useful.

Personally I use Google as much as d.o Solr for searching, and when I see a stackexchange hit, I know it's usually at least worth a look. Let's help the community work better, and let our infrastructure catch up as time and resources allow.

Michelle’s picture

@juan_g: Yes, it's definitely long term. I only mentioned it because it seemed like the argument was boiling down to "use core's limited forum which many people don't like" or "move support elsewhere" or "build a SE clone" and I wanted people to be aware, that we could have a decent-ish* forum here without a lot of effort by using existing modules.

Michelle

*I have to qualify the "decent" because there still are some sticky points, even using contrib, which is the whole point behind my rewrite. Still, though, we could be much better than we are just by adding some contrib.

DjebbZ’s picture

@erdavis, agree with you about "keeping our options open". It also means a non clearly defined perimeter so it may take even longer... Which is clearly in favor of endorsing somehow Drupal SE (available now). But I clearly understand the argument of eating our own dog food. laura s in #111 is right, analysing Drupal SE could give us some insights to build a good in-house solution.

bojanz’s picture

Note that for Artesian to be used we'd first need to move drupal.org to D7, and that's a whole different story.

Michelle’s picture

@bojanz: You don't think that d.o will be on D7 a year from now?

Anyway, I think my point got a little muddled. What I was trying to say is that there are modules that can be used right now to make the forum better with no more work needed than maybe a security audit. It won't get us all the way, it won't be as good as SE is, but it would make things better. And then, in the long term, I'm working to make a better forum module that will take it even farther an hopefully get rid of all the complaints about the forum. In the end, it will still be a forum, though, so the first thing to decide is if a forum is what's wanted for support. If not, then it needs to be decided whether a forum is wanted for anything. If it is, then let's make it better.

Michele

bojanz’s picture

@bojanz: You don't think that d.o will be on D7 a year from now?

I'm hoping it will. And the community joining forces behind Artesian would be great. Still, I wanted it to be noted.

Michelle’s picture

"And the community joining forces behind Artesian would be great."

LOL! Now you're just dreaming... :P

Michelle

VM’s picture

How can we successfully provide support to Drupal users, new and old, over the long term?

A few of us in this thread have been and currently continue to successfully provide support to users via the support forums. Perhaps not at the same level that the upper echelon would or could but we try pretty darned hard to be successful at it. They do so in spite of the missing tools many have deemed necessary for them aid in supporting users asking for help here at d.o.

That said, those who have such a strong negative feelings about the forums, and the way support is currently provided here at home and who've moved on to providing support elsewhere, really should take the time to ponder what d.o. could do short term to get them involved / re-involved here .... at home.

I've only needed questions to be asked by users so that I can aid in answering them. The users are doing their part by placing the questions in the forums. The few people who support via the forums are doing their part by trying to answer those questions to the best of their ability.

lisarex’s picture

It's probably worth asking the forum-based support-seekers what *their* needs are ... perhaps start a discussion in the General discussion forum, and link to it from the description of http://drupal.org/forum.

* beyond "I'd like the right answer as quickly as possible.

jerdavis’s picture

@lisarex It'd probably be good to expand that approach to other support venues as well, including asking folks why they use SE rather than the forums, etc. I'm guessing folks just congregate to what they're comfortable with, but what makes them make those choices?

jerdavis’s picture

Double post.

chipcleary’s picture

If we are considering eat-your-own-dogfood alternatives, I can offer another one. The Answers module provides a Q&A service something like StackExchange (it is partially inspired by StackExchange). It is meant to support focused Q&A, unlike a traditional threaded forum, and has supporting modules to enable selecting of best answers and manage a point system with user reputations. Demo at http://www.answers-userpoints.chipcleary.com .

I do see how using something like this would enable the benefits of integrating into d.o., like having a filter to make it easy to link to other items on d.o, or integrating with user profiles, or taking on the "support" requests in the issue queues.

At the same time, as the maintainer of Answers, it's also the case that this one is not as mature as StackExchange ... and DS has a vibrant group working. I can see the wisdom of jumping on the train that is already moving.

juan_g’s picture

Many users, many reasons. The fact is that, even when the "Forum" tab at the top of all drupal.org pages disappeared with the redesign, and now the forums are pretty hidden in a list of Community & Support, even so a great number of users keep posting on them. See for example the busy Post installation forum.

But of course the forum needs improvements, as discussed on this issue and on the already quoted #1239326. And a research on the many, many different needs of the users is basically: see where they go and what they ask. A very simple summary of what I've seen on the forums is reflected on that issue, in the proposed structure update for the support forums.

I think the growing number of Drupal users need as much support as possible. Any cooperation, at Drupal.org and elsewhere (SE, etc.), should be very welcome.

The title of this issue is "Endorse drupal.stackexchange.com". In my opinion, yes, if it helps Drupal users, that's the only really important point to consider. Let's have a link to it as one of the recommended resources, in addition to the Drupal.org's support resources.

jerdavis’s picture

I agree with @juan_g, we should probably go ahead and add drupal.stackexchange.org to the list at drupal.org/support as a venue for getting help. We can then still cary on this discussion of do we make that the only support avenue and fully embrace it, or do we still improve a support experience on D.O.

juan_g’s picture

@jerdavis, yes, the Drupal community is expanding, and there are many good external Drupal resources available. Drupal.org is not all the Drupal world, it's just in the center of it.

jerdavis’s picture

Yes, and regardless of how we decide to improve things here, we shouldn't hide a valuable support avenue from folks looking for help. I don't think anyone's questioned that drupal.stackexchange.com is a good resource.

bojanz’s picture

So...
1) Link to drupal.stackexchange.com from drupal.org/support. Effectively, endorse it, as the issue suggested.
2) Keep the d.o forums
3) See what we can do to improve the forums (and see if DA can help us with that).

VM’s picture

Just a thought:

Is there not a way for d.o. to monetize the forums (in some way) to aid in offsetting the cost of resources used and aid in rectifying some of the long standing feature issues? Ads perhaps?

While money hasn't been my motivation, it motivates a great deal of people. Perhaps a feature/forum could be deployed that allows for those wanting quick/high level support and would be willing to pay for it. Nickels, dimes, and quarters add up fairly quickly.

Though a side effect of such a decision with minimally paid support services could mean that, that's the only forum where questions get answered by the masses.

joachim’s picture

I can see good reasons for endorsing this, and for replacing our forums with it.

But at the same time, I spoke against the creation of the Drupal SE site in the first place. Here's why:

Consider this document: [#751826]. Sure, it has gaps, and it needs updating for D7. But it's an *excellent* resource for understanding FormAPI. I return to it time and again even though I've coded hundreds of custom forms and form alterations, so much so that when I start typing 'form' into my browser that's the first URL it suggests. Consider the value of this document.

Now consider the equivalent information on the Drupal SE site: dozens and dozens of individual questions and answers, phrased in whatever way seemed to make sense to whoever posted it. Imagine trying to get an overall view of FormAPI with those.

Drupal SE does not solve the problems we have with support and documentation -- it improves the experience somewhat, but also pushes the problem off to somewhere else where we, the Drupal community, don't have to worry about it any more. I am really concerned that shifting support to a shiny third-party platform will make us forget our commitment to support and to turning support into documentation.

juan_g’s picture

joachim wrote:

I am really concerned that shifting support to a shiny third-party platform will make us forget our commitment to support and to turning support into documentation.

About turning support into documentation, I think a good practice is to reply in the support forums including links to the suitable handbook pages. And, when necessary, improving or creating the page in the documentation, to reply similar support questions and to make the information easily available, as you suggest.

For example, some of the handbook pages I started where created because of repeated questions in the support forums, like this one for new users overwhelmed by the learning curve, or this other for builders saying their site is too slow, and so on.

So, turning support into documentation as you say is a useful workflow from forums to docs.

But anyway, the more resources for support, the better. Here, there and everywhere. ;)

WorldFallz’s picture

@lisarex-- definitely. Though most of us that regularly participate in the forums hear it over and over and over again as part of the process of hanging out there, it's always good to ask the user base involved what it is they need.

And I'm not sure why there are some who feel this is an either/or decision. Unless I misread something somewhere, I don't think anyone is suggesting shutting down SE (and there's no reason not to mention it either). On the contrary, I think the onus is on those that advocate shutting down the d.o forums to explain why it has to be one or the other-- with actual logical arguments beyond simply restating 'they suck' over and over.

Frankly, since those that advocate loudest for shutting down the forums completely never even use them, I genuinely don't understand why they even care. Does their mere existence somehow negatively impact their contributions in other areas?

There's no question there are things that can be done right now to improve them-- just like we've done with other areas of drupal.org. And definitely, artesian should be considered for the d7 version of the site.

At this point though, I think we're really just waiting on a decision from Dries-- until then anything else is pretty much academic.

David_Rothstein’s picture

At this point though, I think we're really just waiting on a decision from Dries-- until then anything else is pretty much academic.

The decision from Dries was AFAIK specifically regarding the proposal of redirecting support.drupal.org to the Stack Exchange site (since he owns the domain and is the only one who could authorize that). I don't think discussing other things is academic.

However, there clearly isn't wide consensus in this thread for doing something as bold as a redirect anyway. I think instead, @bojanz's summary in #127 represents the consensus of what to do next. Does anyone disagree with that?

WorldFallz’s picture

Seems reasonable-- the only other thing I might add to that list is to decide if the support forums should stay part of the standard forum structure or move to a q&a format. Since SE fills the q&a format, I'm not sure it makes sense to use q&a here as well but I can see logic for going either way.

DjebbZ’s picture

Agree. I have a question though : apart from typing the url, how do we access drupal.org/support from d.o ? I can't find no link d.o/community.

Mark Trapp’s picture

@DjebbZ I can't find the link either: was it part of the old design?

#127 sounds entirely reasonable to me, but it doesn't look like drupal.org/support is linked anywhere. Providing prominent mention of Drupal.SE as a support option would seem to necessitate a link on one or more of these pages:

Those seem to be the main support pages linked to from the front page.

davidhernandez’s picture

My only questions about #127 is, if it is listed as an "external resource", will that lead to other requests to have sites added there? What would be the criteria for such a thing?

rfay’s picture

groups.drupal.org should be mentioned more prominently in this conversation.

Groups has largely replaced drupal.org forums in my experience for two reasons:

1. They're focused to start. A person can choose a topic to work with.
2. You can get email notifications when somebody responds. That's a key (and easily fixable) flaw with the d.o forums.

I suspect that groups already has replaced the forums, and if we "shut them down" it would not be to replace them with SE, but because they've already been replaced by groups.

moshe weitzman’s picture

I love the brevity of #127, but not its substance. To me, it is a wimpy cop-out. With two peer support sites, you get content duplication on a massive scale. If one lands on the forums and does a search and finds nothing, one posts the question. The same scenario for the same question then plays out on the SE site. So now our brave answerers need to reply twice. And careful askers will search two sites before asking.

IMO, the web is built on hyperlinks. askdrupal.org or whatever will be just one link away from drupal.org. I think the harmful effect of "sending folks away" is small (but not zero). That harmful effect is more than compensated for by the excellent software and social norms that StackExchange has built. Mark Trapp wrote eloquently about those social norms in #83.

WorldFallz’s picture

Sorry moshe-- I'm not really sure I'm understanding your point. Yes, the whole internet is 'a click away' and there will always be multiple places to find and get support (regardless of what we decide here that will always be true).

Truly experienced searchers will google with all it's advanced searchy googleness. And careful askers will always have more than one place post if they so choose. This has always been true, was true before the drupal.stackexchange.com (i still find good drupal answers occasionally and pain old stackexchange.com), and will be true whether or not we shut down the support forums here.

Sorry, I still just can't see not having support mechanism for drupal on drupal.org by drupal-- talk about not eating your own dog food. That's essentially saying, we don't even make dog food worth eating, lol.

catch’s picture

@rfay: groups is also replacing the forums in the list of places to have been overrun by support questions the past few years.

The views developers groups (with a big 'support questions will be deleted' topic up top) has as it's two most recent posts, support questions. The high performance group is in the region of 50% support questions (and 30% job postings). I pretty much abandoned groups a couple of years ago as a place to get work done, and have only been dragged back there by D8 initiatives.

@WorldFallz - the constant support creep on Drupal.org is why so many people who no longer use the forums (or never did) feel so strongly about this issue. Because it is not that the forums are necessarily bad (although a question/answer format, ability to mark posts as duplicate, convert bug reports into real issues etc. would all help for some of the problems with them - I find core bugs reported in the forum a couple of times per month, and what to do with them?). It is more about trying to find a model that could suck back some of the overspill of support questions into all but the darkest corners of Drupal.org.

There need to be places on Drupal.org to get work done (work other than answering support questions), 2-3 years ago a lot of these were quite murky (like #drupal-dev), they are now better documented but also set expectations up front (like #drupal-contribute). However there are lots of places that are bursting at the seams.

WorldFallz’s picture

omg catch-- that is soooo very true. VM and i were tossing this very topic around in email just the other day.

The simple fact is, drupal's meteoric growth and usability improvements are bringing more and more non-technical users into the community who only care about one thing-- getting their support questions answered. period.

And to that end they will infiltrate every thing and every place they can. In the forums themselves, all the forums, even the non-support ones, are being suffocated by support questions (and usually unanswerable ones at that) that drown everything else out. I move dozens of misplaced posts every day. And I doubt very much its because it's so unclear where to post. They simply don't care and will post wherever they think they might get themselves a response.

This is actually a separate issue that should probably get it's own thread. Your point that there needs to be a place on drupal.org to get work done is extremely important. I would go so far as to argue that every place, except those specifically tasked with support, should be non-support zones.

Though I'm sure that will be seen as newbie hostile to some-- I would argue it's actually newbie friendly. Newbie's are usually their own worst enemy and, if anything, their typical behavior only serves to make actually getting support more difficult as they overwhelm and drive away supporters. It's our job to know better and provide the 'tough love' necessary to keep the support environment as productive as possible.

jerdavis’s picture

@moshe it sounds like you're in favor of pointing to SE as de-facto support? Likely to the point of shutting down the forums here on D.O.. Is that the case?

I'm less convinced that's a good idea now than at the start of this thread. Before someone makes that call, I think it would be prudent to survey folks using both systems to get their thoughts. I'd be concerned that pushing all support to SE and shutting down the forums would alienate people who've contributed there.

But that argument is double-edged. I'm also not convinced that the solution to improved support on D.O. is fixing the forums. I'd and advocate of a Q&A format that we can integrate better with other systems like issues and groups.

This goes to some of what @catch is saying as well. Let's have a big bright option on GDO when you're posting there, 'Are you looking for support?' and if they say yes - well then it posts a question rather than dropping something in the group thread. Have GDO be where discussions happen, issue queues be where work is done and Q&A where people get help. If we have everything under one roof and spend time on UX we can start to figure out how to route people where they ought to be.

@worldfallz and @vm I think it's important to point out that the value of your contributions shouldn't be measured by where you've contributed. The discussions here about how much the forums suck and should be gotten rid of aren't meant to discount the work that goes on there. It's about the tools and making the tools better. I don't think anyone here cares if you helped people by carrier pigeon, the value is in the helping of people not how you provided that help. Even if the forums were shut down tomorrow it doesn't mean you've wasted your time. And whatever course we take here I'd hope it included preserving the information that's been built up in the forums

laura s’s picture

+1 on making "Shut down Forums" a separate issue. This is about linking to d.s.c, yes? I don't feel a decision needs to be held up until the status of Forums is settled, or whether a new support system should be created here. Forums architecture has not changed since at least Drupal 4.5. It's no surprise they feel inadequate now. I repeat @rfay's linking to http://groups.drupal.org/support-infrastructure where work on improving/creating a new support infrastructure can happen.

Meanwhile, is there agreement that linking to drupal.stackexchange.com makes sense?

Michelle’s picture

@catch: If the fact that we have support forums doesn't stop people from using g.d.o for support, how would not having support forums do so? I won't disagree that g.d.o (and elsewhere) being overrun with support questions is a problem, but I fail to see how reducing the support options will improve that.

Michelle

VM’s picture

Even if the forums were shut down tomorrow it doesn't mean you've wasted your time.

Jerdavis explain to me how shutting off the forums isn't throwing away my 5 years worth of work?
Explain to me how throwing away that work helps users get support? (How many users find my responses beyond the original aid that went to the OP?)

I understand no one is trying to go out of their way to target supporters specifically. I don't think anyone here is trying to be malicious. That's not why my feelings are hurt over this discussion.

bojanz’s picture

The forums would still be there, only closed for new posts. So users could still search and find your old replies. I don't really see your point.

VM’s picture

well until your comment that's not at all how I took "close the forums" , "turn off the forums" , "shut the forums down". Jerdavis himself wasn't even sure that work there would be kept.

So great drupal.org used to have support forums. Support forums that will still be here, be found, but not be usable. If someone finds one of those threads and requires elaboration, what shall they do? Copy and paste a drupal support forum question from d.o. to SE?

If my contributions and investment in this project will have to be moved, perhaps acquia support channels are better suited for me and people like me. Feeling like I was actually part of the drupal project (without knowing how to code) has been what's kept me at it.

Code or lose my ability to contribute here at d.o. is what this feels like.

jerdavis’s picture

I would hope that the content of the forums is not lost. It's not as if I have any authority. I'd assume the content would always be kept. Who knows in what form. Perhaps the forums are just turned off to new posts as @bojanz suggested. Perhaps the content is migrated into a Q&A system. Perhaps the forums themselves are updated. Maybe nothing happens. I don't know.

It would be a travesty if the forums were just deleted. I don't think anyone here is suggesting that would happen.

Of course I don't know what will happen though - I'm not a webmaster, I'm not a decision maker, I'm not in the D.A. and I'm not Dries :) I can participate in discussions and give opinions and perhaps volunteer on efforts but that's about it.

rfay’s picture

I don't support deleting old forum content; I don't think I've heard that suggested by anyone, and several people said no.

I don't necessarily even suggest getting rid of the forums, but rather just removing them as a support option, making it clear that they're not for support, as better options are available.

So those of you who have invested in content in the forums: Please relax.

webchick’s picture

I am completely and utterly flabbergasted that you can say both that we should remove links to the forums in support because "better support options are available" than our forums, and also "please relax" to people who have invested countless hours of time and energy into helping users here on drupal.org and make this their primary contribution. How is that NOT a slap, directly, in the face? I'm utterly stunned.

WorldFallz’s picture

stunning doesn't even begin to capture it. Perhaps those of us who provide less than adequate support in the forums should leave all support to those who are clearly so much better at it...

... oh wait, none of those people actually provide support.

stunning indeed.

The message in this thread is becoming abundantly clear. So much for encouraging 'community' contribution-- apparently there's only certain contributions that are deemed worthy.

I fear that the damage already done in this thread may already be irreversible-- and Dries hasn't even weight in yet.

VM’s picture

@rfay

So those of you who have invested in content in the forums:

It's that kind of minimization of the work done by those who do the work that keep this cycle going and won't allow relaxation over this issue of shutting down the forums.

Content in the forums? Really? We're content generators and no longer contributors to the project?

dddave’s picture

Sorry, but I do read a difference between "forums suck" and "people doing their work in the forums suck". The former is true as I experienced myself from both perspectives and the latter nobody said.

Do we have after Designers vs. Developers now Devs vs. Non-Coding Drupal folks? I don't think we need to go this route and taking things a little less personal would help tremendously.

What change in the current support system would the support givers deem ok without dishonoring their work in the past?

rfay’s picture

OK, defense, defense!! I was trying to say the opposite!!! Hold on!! Don't kill me!

Dare I try again?
#150, #151: in #149 this is what I was trying to say.

1. I was supporting *not* deleting existing contributions in the forums, that we should not delete the vast array of existing content.

2. I was saying that *if* we move from using the forums for support to using something else, that we might keep the forums around for other uses (there are many) and just instruct that "new support requests are being handled in ... so don't put your support requests here".

Saying "relax" was about the issue of killing off existing forum contributions - I don't think we should do that.

Does that get me out of any trouble? Wow, I seem to have landed in it.

VM’s picture

@ dddave

We're not the in favor of closing the forums because doing so disallows us to contribute. We're doing our work there in spite of the reasons everyone else has given to close the forums. Thus, I don't know how those that do work there can answer your question beyond -- don't intentionally take away our ability to contribute and take away our pride in being part of this project. Removing our direct connection to d.o. does those things.

Telling us not to take this personal won't change the fact that it is personal. It's our workplace. It's our contribution and our ability to continue to contribute. It's our pride. It's our connection to the project. Hell it's our connection to you (our peers) whom we appreciate (even if it's unrequited by many). Sorry but it is personal because we personally invested here and want to be able to continue to do so. Not be pushed off somewhere else because there are too many questions being asked and not enough people answering them. The fact that there are more questions than answers is a reflection on everyone else here who don't aid in answering those questions. It's not a reflection on those who do. Doing away with our ability to contribute does not aid in getting those unanswered questions .... answered. It does the exact opposite.

WorldFallz’s picture

Creating content? really? really?

No, I don't consider myself a 'content creator' -- that's what the documentation team does (and does fantastically as a matter of fact).

I help people-- you know human beings? You remember them... I know you do... they are the ones that actually use the sw produced by our talented group of core and contrib developers. The same ones that fumfer around in those usability videos trying to figure out how to make drupal actually do stuff.

And I've done it all this time without demanding fancy shiny badges, post counts, karma, thumbs up, or userpoints in spite of the crappy forum tools we have... certainly not because of them. All the while nuking spammers, moving misplaced posts, deleting duplicate posts and content, clarifying documentation, keeping misbehaving users inline, submitting contrib patches, and even trying my hand at the occasional core patch when I feel I can.

Not to mention being continually harassed on a daily basis (in my own personal email) being told that it's not enough, that i suck, taht drupal sucks, that the forums suck, that i should do more, how I should do more, how i should their work for them, and on and on and on.

And now, my own 'community' says thanks but no thanks--- and then relax?

seriously? where is allen funt....

dddave’s picture

@VM and WolrdFallz

As someone whose main contributions apart from translations come from giving support I can certainly relate to your feelings. I too see with unease how lightly some suggest major steps against the forums. BUT apart from the personal problems this issue here already created we as a community have to face the support gap that exists although people like you do tremendous work.

You have good arguments against completely redirecting support to SE and my enthusiasm in favor of this solution has vanished over the last days. But reacting as strongly as you do now makes it really hard to consider any change at all. And the question remains: What change would you approve? ;)

jerdavis’s picture

This thread is quickly falling off track now. Perhaps things aren't being worded well, perhaps there's misunderstandings. Nobody is questioning anyone's dedication here or the quality of their efforts. I've only seen people praise the folks working hard in the forums, including @worldfallz and @vm. I think everyone here appreciates your efforts.

@dddave is entirely correct. There is a HUGE difference between "the forums suck" and "the forums are completely worthless" or "the people on the forum suck". What is being put forth in this thread is that the forum software is not working well enough to provide a great support experience. That's not to say that there aren't a LOT of people still trying to use them to help people or ask questions. As @worldfallz said, people are using the forums in spite of this. No one is saying that those people are stupid for doing so or that their contributions are worthless.

Our focus in this thread, or wherever appropriate, should be about how to help people find and use good tools to get the help they need, and how do we improve on the experience for support seekers and support givers.

With whatever decisions come out of those conversations I sincerely hope that not only do worldfallz and vm continue their contributions to this project and community, but that others are more encouraged to join them because we provided a better experience.

WorldFallz’s picture

Lighting someone's hair on fire and then admonishing them for running around in response is disingenuous at best.

And unlike others in this thread, I don't see this a matter of what "i approve". As we do everything else, we should arrive at a decision as a community.

However, every time I get fine with the track the thread is taking someone else pops in to insult or minimize our contributions or give us a quick little pat on the head to push us aside and pursue their own agenda.

If you want us to stop running around stop setting our hair on fire.

VM’s picture

noone is arguing that the forum software is great. That's where the coding sector of the community may want to step in to make it better. This allows the coding sector of the community to support the support givers. Give me a mentor and teach me how to make it better. I'll rise to the challenge and work to do so for everyone including those who still won't take the time to answer questions there.

Shutting the forums off and removing my ability to contribute don't make the software better.

I'm not going to apologize for being passionate about this issue. Using my passion to state that I'm in any way obstructing a decision from being made at all, is bogus. I'm fighting for what I thought being part of this project and community was about and fighting for it to continue to exist.

webchick’s picture

Woah. Sorry, rfay, for completely misinterpreting your post. Though it's a sentiment clearly echoed by some other people in this thread, I certainly don't want to attribute it where it wasn't intended.

VM/WorldFallz, Ok, what if we tried to reset this conversation a bit, more around what would make your contributions easier (which is exactly where this conversation should've started, but it's a little too late for that now :\).

There's a list of proposed features outlined at http://drupal.org/sandbox/HedgeMage/1088372. Do you agree with that list? Are there things in that list you would prioritize over others?

Do you have ideas based on your experience on what motivates people to contribute to the forums, or input on what causes active contributors to the forums to leave?

I'd love to see us develop a concrete plan around this, but about 98% of the people who've chimed in on this thread so far don't have the capability to come up with one because we lack your experience (or have let it atrophy). We're guessing at the needs of our users, and we're guessing at the needs of our support contributors. We need your help.

VM’s picture

Webchick,

I don't think my prioritizing that list aids. I'll do it though and I'll start right after this comment.

I think we're all in agreement that what is needed is more people to answer questions. Thus how do we get others involved in that capacity. So as to limit the # of unanswered questions and get the individuals who know more than me involved in support. The support channels will always languish regardless of where they are placed if those who know more aren't around to answer questions. IMO this gap happens because their knowledge doesn't get passed down to those who spend the most time doing the support.

So it really boils down to how do we get people in the upper echelon involved in answering the questions general supporters can't. For when the upper echelon does this, general support learns from them and then passes that learning on. Otherwise there will always be a gap in what can be passed on.

That said and not to intentionally single anyone out, I really think the question is what would aid people like rfay (those with his type of knowledge) to contribute to support beyond IRC where that knowledge doesn't come back to d.o.

When I first started, I was always thrilled to have a response overridden by someone with more knowledge than I. I was thrilled because I learned something and could pass it on.

Trust me, I know how busy the upper echelon is and how difficult it is to make time for that kind of commitment.

davidhernandez’s picture

webchick, I hate you for stealing my pragmatic thunder. I had a nice big rant I was going to submit?

Anyway, I support WorldFallz, VM, or anyone else opening up some actionable discussion in the group - http://groups.drupal.org/support-infrastructure . It should be a general group to discuss this matter, even if some of the focus is shifted to the forums.

There is a lot of complaining going on. People who hate the forums need to respect the opinions of those who use them, but the people who use them should also try to understand why those people hate them. Often the reasons why people don't use a service are as important to understand as the reasons people do. I think it would be a nice goal to try to drive more contribution, to assist people like WorldFallz and VM in their endeavor. Everything I do here is with the mindset of trying to be as inclusive as possible, listening to all opinions, and getting as many people involved as possible. I want to help you regular forum users make it a better place, so some of those people that complain can be enticed back. And it would be far more productive to keep these two groups communicating, so we can make some improvements. The last thing we should want is for people to leave the discussion and have decisions/changes made without their input.

juan_g’s picture

A problem is that this issue is a mix of several questions. A possible solution would be to solve those issues in their respective appropriate places. An overview of the different questions can be the following:

  1. Should we endorse drupal.stackexchange.com?
    That's basically the title of this specific issue. It seems we agree that SE is a good technical Q&A site, and is helping support Drupal users. Since we should focus on the user needs, probably we should consider, when linking to resources, not so much if they are or not at drupal.org, but rather if they are useful for Drupal users. However, this has to be decided on this issue.
  2. Should we slap a support.drupal.org CNAME over drupal.stackexchange.com?
    That was basically the initial subtitle. ;) There are different and passionate opinions, but not a clear agreement supporting this idea. One of the several reasons is that SE is good for short technical questions, but for example not a suitable place for new Drupal users. So, if we shut down all support at drupal.org, and new users do not have a place to go, they are going to flood groups.drupal.org and other workplaces.
  3. What kind of support should we have at drupal.org?
    Assuming we don't remove all support, should it be a discussion forum like now, or a Q&A system similar to SE, or both, or what? The place where a new support.drupal.org is being discussed is the new Support Infrastructure group. However, this and SupportDO are probably going to be a medium or long term work, so we only have the support forums for now.
  4. Should we improve the support forums?
    There seems to be agreement about this, at least until a future drupal.org support system, with or without forums, will be in place. Suggested ideas, that can be easily done right now, have been the following:
    • The addition of some Q&A functionality to the forums. SE has for example topic tags (similar to free tagging), and voting. I think that, at this time, there are not drupal.org webmasters issues discussing adding free tagging, etc. to the support forums. Naturally, if there is enough interest in a feature, a new issue should be opened.
    • Other suggestion has been to use Drupal's Advanced Forum for now, and later Artesian Forum, for the support forums, in order to have similar features to standard forums like phpBB, etc. This would also need its own issue.
    • Another proposal that is being discussed is the issue #1239326: Update the support forums structure according to the current needs, for example to adapt the support forum to the growing numbers of new Drupal users, etc. This specific proposal (new forum structure) can also be done very quickly as well, if approved.
webchick’s picture

davidhernandez: hehe, sorry! ;)

VM: So I can personally speak a bit as to my transition as one of the "upper echelon" (I don't actually like this distinction; everyone dumping hours of their own time into improving Drupal/Drupal.org is equally "upper echelon" as far as I'm concerned) from someone who used to spend a very significant amount of time contributing to support back in 2005, to someone who now generally avoids support channels altogether in 2011, despite my innate drive and desire to help people not having changed at all in those 6 years.

In 2005, I was accepted as a Google Summer of Code student. Mind you, I didn't know anything about Drupal back then, and had never even installed it. But I needed to go from a place of zero knowledge to module contributor knowledge in 2 months. By far, the most effective approach for me to get up that learning curve was to hang out in #drupal, #drupal-support, and the forums, look for questions that I thought I might be able to figure out, and attempt to figure them out. Either I would, and I would learn something, and immediately be able to help someone else, or I wouldn't, and could learn from others' answers. I remember the first time I actually answered someone's question I felt SO bad ass. :D I also spent a significant amount of time back then adding/editing documentation as I was learning: pulling useful tidbits from the forums/IRC and putting them into the handbooks so both myself and others could find them later. I probably spent 20 hours a week doing support/documentation work for fun, and it was great.

However, as I gained experience with Drupal, I also gained additional responsibilities. By the end of 2005 I was working on the core development team. By 2006 I was also on the webmasters team and the security team, and was also managing Drupal's Google Summer of Code involvement. In 2007, I was on the team who helped found the Drupal Assocation, and led efforts around the Google Highly Open Participation contest. In 2008 I was promoted to a core maintainer, and that was pretty much the end of my support career (and everything else), beyond the occasional thrill I get to walk someone through their first patch on IRC or at a code sprint. The last kiss I gave support was writing the Using Drupal book, to try and spread around what I knew to as many people as possible.

Basically (prior to joining Acquia, anyway), the amount of volunteer time I can give to give to Drupal has remained relatively constant over the course of 6 years, but the things demanding attention for that time have grown exponentially, particularly around things that only I can do. As a result, I can honestly not really think of anything that you could do to get me back into the forums again. :( If I have 10 minutes to give to Drupal, it's simply much more effectively spent reviewing/committing patches than it is helping someone figure out Views.

I am a bit of an extreme case, I realize, but this same story rings true for others in this thread, as well. chx went from being a general member of the core developer team to one of the chief engineers working behind the Form API, menu system, database system, field system, testing framework, and update system. rfay went through a transition from someone who just wanted to fix a freaking bug to the leader of the Examples project, one of the primary architects of the automated testing framework.

We all (core developers, testing framework maintainers, documentation authors, and support question answerers alike) share a common bond: we have the personality type that we see something wrong we want to fix it. We also have a personality type that it's hard to say "no" to things, because we genuinely want to help. But when you're honestly maxed to the brim, something has to give or you completely flame out.

Support, unfortunately, is often that "something" for more advanced people. Because as you gain in experience, you no longer are scratching your itch when you help with support (as I was in 2005, when it directly helped me to learn Drupal). Now, you are either performing charity work or doing it because it's something you enjoy, and it's a hobby. Most of the people in this thread, particularly those you refer to as "upper echelon" don't have the luxury of hobbies/charity when it comes to Drupal.

Support is, however, one of those things that is accessible as a contribution method to pretty much anyone with even a passing understanding of Drupal. I go around and give this talk at DrupalCamps a lot about how if you have even installed Drupal, you already know more than a whole bunch of other people, and you can help them. There are also, just by sheer numbers, a lot more people who are in that "scratch your itch" place (actively in the process of banging their head against the learning curve) with support than are with development tasks. So support can scale, if we can figure out how to encourage people to do it.

VM’s picture

Then I've misunderstood what people are talking about when they speak of "the gap" that exists. I was associating it with gap of knowledge.

Sorry about the verbiage choice, exchange upper echelon for more advanced/knowledgeable users.

At this point, I think I'm going to tap out and submit to the democratic process.

+1 for SE -- or other alternative --

Meant to attach this list: As someone who worked in the forums supporting drupal.
None of this was necessary to get me involved in aiding users. Therefore, I don't really see any of it as a priority over the objective and can't really see any priority in one over the other.

The only one that makes any sense to me is possibly the vote up idea. Specially on the question of comments, which would be the answers given. Thus aiding in not confusing users when there are multiple answers some of which are clearly incorrect and clearly correct. As we know sometimes with Drupal there is more than one correct answer.

Tagging questions with version # may also make sense. However (no idea of performance implications) may make more sense to have a Forums based on version. Deprecate the support area (read only) as versions of Drupal get sunset. Thus only keeping active versions of Drupal, active in support.

I'd be a bit careful with flagging for moderation. I can see users abusing this if they think it will get them an answer.

#1122212: The ability to vote questions and answers up or down. (What about comments?)

#1122218: The ability to tag questions by topic and Drupal version so that people can follow exactly the content that interests them

#1122202: The ability to post and answer questions
#1122204: The ability to post short comments on answers and questions
#1122208: Some sort of points tracking
#1122216: The ability to close or merge questions that are off-topic or duplicates
#1122222: UI feature that will suggest possible duplicates as the question is typed.
#1122198: Filter to make linking to drupal.org nodes faster/easier
#1122230: The ability to flag content for moderator review.

bojanz’s picture

I am glad we've taken a step back and started talking about what we can gain from all this talk and emotions.

The main problem that Drupal support has in any of its forms (issue queues, groups, irc, forums, stackexckange) is the problem of noise.
An increase in users also brings an increase in support requests everywhere, and many times those people are asking the same questions, but getting different (sometimes half-baked) answers. And this noise and this repetition leads to exhaustion on the behalf of support givers. I've spent most of last winter in the Views queue as a part of the bug squad, and could catch myself becoming more and more irritated, sometimes even snide, because of that noise and that repetition, lack of energy put into asking questions. And I really don't want to become a "RTFM, idiot, noob!@#%" kind of person. I'm not alone in this, and that brings down the general friendliness of the community, and leads to busy people avoiding support altogether (#drupal, groups, support requests in the drupal queue, less and less active coders in the forums).

What I found in StackExchange (and I'm sure many other people who are also a bit disenchanted did) is a way to reduce the noise. A question can be downvoted, and so can an answer. A question can be closed, or pointed to an existing answer. The goal is to have a definitive answer to an exact question, and to build upon that, instead of asking the question again and again with different feedback in return. And the system of points regulates that, while also giving a reward to occasional support givers. If you have 100 people answering 10 questions a week, then you don't need 1 person answering 1000 questions a week. Occasional contributors need a way of knowing their work is recognized, and on d.o currently that is very opaque and invisible. I look at the nicks "VM" or "Worldfallz" or "rfay" and I know the awesome effort that they've put into the project, and respect them for that. Someone new might have no idea. The git migration has improved this with a way to give credit to commits, and you can see that the people have been very enthusiastic about it, it has shown to be more important than we might have anticipated. When "community karma" is visible and in sight, people know that improving it also improves the rewards they get back (faster support from others, attention to feature requests, or just getting their voice heard through the noise).

So, I'm opening all sorts of themes that could go on for days, and already have on other issues, drupalcons, etc.
The TL;DR version is this:
1) Better ways of regulating the signal vs noise ratio
2) Better ways of rewarding and attracting occasional contributors
3) Better ways of turning support into facts and definitions ("promote to docs", "promote to issue", "vote on answers")
4) All this leading to better ways of channeling people to right support places, reducing frustration on all sides.

And then recognizing that the reasons above are why people are advocating for SE. And I am sure that even if we kill our forums completely and redirect everything to SE, this problem will remain if the underlying issue doesn't get enough attention, because fundamentally it is about providing direction to people, and helping them help each other.
You could then tell this same story, but replace support with code contributions, and it would still matter, except that when it comes to code we've been faster in improving it.

VM’s picture

bojanz

I wish you (and everyone) the best of luck with finding some place quiet to work. However, I really don't see that happening when it's publicized where you're working. IRC was fairly quiet (if I recall correctly) until it was widely publicized that answers can be received there. G.d.o didn't seem that noisey until the noise could float there without having to register and login specifically to g.d.o.

The only way I can see to legitimately reduce the noise is to educate the noise makers. Voting the noise down won't make the noise stop. If anything this will increase the noise made. If the noise isn't made here it will be made elsewhere.

Note: This is not advocating any idea to start making drupal more difficult to deploy.
Ultimately now that grandma has a much easier time installing drupal and using Drupal out of the box. She doesn't have much of an idea of how anything else works. The easier it becomes to deploy Drupal and do a myriad of things. The more the noise makers will have more in depth questions that they have a difficult time asking and support will have a difficult time answering. I know I've spent many hours, over multiple days, whether in the forums or issues queues in what began as a single question from an OP and often times needed to drag enough information out of the OP in an effort to understand the root of their issue. Add that to the plethora of follow up questions that a support answer can generate for the OP and one issue can take several days to square away. I never gave up on answering a users question regardless of how long it took or the level of silly/frustrating the thread got. I stuck it out. I did this because in the end I always felt it better that the user was always seen to be getting help and not just voted down into obscurity. When drowning I don't think anyone is going to care much how they were saved. Just that they were saved. Even if the technique used in the saving of victim he/she felt they were man handled a bit. Perhaps they just won't thank me afterwards.

Best I can tell, it's just the nature of the beast. From where I've sat these years the noise has always seemed to follow wherever it is that Drupal people are or are thought to be. I understand this is an open community. It's just what I've recognized in my time here. The thinner the walls in a structure are, the more noise that travels through them. This is not put forth in a way to tell you or anyone who wishes for a quieter working space to wall yourself off.

cweagans’s picture

Status: Needs work » Postponed (maintainer needs more info)

HedgeMage wrote a really detailed post about Drupal.org support back in January. http://www.binaryredneck.net/node/175

@VM, I think you're missing the point that Bojanz was trying to make (correct me if I'm wrong). If we focus on the questions that they need answered, and find a way to concisely answer that question (or work with that person to develop their question to the point where it is answerable), then we will have come up with a successful (IMO) support system.

I know that I don't spend a lot of time in the forums (as in, I don't spend -any- time in the forums) because I believe forums in general to be a waste of time. They are great for conversations and discussing things, but I don't think forums = a quality support mechanism. Again, I'd like to reenforce the idea that there's nothing wrong with the /people/ doing the supporting - just the technology used for providing the support.

StackExchange is a good alternative that has already solved a lot of these problems. If we are set on rolling our own, then we need to come up with a spec (most of which is already spec'ed by HedgeMage: http://drupal.org/sandbox/HedgeMage/1088372).

Personally, I think that we ought to do one of the following:

  1. Deprecate the forums in favor of StackExchange (do not delete the data - just disable new threads and refer people to SE); or
  2. Deprecate the forums in favor of a yet-to-be-spec'ed-and-constructed support system built on Drupal hosted on our infrastructure

I realize that it would be quite a change, but I really believe that it would be for the better. In my opinion, forums are going to go the way of the dinosaur in favor of more focused communications mediums, such as StackExchange or similar.

So who's going to do all of this work, you ask?

I will volunteer a reasonably significant amount of time if there is demand for a system as specced by HedgeMage. I'm sure there will be a couple of other people that would be willing to help as well.

The main thing is, we can't bikeshed forever. At some point, we need to take action. However, we're sort of blocked on a decision from Dries. So marking this as postponed until we have a decision.

VM’s picture

Status: Postponed (maintainer needs more info) » Needs work

The main problem that Drupal support has in any of its forms (issue queues, groups, irc, forums, stackexckange) is the problem of noise.

K, anyone wanna clarify for me what is meant by noise? Since I'm clearly living up to my SN ?

VM’s picture

Status: Needs work » Postponed (maintainer needs more info)

erp, cross post

catch’s picture

omg catch-- that is soooo very true. VM and i were tossing this very topic around in email just the other day.

The simple fact is, drupal's meteoric growth and usability improvements are bringing more and more non-technical users into the community who only care about one thing-- getting their support questions answered. period.

I completely agree on this but would also qualify it. In 2005-8-ish a lot of people asking support questions were also 'scratching their own itch' - building sites for themselves, trying Drupal out etc. that's how I got into Drupal myself.

In the past 2-3 years there has been an increasing number of people who are working on development projects for their day job (i.e. already PHP developers, themers, designers etc.), however they also have that same attitude that they want their support questions answered and that's it. There can be various reasons for this (they may have just walked into a job and they happened to be using Drupal, and had no part in making the decision to use it, or some people are just entitled arseholes), but it's a general pattern that is becoming more and more prevalent.

With the first group, they are spending their free time figuring out Drupal (as was I when I first posted on Drupal.org), and so am I (albeit with 6 years more experience). If you answer their question then they may or may not one day answering someone elses, but it feels like an equal exchange (especially if they ask the question well and made some effort to research first).

With the latter group, they are getting paid to answer the question, but the person answering is doing so as a volunteer (who might earn a living from working on Drupal, but not at the moment they're answering that person's question). This is not really an equal exchange any more, and it doesn't help when some of the people answering questions are more demanding than paying clients.

webchick made the point about charity.

This varies culturally, but in the UK if you see someone trying to carry a pushchair with a kid in it up a flight of stairs, a lot of people will stop to give help them carry it. Sometimes you yourself are carrying a heavy bag or a pushchair but no-one is going to try to guilt trip you for not doubling the load and risking you all ending up in a broken heap at the bottom of the stairs.

However, if you happened to be walking past someone's office and they ran out and started screaming at you to help them with their filing you would run away very, very fast, or punch them in the face then run away fast. (this probably does not vary culturally so much).

I also think we need to clarify what's meant by support. If webchick is working someone through their first core patch that is 'giving support' in the colloquial sense, but we (or I at least) don't think of it as answering support requests - it's helping people to contribute to Drupal.

Some people start by thinking they're going to just get their bug fixed, but end up re-rolling a patch, testing it, writing a test or whatever by the end - that is how people get hooked. However the massive, massive number of people who are firmly 'consumers' (or if they persistently ask very basic questions without research, 'help vampires') of the various support and non-support channels mean that any attempt to answer questions runs the risk of turning into a multi-hour marathon, VM described the same thing.

And to that end they will infiltrate every thing and every place they can. In the forums themselves, all the forums, even the non-support ones, are being suffocated by support questions (and usually unanswerable ones at that) that drown everything else out. I move dozens of misplaced posts every day. And I doubt very much its because it's so unclear where to post. They simply don't care and will post wherever they think they might get themselves a response.

Right, there are 'good' support questions and 'bad' support questions. Part of what StackExchange has, and we absolutely don't is a way to promote the good questions and demote the bad ones. The issue queue has a way to demote the bad ones (postponed 'needs more info' and all the other various black hole statuses), but no way to promote the good ones so it is not able to accomplish this either yet.

This is actually a separate issue that should probably get it's own thread. Your point that there needs to be a place on drupal.org to get work done is extremely important. I would go so far as to argue that every place, except those specifically tasked with support, should be non-support zones.

When chx opened this issue he also opened #1236280: We need a central place for core contributors broadcast their business within the same few minutes. They are definitely two discussions, but I think this issue is finally getting towards what I hoped it would be about - which is neither just the quality of support infrastructure (as opposed to support givers), nor the lack of places to get work done, but what the problems really are at the moment and how they are affecting people - which is often not being openly discussed for what it is.

Though I'm sure that will be seen as newbie hostile to some-- I would argue it's actually newbie friendly. Newbie's are usually their own worst enemy and, if anything, their typical behavior only serves to make actually getting support more difficult as they overwhelm and drive away supporters. It's our job to know better and provide the 'tough love' necessary to keep the support environment as productive as possible.

Yeah exactly, I wouldn't use the term 'newbies' necessarily (although good newbies often do basic research first so they hide their newbieness when actually typing the question), but there are lots of people who use Drupal.org dysfunctionally. Some of the behaviour on Drupal.org I would class as worse than spamming us with jewelry ads. Deleting a jewelry ad only takes a couple of seconds, responding to "I want facebook clone one week scale 1 million users" or navigating duplicate issues sucks time and motivation from people who could be doing other things.

K, anyone wanna clarify for me what is meant by noise?

These are probably not great examples, but they are ones which I recently ran into.

There is this bug report about PostgreSQL public schema #1060476: Multiple issues when PostgreSQL is used with non-public schema. It's a pretty low level bug in core that only a handful of people are qualified to fix.

There are several duplicates in the core issue queue #1219674: Postgresql not public schema upgrade error, #956942: Use current postgresql schema instead of hardcoded "public". , #1227004: Multiple issues with PostgreSQL driver and public schema. Then there is this forum post in 'post installation' with the same bug http://drupal.org/node/1177322. What can be done with that forum post? (apart from adding a link to the issue) - can't mark it as duplicate like the core issues can be. What can be done to force people to search before opening new issues (well we know the answer to that, it's #19386: Automatically search for duplicate issues/questions before submitting new issue/question, but that issue doesn't yet deal with duplicates opened outside the issue queue).

There are a tiny fraction of the Drupal community who understand PostgreSQL, there is an even tinier fraction who also works on core bugs. Every time people create duplicate issues or forum posts they decrease the ability of the people who can fix those issues to actually do so. I'm not going to fix PostgreSQL issues but there are a load of other things I can be doing in the core queue other than closing duplicates.

And this example is a very focused, very specific bug report. I'm sure there are some questions that have been asked maybe hundreds or thousands of times by now on Drupal.org

think we're all in agreement that what is needed is more people to answer questions.

That is definitely part of what is needed, but ways to ensure that those people aren't answering the same questions dozens of times is another way to look at this, and that is part of why people are looking to SE for an alternative to the forum format.

VM’s picture

Thanks for clarifying catch. I appreciate it. As stated if SE is thought by the majority to be the answer to these issues then +1 to fixing them the way the majority sees fit. I certainly don't want to be/feel like I am the albatross around the neck of the drupal project or feel like I'm an obstacle for others to support the users of the project. It will be an interesting human experiment to watch from the stands.

I'm all for making the project better for everyone.

webchick’s picture

I don't disagree with anything catch said, but I do still fail to see the correlation about how endorsing SE magically fixes all (or even any) of these problems. New and/or "vampire" users simply do not have the experience we all do on drupal.org to know that node type X is for this and node type Y is for that. I think that VM has a much more realistic view of the situation when he said earlier "the noise has always seemed to follow wherever it is that Drupal people are or are thought to be." The answer to this problem is patience with educating people and/or acceptance of this reality, not putting our heads in the sand and attempting to redirect these people out of our community.

webchick’s picture

Or, of course, combatting this with additional Drupal.org website features. I'm starting to get a better idea of what that might look like:

1. The ability to seamlessly move things from the place where you report bugs to the place where you post support questions (and vice-versa).
2. Make it possible to rate quality of content on Drupal.org, and make it clear to a casual visitor "This is a really good reply" vs. "This question is crap".
3. Make Solr aware of that, too, and have it impact the search results so people can better find answers to their questions without having to ask them multiple times.
4. Some sort of a workflow for support questions, so it's clear to tell if they're "answered" or whether they're "duplicate" or whether they're "I have no idea what you're talking about," like we have for issues.
5. Auto-search for possible duplicate content when submitting a support topic/issue and present people with a list to look at before letting them post.

Those all seem pretty doable to me, especially if there's funding involved. But it would be easier to justify funding this effort if support providers, support desirers, and code contributors were on the same page that these were desired improvements to our infrastructure.

juan_g’s picture

Yes, many of us also think that improving and increasing, not reducing, the current support resources is the way to go.

Logic says, the less available support resources to attend users (like removing the support section, and at the same time removing the motto on the homepage, "Come for the software, stay for the community"), the more support questions will overflow, not only SE, but also non-support places at drupal.org. And on the contrary, the more and well advertised support resources where users can go (at drupal.org, and SE, etc.), the less unwanted support noise in workplaces. So, yes, let's increase and improve the support resources.

I was thinking of a possible future conversation:

- About that CMS, Drupal, that you proposed for our organization site, can it help manage user support?
- Yes, of course.
- But see, the official Drupal site is not capable of it. They outsourced all user support to Stack Exchange.
- Oh...
- We are considering now Microsoft SharePoint. Can you help with that?
- Er...
- That's all, thank you.

VM’s picture

it would be easier to justify funding this effort if support providers, support desirers, and code contributors were on the same page that these were desired improvements to our infrastructure.

If the tools were available, I would certainly use them.

A thought:
What's not addressed by any of these things is the old crusty stuff that users shouldn't be using any more and that aid in causing confusion. Currently, if someone finds and implements a 4.6.x response to a question regardless of how good the response is/was its unlikely to work for them.

IMO there needs to be a way to be able to deprecate old stuff (yes even if it's mine) so that it can't easily be found by users who couldn't possibly be using that version of Drupal (and if they are they shouldn't be.)

I don't have a clear way to articulate this idea at moment beyond using forums as a diagram. NOTE: the end result doesn't have to be in the form of forums.

Drupal 6.x
Themeing Questions
Module Questions
blah blah

Drupal 7.x
Themeing Questions
Module Questions
blah blah

The above topics are merely generalized for the sake of getting the idea out of my uber thick dome.

When Drupal 6 is sunset those forums get turned to read only. Perhaps until Drupal 9 is released when those forums could possibly be completely removed (if DB size/resources are an issue) or hidden (if we want to keep legacy stuff for archives). This would allow for datamining if necessary or pulling stuff forward if it's still relevant to newer versions of drupal if possible.

I believe this minimal degree of separation will aid in concentrating users and allowing for focused aid to those users. I also think it may aid with getting users helping other users because; if someone is using D6, and they are in the D6 support area, then it's highly likely the others in that area are familiar with the same version and doing the same kinds of tasks with the same modules and same versions of those modules.

This may also minimize the situations where the first response from support is a question in the form of "what version of drupal is in use?" or the responder feeling as if they need to try and cover both versions of drupal in the same response and has to recite two different methods in the same issue if they know it's drastically different for each version. Think of the way paths have changed between 6 and 7. Anyone using 7 who comes across a response of mine from D5 or D6 goes off to find admin -> themes -> configure to change post display information won't find it there in 7.

Realistically a great deal of the problem with some of the old forum posts (now and in the future and whether Q&A format or forum format) are the same things the doc team runs into with documentation where tagging per version isn't enough thus two different explanations about how to accomplish the same task reside on the same doc page. Some users pay close enough attention to what they are reading ... others don't.

@all
My argument here hasn't specifically been to keep the forum format and I hope the above isn't seen as such. My argument is also not an attempt to stop SE or limit where support can be found. My argument has been to be able to remain a part of this community not have to go off and be part of someone else's community.

** I'm sure I'll edit this at least 10 times for clarity. Yes I typed it offline first.

juan_g’s picture

webchick wrote:

The answer to this problem is patience with educating people

Yes, I think you mean, for example just replying to misplaced support questions:

"Please go to drupal.org/support"

where there are links to documentation and support forums, and where an also useful link to SE as a "third-party technical Q&A site" can be added.

And as well, instead of just having a notice like (a real example) "This group is NOT for seeking Drupal support, or asking questions about how to make Drupal work.", adding as well something like "Support is available at drupal.org/support."

DjebbZ’s picture

@webchick in #175 : basically you're trying to create a Q&A system in d.o, a bit inspired from Q&A Drupal SE (edit), but fully integrated with d.o. I have a feature request : as we all want for issue queues, we also need here the famous "Subscribe/Unscrubscribe" button, and also some public statistics like in all forums (and Drupal SE as well) so we have a way to analyse and reward help.

The key feature in your "getting a better idea of what that might look like" to keep the support under our home and transform it into contribution, is the ability the move from "forum post" to "issue queue" and vice-versa. This is where I see the forum useful, because it won't be anymore this place where mostly comsumers and help vampire wander. If this is the direction we're taking, I'm all in favor of changing my mind and keeping the forums altogether (but still endorsing Drupal SE in d.o/support as "third-party technical Q&A site"").

All the other features you mentioned serve another purpose, very well described by @bojanz in #167 : noise reduction. If we put energy in making a good Q&A system at home (d.o) that helps to reduce noise, answer questions and move support seekers to contributors, it's a win.

We just have to admit that it will take some time, energy, and I'm pretty sure money (DA, I'm looking at you), that in the meantime noise will continue to raise in our current channels, and that in the end, we'll be competing with Drupal SE (an internal Q&A platform vs. an external one - but do we really care ?).

catch’s picture

Just to clarify:

and that is part of why people are looking to SE for an alternative to the forum format.

..since I think that might have been misconstrued. I meant SE as in Q+A format, and also from the point of view of wanting it ASAP, not necessarily just pointing everyone to it.

Actually moving stuff to SE in an organised way would not be a tiny effort, we have mothballed development forums on Drupal.org that have been 'closed' for over two years yet still get posted on. And it still doesn't deal with the bug vs. support request issue (where you want to actually move those to the issue queue or have them clearly marked duplicate).

If we're able to do the things in #175 though, and quickly then that'd be great.
#175 includes things that have been wanted in the issue queue for many years, so we are not only talking about completely building new infra on d.o for support but also improving the issue queue for bug reports etc. some of which can be re-used by a dedicated Q+A site if that's part of it.

This doesn't stop just adding a link to SE from http://drupal.org/support. It links to commercial companies that you have to pay for support already.

juan_g’s picture

Also, as suggested by DjebbZ and Mark Trapp in #134 and #135, links to drupal.org/support should be added to d.o seccions like Community & Support.

See also #314325: Redesign the support page. A possibility would be to have a header tab for Support, in addition to the current one for Community.

Damien Tournoud’s picture

The "Drupal community" should not and is not limited to "*.drupal.org". With the growth of Drupal, we need to provide a better support experience for new users. That comes through:

  • Better tools. [... a lot has been said in here already, but it starts with input formats: to post on Drupal.org you need to understand HTML. That's not something a lot of new users know]
  • A large community of people answering questions, and a way to transform people asking questions into people answering questions.

Stack Exchange provides us with both. Let's be smart and take up the offer. As I said before, the community is *already there*, so I see no reason not to link to there from drupal.org, as I did even before this thread was open (my edit of the /support page was reverted by killes).

WorldFallz’s picture

Seems to me this is still somewhat theoretical until a decision is made, but assuming we go with improving our own tools, #175 would be an amazing start!!

Some other things I would add to that list:

  1. tag vocabulary
  2. in conjunction with #1 a tag cloud—they can help people find issues using terms they might not otherwise know (especially when it comes to our Drupal jargon).
  3. perhaps a moderator only field for marking stuff deprecated. I’m not sure about the deprecating forums by version idea since a lot of stuff applies to multiple versions which means valuable info will be automatically buried/lost when a d6 forum is deprecated even though some answers will still be usable in d7 and d8.
  4. regarding input formats/filters:
  5. a more detailed search interface (more like the exposed filters in the issue queues than the 1 or 2 solr facet blocks we have now)
  6. some good associated block views right on the main forum page-- top voted posts, most recent posts, most viewed posts (though I’m not sure we track page views in a way we can report on), most active posts, etc.
  7. some form of site maintainer / forum moderator/ user admin ‘badge’ – not for accolades' sake, but for practical reasons. For example, when a post is moved by a moderator it should stay moved— having people understand that a mod moved it will lesson their desire to move it back.

And though by no means ‘critical’, a bbcode or bueditor type post helper would be welcomed by users.

And of course, the #1 request has always been 'subscribe' capability-- but that's already being addressed in one of the other improvement issues.

Also, I’m totally -100 to a moderate for review flag— it’s not a question of ‘if’ it will be abused but how quickly it will be abused.

kiamlaluno’s picture

To notice that it could also possible to link back Drupal Answers to Drupal.org. In other SE sites, users are asked to propose ads that shown an image with a short description and that links to a site; users propose the mini-ads, and the most voted are using in the site.

It is not possible to do this on Drupal Answers, as the site is still in beta phase, but other sites decide about those ads every year. If that would be done for Drupal Answers too, it would be possible, for example, to add an ad for a specific initiative that requires help, or add an ad that points to the group page for the proposal requests; it could also point to the project page of a module that Drupal.org wants to use, and that needs some help in the developing. The text in the ad should made clear that is not linking to a module that is being suggested to be used.

As Drupal is a GPL project, there should not be problems on creating such ads; at least nobody would be complain that the ads are added from who takes a direct financial benefit from showing them.

catch’s picture

I've helped admin a site which has a flag for moderators thing, and I agree it's a complete nightmare. I'm sure people click it not knowing what it even does.

ccardea’s picture

1. The ability to seamlessly move things from the place where you report bugs to the place where you post support questions (and vice-versa).

I think this would probably look something like what we have with project issues, where they issue can be moved from one project to another. I would like to see both a support queue and an issue queue for each project, with the issue queue being restricted to problems that have been identified as actual bugs, and with special permission required to move items from the support queue and to post in the issue queue.

juan_g’s picture

WorldFallz wrote (#73):

I've been completely unsuccessful thus far in getting any improvements actually brought to fruition (after all, we're still left without even so much as a forum search on the new drupal.org even after all this time).

Indeed, there is something here. We can try to analyze a couple of the several causes of the current situation.

So, let's say an innocent new user comes to the Drupal homepage, and see:

Refine your search

  • All
  • Modules
  • Themes
  • Documentation
  • Forums & Issues
  • Groups

He selects forums, and lands in the issue queue. He thinks, "How nice of them, to have this support forum to spend their time helping me build a YouTube killer on my Go Daddy shared account".

Why in the world are forums and issues in the same option?

And another even more serious problem. If we remember, until October 2010 there were two header tabs for support on all Drupal.org pages: "Support" (drupal.org/support) and "Forum" (drupal.org/forum); see a screenshot.

And now there are zero support tabs. But wait, there is "Community & Support" now. Well, that's not a support page, it's a community page (drupal.org/community), with the forums lost among many community temptations like groups.drupal.org, mailing lists, etc. Naturally, because of the page title, new users think all the community is a support place.

And we wonder why support questions in non-support workplaces have recently increased...

joachim’s picture

> The ability to seamlessly move things from the place where you report bugs to the place where you post support questions (and vice-versa).

Unfortunately, moving things around when that entails changing the node type is not something Drupal's good at at all... I'm wondering how we can approach that.

ccardea’s picture

5. Auto-search for possible duplicate content when submitting a support topic/issue and present people with a list to look at before letting them post.

I think duplicate questions provide useful feedback. I think it would be possible to build a system in which questions that are flagged as duplicates could be used to generate a FAQ. Of course that would require a human being to flag the duplicates.

ccardea’s picture

@joachim That's a good point. Maybe it's not possible to have seamless movement if you're using different node types. Then it becomes a question of what is more important to you, the ability to move seamlessly or the ability to separate support questions from bug reports?

Michelle’s picture

There's a node convert module that changes node types quite nicely. I have no idea if it's "d.o worthy" but I've used it in the past and it's worked great.

Michelle

ccardea’s picture

Re: my post #190 - Page views is probably a better measure than duplicate questions.

juan_g’s picture

@Michelle: I think node_convert supports forums, but not sure if it supports issues.

ccardea wrote:

what is more important to you, the ability to move seamlessly or the ability to separate support questions from bug reports?

Prevention is better than cure...

juan_g’s picture

About the support bugs mentioned in #188, these are two screenshots of how it looks now, and how it can look after fixed, with the Community (/community) header tab named correctly, the Support (/support) header tab restored, and separate search options for Forums and Issues.

juan_g’s picture

Other possible fix:

cweagans’s picture

Unfortunately, regarding #195 and #196, it's not quite that simple. IIRC, those options are Solr facets. Will take a bit of work to make that happen.

In any case juan_g, you're off topic for this thread. That change should be proposed in a new issue. This issue is about endorsing SE as an official support channel.

IMO, the discussion about providing a better support experience belongs in another issue as well.

Damien Tournoud’s picture

See #955928: Split apart "Forums & Issues" in search filters. This facet was implemented that way because it was explicitly designed this way by Mark Boulton.

juan_g’s picture

cweagans wrote:

Unfortunately, regarding #195 and #196, it's not quite that simple. IIRC, those options are Solr facets. Will take a bit of work to make that happen.

Those in the screenshots are two usability bugs and fixes. The one of mixed Solr facets in drupal.org search is being taken care of in #955928: Split apart "Forums & Issues" in search filters, issue opened by webchick. I knew about this just after writing here, and the issue has also been mentioned here by Damien in the previous comment.

The other usability bug is much easier, not about search but about navigation links. Those affected seem hardcoded in the design rather than primary links, etc., but anyway this fix in particular is extremely easy. See the details in its issue, that I've just opened: #1242094: Restore "Support" in header navigation

That change should be proposed in a new issue. This issue is about endorsing SE as an official support channel.

Sorry, you are right. The two issues are those just mentioned by Damien and me.

About other issues for the various tasks, that's the same I said in #164, including a list of different questions discussed in this issue, and some links and suggestions about the places where they can be more effectively discussed. One of them was the issue #1239326: Update the support forums structure according to the current needs, that I also opened and first linked in #103.

IMO, the discussion about providing a better support experience belongs in another issue as well.

Yes, that should also be done. The support topic in this SE issue has included a great part of the discussion, because they are very interlinked questions.

For example, as we know, the SE endorsement can be done in two really different ways:

One would be to include, among the support resources on the /support page (documentation, support forums, etc.), a link to SE as a third-party technical Q&A site, which can be very useful like other resources to help support the growing number of new Drupal users.

The other way is completely different, and was the original proposal in this issue, a CNAME from support.drupal.org, which implies the removal of all user support at drupal.org.

So, you see, drupal.org support and endorsing SE can be two really interdependent topics.

From the discussion on this issue, new issues are being opened. I've already opened two related, and probably others will open some for the different tasks suggested, like webchick's several ideas in #175, or the proposal of using Drupal's Advanced Forum (more similar to modern standard forums), or free tagging (similar to the one at SE) for the support forums, and other previous suggestions in this discussion.

For the medium to long term work for a new support.drupal.org, probably including Q&A functionality, there is the Support Infrastructure group, as already mentioned.

For the short term work, that is improvements to the current support resources, basically the forums, a meta-issue to keep track of the different tasks could be opened, to replace the current support discussion in this SE issue. I would do it, but my time is going to be limited since this Monday, and of course surely others will open the necessary issues.

cweagans’s picture

Title: Endorse drupal.stackexchange.com » Decide on a course of action for improving support options on Drupal.org
Issue tags: -Needs issue summary update

Per the issue summary and discussion in the last 199 comments, changing the title.

This can include endorsing drupal.stackexchange.com as an official support venue, creating a plan for implementing a home-grown support.drupal.org, defining a timeline for completion, among other things.

Also, removing the issue summary initiative tag, due to the presence of an issue summary.

chx’s picture

We can not pull off any major improvement properly. We did the Great Git Migration yes but the compromises and agreements on what we punt on was completely ignored. We had agreed on having a manifest file in place of $Id$ and also I commented "it's absolutely crucial for security and core development both that we are able to quickly grep all of contrib for a given Drupal version" at http://drupal.org/node/1057386#comment-4131242 it was downgraded from critical to normal and noone fixed it. This is what happens: we have a major drive and we do something and then ... end.

webchick’s picture

Drupal.org is now on Git, and many thousands of commits and projects have been released since. I'd say we did a major improvement just fine, and some peoples' definition of 'critical' needs a bit of work.

ccardea’s picture

I don't want to beat a dead horse here, but I've been looking at a lot of cms sites recently, and I've noticed a few things.
1. Everybody is still using forums for user support.
2. Everybody has clear separation between user support and developer issue tracking.
3. Only Joomla provides any kind of user support for contributed projects, and they do it with a link from the project page to ...you guessed it, a forum.
4. Nobody provides bug tracking for contributed projects.

Most other sites are much more user-centric, rather than developer-centric, as Drupal is. I think that probably the reason why Drupal is so popular is because its popular with developers, because its a good place for them to hang out and talk to each other, and do what they love to do, which is eat, drink, sleep, and breathe code.

I think that what's happening here is a typical programmers response to every problem, which is, we have to build better software. But everybody else is still using forums, so, is the software really the problem? If not, then what is the real problem?

I think the leadership at Drupal needs to decide who they want to be. Do they want to try to be like everybody else and become more user centric? Or do they want to embrace who they already are and continue to be more developer centric? Not that you can afford to ignore your end users at all, but you have to figure out who you really are.

Terrific issue summary, by the way. It really adds some clarity to the discussion.

VM’s picture

@ ccardea

In defense of those who code, a large part of the problem with the drupal forums is the lack of tools. The same tools that are commonly provided by forum specific scripts aren't available in the core forum module. Drupal's core forum module is essentially a proof of concept and a rudimentary forum idea brought to life with code that has in large part become antiquated.

While it is true that forums are the engine on which support for the projects mentioned are driven. It is the features that when compared aid in the negative reactions toward the drupal.org forums. Joomla uses phpBB which comes with a plethora of features that aid in running/using/communicating in forums. phpBB's main focus has always been forum style communication. Thus they are heads and tails above what the current core forum module is capable of.

EvanDonovan’s picture

I agree with @webchick on this one. Stack Exchange is very well-designed, but I don't think we can bless it without further encouraging people to view Drupal as simply a tool that they use, rather than a community in which they participate.

I used to love supporting people, but now that I've learned how to develop for Drupal (through the support I received, primarily in issue queues, and later, IRC), contributing modules & reviewing patches is a much more effective use of my time, though I still try to do docs work occasionally. I think, as webchick stated, that is why many people no longer contribute to forums, not because they don't like to help others.

As has been stated many times, the forums also offer a poor user experience, technologically. From my observation, I think a lot of this is actually due to the threaded nature of the discussions, which makes them difficult to follow. Some way to rate responses and to flag a response as accepted would go a long way toward fixing this, I think. Some of the issue queue features (like retitling posts & issue summaries) could also be useful, since oftentimes the original poster of a forum question doesn't word the question precisely enough. It would also be great if issue queue nodes could be moved over to the forum. A good per-forum search feature would probably be the most critical though.

I don't believe making Drupal.org forums more usable would be as difficult to do as creating a 1:1 clone of Stack Exchange. On the other hand, if we simply link to Stack Exchange, then we are guaranteeing that the forums will stagnate.

I think we have to put effort toward improving the forums, even if it would be easier to just tell people to go offsite. To mention WordPress again, they keep their forums prominently in the top bar of WordPress.org. The forum categories seem clean and well-described also.

Stack Exchange will continue to exist no matter what, and people will get there from Google, since they have great SEO. But we shouldn't cede support over to them completely.

ccardea’s picture

@vm

I have to agree that, compared to what's available on other sites, Drupal's forums are antiquated. But the problem isn't limited to just forums or user support. Just about everybody else is using third-party software on their sites. I think that a third proposed solution would be to integrate more third party tools on Drupal.org.

VM’s picture

I disagree with the assertion that 3rd party software should be deployed on drupal.org. That would cause issues for infrastructure in that they would not only have to keep drupal updated but third party software as well. It's important to recognize security issues with such an idea.

cweagans’s picture

Deploying 3rd party software is absolutely not an option here. That would be a nightmare to maintain and keep secure, not to mention that the entire infrastructure is built to run /drupal/. Oh, and that part about *we're DRUPAL.org*.

ccardea’s picture

Deploying 3rd party software is absolutely not an option here.

I think maybe we're getting to the root of the problem. Drupal.org already has a problem with important contributed modules not being maintained and upgraded. I think using software that is maintained and used independently of Drupal might actually make the site easier to maintain and upgrade. IMHO what's needed is a change of culture, organization, and infrastructure.

Michelle’s picture

Adding a 3rd party forum here would make things easier in the short run. In the long run, it would spell the death of a forum module in core. The only thing keeping core forum alive at this point is that it's needed here on d.o so the core devs grudgingly give it some love at some point in the dev cycle.

Michelle

ccardea’s picture

Now that's a novel idea, taking something out of core. I wonder if that's ever been done before.

Michelle’s picture

@ccardea: Yes, several times, but that's getting OT. My point was simply that, if the result of this issue is cause drupal.org to not use the core forum, there is a consequence that needs to be considered.

Michelle

catch’s picture

Drupal.org upgrades being held up by contrib modules is a feature, not a bug.

cweagans’s picture

I think using software that is maintained and used independently of Drupal might actually make the site easier to maintain and upgrade.

With all due respect, have you ever maintained a large site with this setup? I have, and it was a frigging nightmare. We can change culture, organization, and infrastructure all we want, but that still will not mitigate the problems that can occur when two different systems with entirely different maintenance requirements are progressing at different rates.

It's a PITA, and I'm sure killes will agree.

Also, I think we're missing the point: in my opinion, a forum is simply a place to talk. Something like StackExchange is more suited towards structuring Q&A data and allow other users to quickly find the answers that they're looking for.

@ccardea, if you want to remove Forum module from core, open an issue. For your information, there were a number of things removed in the D6->D7 cycle, such as Blog API (which I am now the maintainer of).

@catch, I'm not sure I understand what you mean. Why is that a feature?

Michelle’s picture

Actually, no, don't open an issue. If you really want to discuss it, here: #881916: Should forums be in core or not?

Michelle

catch’s picture

@cweagans, just in the same way that any other large Drupal site using contrib modules is a feature, there's an incentive to help port them etc.

A lot of people go from some cms + phpbb to Drupal + Drupal forums and there are good reasons we don't hear about people doing it the other way. Managing multiple cms apps with completely different cycles but on the same site and userbase is a pain.

Michelle’s picture

@catch: Ah, I understand what you're saying now. The reason your comment came off so cryptic was that the immediate prior comments were discussing using Drupal core forum vs using something like phpbb so by contrib it almost sounded like you were talking about a bridge module.

Michelle

ccardea’s picture

First of all, this discussion started as a proposal to endorse outsourcing Drupal.org user support, because frankly what Drupal.org has now isn't all that good. Then the discussion turned to well, maybe we should build our own Q & A system. None of Drupal's competition uses a Q & A system, so really what you're doing is trying to compete with SE for user support, and in doing that you're losing focus on your core capability.

Everybody seems to agree that Drupal's forums are inadequate, and d.o. is really behind the competition in that area. So if using third party software get's you up to par with your competition, then that's something that needs to be considered. Arguing that it's a pain to maintain I think is weak. If all of the competition can do it, then d.o. should be able to do it too.

I do not see a trend of people trying to build and maintain their own proprietary forums. In fact the trend I see is exactly the opposite. Large sites are focusing on the things that they do well, and using third party software for the things that others do better. You can see this on Wordpress and Joomla, using third party software for user support, issue tracking, and api documentation. You can see it on phpBB, using Symfony, MediaWiki and others on their development site. You can see it in Symfony, using Doctrine, Monolog, and Swiftmailer. Drupal forums are fine for small sites that want to add a forum without the hassle of trying to maintain more software, but they are proving to be not that great on d.o.

Thinking about how to improve user support on d.o., there seems to have been an implicit judgement that forums are a thing of the past for user support. Looking around on the internet shows that's not necessarily true. Maybe the Q & A format is the next greatest thing, and by building their own, Drupal would be ahead of the competition. But Q & A systems are only effective if you have a fairly large group of people who contribute on a regular basis. So probably not that many Drupal sites would use it. Then d.o.would wind up with another big module that's only used on d.o., and it would eventually go the way of all such modules.

The most successful sites today incorporate third party software. I believe that will continue into the future, and I think Drupal should take note of it.

Michelle’s picture

A forum and a Q&A aren't that fundamentally different. It wouldn't be hard for one module to do both. Unfortunately, the issue is time. drupal.org needs a support solution now and not in a year when my module is ready. :(

So, first we need to decide: where will support be?

A) Totally outsourced.
B) Local support but acknowledge SE (and others?) by linking to them
C) Local support and pretend others don't exist.

Based on that:

* If A) Drop or archive the current forums and do a redirect to SE.
* If B) Add a link to SE and work on improving our support offerings.
* If C) Work on improving our support offerings and get therapy for being childish and pretending this site is the only important one in the Drupal community. :)

Furthering on B/C, we need to decide:

1) Continue on with the forums just as they are.
2) Continue using core forum but revamp the forums and categories.
3) Enhance the core forum with existing contrib .
4) Enhance the core forum with existing contrib and revamp the forums and categories.
5) Ditch forums completely and build our own Q&A solution.
6) Do 1, 2, 3, or 4 and revisit this in a year when there is a better native Drupal forum solution.
7) Install a 3rd party forum.
8) Install a 3rd party Q&A system.

Thoughts? Have I missed any decision points?

My preference is for B followed by 4 and 6.

Michelle

[Edited for clarity]

WorldFallz’s picture

excellent summary michelle-- imo that's pretty much sums it up. And my vote is the same-- 1b with 4 & 6 and i'm willing to put even more skin in the game by helping with 4/6.

Michelle’s picture

I don't know how much help I'll be with 4 right now but I'm onboard for 6. If I can pull off everything I have planned for Artesian, I think it will be a great module for support here on drupal.org. Unfortunately, I can't go back in time a year and make it ready now but If we can just hold on for that, doing what we can with what we have now, I think it will be the best ultimate solution. I'll be starting work on it again in September and hope to have a base and a roadmap so other can help within a month or two after that.

Michelle

kiamlaluno’s picture

I am for 6.

Please don't remove the forums, because a forum and an SE site are different things. The first error made from who is not used to an SE site is to consider it a forum; they then start adding a question as it would be an answer (because not all the users can write a comment in everyplace), which is then deleted.
There are cases where a forum allows a workflow that is not possible in an SE site, and in some cases that workflow is preferable.

I think it would also be preferable to let the users decide where they want to ask for support; there is who prefers Drupal Answers, and there is who prefers Drupal.org. I don't think that forcing somebody to use Drupal Answers, when they feel more comfortable on Drupal.org, is something we should do.

Fidelix’s picture

My vote is for B and 5.

Let's evolve, and not worry with what other CMS's or communities are doing. We need to do what we think is best for US, the community.

@kiamlaluno, it doesn't mean that because SE forbids new users from answering questions, we have to do the same. Remember, we could build it JUST the way we think would be better for the Drupal Community.

@all, I understand the hesitation for ditching the forums, but I think we can find a way to merge the forum's actual content with the new Q&A system, and that would make most people happy.

As Drupal SE is proving, the Q&A format is the community's preferred method for getting support, and I consider it the most efficient, in a number of ways.

WorldFallz’s picture

...the Q&A format is the community's preferred method for getting support...

Sorry, but based on what? It's a long way between a personal opinion (and again, from someone that doesn't use the item in question) and 'the community's preferred method'.

At the very least, I would say this thread is actually evidence to the contrary.

I'm not saying it's not true-- but I have yet to see anyone back that up with any type of facts. I'm honest enough to admit I personally don't know what 'the community' prefers at this point-- especially since we haven't even polled it.

Lets please stop speaking as if personal opinion is the will of the community while await Dries' pov.

kiamlaluno’s picture

@Fidelix You misunderstood what I wrote. SE doesn't forbid new users to answer questions; users on SE are not allowed to post as answer a totally different question.

In a forum you could comment about the answer given from somebody else, and that user could (replying to you) report a new snippet; if somebody else would want to know if the snippet would work in a different case, he would ask in the same forum post.
In an SE site, if you ask if the snippet shown in an answer works in a slightly different case, and you post it as an answer, the answer would be removed because it's a completely different question.

The workflow used by an SE site is not the same workflow used in a forum. In an SE site, somebody asks a question, and somebody answers; that's all. There isn't space for a continuous feedback between who asks and who answers. If you didn't write a question that is clear, then the question will be closed as "not constructive" or "not a question." If you would make clear what you mean in a comment, then you would be said to change the question you wrote, and who keeps changing the question is not much welcome: You first need to think of what your question is, what to write, and which details to report.

The point is that an SE site is different from a forum. We don't necessarily need only an SE site; we can have both our forums (which can be improved where necessary) and Drupal Answers. Having just Drupal Answers would be an error, IMO.

Fidelix’s picture

At the very least, I would say this thread is actually evidence to the contrary.

Evidence to the contrary? Can't you interpret that people spitting on the forums, and trying to fully endorse DSE is a clear evidence of that?
Look at how many users are using Drupal SE. Those people could be on the forums, don't you think? But why are they there?

It's a long way between a personal opinion (and again, from someone that doesn't use the item in question) and 'the community's preferred method'.

Now, where's YOUR evidence for saying that? I've used DSE quite a bit, and I've been looking for answers on Drupal forums since I started in the Drupal world.

You are clearly biased, and you are trying to force your opinion, not me. I understand you are a long time forum user, but you can't mess with the facts.

@kiamlaluno, your words make sense, and I agree with everything you said.

We don't necessarily need only an SE site; we can have both our forums (which can be improved where necessary) and Drupal Answers.

Alright, but we can't EVER let the forums and the Q&A site overlap each other. THAT, would be a mistake.
Maybe if we restrict the type of support requests that a user can ask in each of those?
Reduce the forum to a specific number of categories? What do you think?

Mark Trapp’s picture

Issue summary: View changes

Removing references to specifically improving forums or the forum module. While a potential option, it's not the focus of discussion. We should keep our options open, including building a Q&A system. The goal would be an improved support experience, not work on a specific module.

Mark Trapp’s picture

Just as an update, Drupal.SE is leaving beta likely this week, so if people were concerned about it being in beta, there's that.

webchick’s picture

The people who actually help by providing user support here on Drupal.org's forums are vastly more credible on what Drupal.org's support resources need or don't need, and we should defer to them on the best route forward. Same as we defer to Drupal core contributors on Drupal core decisions and the documentation team on documentation decisions.

And FIWW, the vast majority of people in this thread help on neither the forums nor on SE. If you're in that boat, you should probably bow out and let the folks who actually do the work of helping our users with support questions come up with a suitable plan.

rfay’s picture

@webchick, welcome back!

Let's include the people doing great work on stackexchange in your statement when you say

let the folks who actually do the work of helping our users with support questions come up with a suitable plan

webchick’s picture

Right, that's what I meant, in case that wasn't clear.

cweagans’s picture

Personally, I won't use the forums because it's never a simple Question and Answer sort of thing. It's always something like this (in my experience):

OP: "I'm having this problem with a views argument. I can't get it to work quite right. Here's some technical data about the problem" /* snip */

Followed by 3-300 comments that read something to the effect of "+1, yeah me too - I'm having that problem as well", loosely interspersed with other random conversations ("I'm having that same problem, except I'm not using views. I'm building my own list page. I should be looking at $_GET['q'], right!?")

Then, somewhere down the thread, somebody comes up with a solution.

For somebody that's never participated in the thread, finding which answer is "correct" is a pain in the neck because of all of the noise in the thread.

Granted, it's nice to know if there are other people having the same problem, but I don't think a forum is the right way to communicate that. The data should be a little more structured so that it's /useful/.

DjebbZ’s picture

I provide support in the issue queue and in IRC, so I bow out.

Michelle’s picture

Once drupal.org is upgraded to Drupal 7, you can add fields to comments. Add a field to the comments like "Answer is:" with "Not helpful, helpful, off topic, solution, etc" and then color code them. That would make the forum much more usable with minimal custom coding (a form alter so the field doesn't show up on comments elsewhere, maybe? Some theme code? Anything else?)

Do we have any sort of ETA on going to D7? If it's anywhere in the next few months, might be best to table this discussion until then, aside from deciding if we want to link to SE, which I, personally, see no harm in doing in addition to our own support efforts.

Edit: Ok, maybe a bit more coding to handle who can mark the comments. Maybe Flag would be better than a field since using a field requires editing the comment?

Michelle

rfay’s picture

@DjebbZ, no, really @webchick is saying I hope "Let's make sure we focus on the people who actually provide support".

I do think that's the issue. How can we help the people who actually provide support do it better and more efficiently and to more profit? And webchick, that includes many of us who don't go near the forums.

This issue is not about forums. It's about support.

catch’s picture

Right, I don't agree with limiting the discussion to people who currently post on the forums.

If we're discussing limitations in the general support experience on Drupal.org (for both support givers and receivers), then people who used to give support in certain areas of the site like the forums (or #drupal-support) but no longer do is a very valid viewpoint - especially if they still answer support questions elsewhere as many of us do. Same as people who never used the forums but do support in other places.

This applies to core too. I might give more weight on opinions about core code from people actively involved in core development currently. But if we were discussing problems with the process, then I'd definitely want to hear from people who used to be involved but aren't any more, or contrib authors who actively avoid it.

webchick’s picture

That's fine. I just don't want to keep seeing the opinions of people who actually provide support continue to be attacked/dismissed as #226 and some others have done.

I think Flag would make more sense on the "this comment answered the question" side than fields would. Fields are only editable by the person who posted the comment, whereas what you really want to know is if other people found the answer helpful. #1133956: Improve efficiency when displaying lists of flaggable comments is a blocker to deploying Flag on Drupal.org.

Michelle’s picture

@Webchick. Yeah, I realized that after I submitted. I guess I should have just done a follow up instead of an edit. I forget that many people read these things via email.

Michelle

Fidelix’s picture

Oh, that easy to throw rocks?
What makes you say I don't provide support? Because I'm not an "active user" on the forums?

If you try to read #223, you'll see that I, "a person who actually provides support", gave my opinion, and that it was easily "attacked/dismissed" at #224, and now you reversed the picture. Nice.

I'm trying to look at the problem from both perspectives, as a support giver, and a support seeker, and did not dismiss anyone's opinion.
Personally, I think all of the decision makers (apparently too few) need to look from the side of the support seeker as well, and this is not happening too much on the moment.

Michelle’s picture

Please tell me you didn't just accuse webchick of "throwing rocks". I can't think of anyone in this community less likely to do so... LOL!

And, of course, we need to take the needs of those seeking support into account. But when it comes down to getting something going, it's the ones that will be doing the getting going that end up having the most voice.

Michelle

catch’s picture

Here's an example of issue queue support #1249172: PDOException: SQLSTATE[HY000]: General error: 1364 Field 'router_path' doesn't have a default value: INSERT INTO {menu_links} (m.

That could've been a core bug, it turns out the person used some horrible script that destroyed their database that was linked from somewhere instead. This is the kind of thing I'd like to be able to move in and out of the core queue.

Fidelix’s picture

@Michelle, I did. That's exactly what it is on the first paragraph on #236, and the last on #228.

But when it comes down to getting something going, it's the ones that will be doing the getting going that end up having the most voice.

I agree completely, but I don't think opinions with less "voice" should be completely "attacked/dismissed". After all, there are people who avoid doing the doing on d.o, but do it somewhere else, as this is my case and some more support givers I know.

The vast majority of people in this thread help on neither the forums nor on SE. If you're in that boat, you should probably bow out and let the folks who actually do the work of helping our users with support questions come up with a suitable plan.

I'd really like to do what I do on support.d.o, and that's why I'm giving my opinion and trying to help with the planning. But since I was just dismissed (kind of, because I actually give some support on SE) by the webchick, I have to bow out.

@webchick, addendum:
Current forum users will likely be biased/inclined when considering dumping forums, and that's perfectly natural. So please let the rest of the community have a little more voice IF possible. Thanks.

Michelle’s picture

I think you're totally misreading that. Of course if you are planning on jumping on board to help, we want your input, too. But we have people who don't even do forum support and have no interest in starting jumping on the "dump the forums" bandwagon, which gets annoying. All that's being said here is if this doesn't directly affect you, butt out. Don't try and determine what's going to happen on a part of d.o that you don't even and won't even use, especially when it goes against the wishes of those who do use it.

Michelle

VM’s picture

@ Fidelix current forum users are only as naturally biased/inclined as those who dislike them.

juan_g’s picture

In fact there is a good number of support givers participating in this issue. Some very active supporters are for example VM, WorldFallz, Michelle... (Drupal.org Forums), kiamlaluno, Damien, Mark, rfay, tim... (Stack Exchange), others on IRC, and others that also help a little bit, like myself from time to time on the forums.

Naturally, among the views and ideas that all of us can share, the most experienced supporters should have more weight in the design of possible solutions, in addition to other remarkable members of the Drupal community (Dries, webchick, killes...).

Probably, regarding experience, etc., we should also look at what other notable open-source projects are doing for support. Something that seems unique to this issue is the idea that there must be just one, and only ONE, support place on the entire Internet for all Drupal users (drupal.org, or SE). This peculiarity does not seem to happen for other open-source communities with large numbers of users.

For example, WordPress. They have the official WordPress Support Forums, and there is also wordpress.stackexchange.com (Q&A). They are not shutting down the forums or the SE site.

Another well-known example, Ubuntu Linux. Again, in addition to the official Ubuntu Forums, and their official technical answers system, there is also... you guessed correctly, Ask Ubuntu at Stack Exchange, previously known as ubuntu.stackexchange.com.

So, before deciding what kind of support -if any- should drupal.org provide, for example the current forums with improvements like answer voting and topic tags, or a future technical Q&A system like the one discussed at the Support Infrastructure group, or both forums and Q&A like complementary support services..., before all this probably we should clarify this basic point of either one and only Drupal support resource all over the internet or, on the contrary, as many resources as possible to help support the growing numbers of new Drupal users. Here, there, and everywhere.

Fidelix’s picture

@VM, I agree.
That's why I'm asking that everyone's opinion, not just mine be heard, instead of "bow out and let the folks who actually do the work of helping our users with support questions come up with a suitable plan" as @webchick so boldly said.

@juan_g, as I said before, I agree we should take good examples out there in consideration. But we shouldn't take them as guidelines. The Drupal community is arguably the best in the world, and you can call it the vanguard of the web development, so it makes more sense if others copy our actions.

...the idea that there must be just one, and only ONE, support place on the entire Internet for all Drupal users (drupal.org, or SE). This peculiarity does not seem to happen for other open-source communities with large numbers of users.

IMO we should consider the pros and cons of this. This has been the case for Drupal in the past (always on d.o, but in multiple ways), and IMHO this ended up strengthening the community. What do you think?

juan_g’s picture

What do you think?

IMO? First, focus on the users' needs, provide them with all the internal/external resources they want to use -support forums, technical Q&A, documentation, etc.-, and let each user choose for each different task. That's it.

DjebbZ’s picture

Refocusing the discussion.

If I follow webchick's point of view : "let's make d.o support a better place by making drupal a better software", we could also look at one Drupal distro : Open Atrium. If it's perfectly valid for a company intranet, with a case tracker + wiki + other stuff, the case tracker can also be used (and is often used) as a support place. Because it looks like an issue queue. It's comments on steroids, like Project module for d.o issue queues : tags, category, flaggable, far better than a simple post in a topic-structured forum. It's one of the idea amongst many here.

And there's this saying : better safe than sorry, or an ounce of prevention is worth a pound of cure (thx Google). What I mean : the best way to provide support is surely to have a great documentation. If Drupal is well commented, it lacks a structured that is hard to add by simple commits. I'm using other tools than Drupal out there, and really the docs are so clear, you don't need to google for help. Drupal help at contrary is spread between d.o, blog posts, video of presentation during cons, screencast, paid training, books... I know it's the case with all other tools, but at least the central doc is limpid and easy to grok. For Drupal we need some of our top contributors to write and sell a huge book ("Definitive guide to Drupal 7") to learn how the software (core and contrib), the community, the sys admin stuff, the business side etc. work. No way to (at least easily) find all this info on d.o, and that's why all support places related to Drupal, whether on d.o or not are full of people. Adjusting the documentation variable should re-balance the support equation.

VM’s picture

@Fidelix SE already exists as do other support venues. No one is trying to shut SE or any other external support venue down. Just because I support via the d.o. forums does not mean I wish any ill will on SE or any other venue or on any of those users who support there.

SE exists, it lives and will continue to live. I'm not sure there has to be a competition between SE and support on drupal.org. Support givers can aid users/support users where they feel comfortable. Support seekers can ask for aid where they feel comfortable.

As far as documentation goes, it would help. However, it's another area that requires a massive undertaking, and continues to grow exponentially (though our current docs contributors and maintainers do as much work as they possibly can). Based on my experience working in the forums, documentation only helps those who actually seek it out, read it, and can understand it. With a great many support questions that are posted it is often quickly recognized that the user didn't search the forums or the documentation.

The unfortunate aspect about support be it forums or docs is that those users who did all the right things, searched first. found them, read them, worked with them, and solved their issue is more often phantom success. I'd like to believe that more people have been helped than we realize because they did take some initiative.

giorgio79’s picture

Why would it be a hundred thousand dollar project to build a clone when you can already do it with a few clicks:
http://drupal.org/node/1210616

Also, I can guess Dries' decision based on one of his old blog posts:
http://buytaert.net/drupal-and-eliminating-middlemen
"Similarly, I think there is great value in perfecting technologies that set out to eliminate the webmaster, the developer, and the designer. This is what Drupal is all about. "
So following this if Drupal cannot eliminate the stackexchange drupal answers site, Drupal could end up as a failure...

coderintherye’s picture

Take a look http://developers.facebook.com/blog/post/545/

Facebook is making their developer forums "read-only" in order to make Stack Overflow their primary help point for developers.

I've been involved in helping people on the various IRC channels in the past, though rarely on the forums. I was also involved in getting the Drupal StackExchange site going, though like others, no where near the support some offer. I also made my own StackOverflow clone using the ArrayShift installation profile about two years back now, something I talk about here: http://www.coderintherye.com/staticbuzz-community-building As noted there, it was relatively easy to get going, but that doesn't make it successful, nor better than the alternative of merely using StackExchange.

I'm very much in favor of pushing support to StackExchange, or at the very least, linking to it prominently. I love Drupal, suggest it to everyone, and spend my life working on it, but I also realize that there is plenty for the community to work on, so why take away resources to reproduce effort just so something can be done in Drupal on drupal.org when it has already been done elsewhere. I also agree with other posters that it doesn't mean StackExchange becomes the only place to get support, it just makes it more prominent, and takes or pushes away the prominent blemish that is the d.o. forums.

catch’s picture

Here's a good example of the kind of critical bug reports support requests we get for core.

#1247560: allow validation of feed data format prior to import in aggregator_save_item().

Fidelix’s picture

@nowarninglabel, I strongly believe you are missing the point.

What we need is a way to streamline support requests, and remove as much effort as possible from support givers (which include module developers, theme builders, etc.).

This will not happen with StackExchange.

JamesK’s picture

+1 for keeping support on d.o
At the very least, it would be great to have a "Support" link going to anywhere other than the issue queue.

juan_g’s picture

JamesK wrote:

it would be great to have a "Support" link

That's #1242094: Restore "Support" in header navigation

JamesK’s picture

Thanks I must have missed that (252 posts; I mostly skimmed).

Mark Trapp’s picture

As an update, Drupal Answers is out of beta and has fully launched.

Mark Trapp’s picture

Issue summary: View changes

Update beta status

cweagans’s picture

Hmm. Well, I have this askdrupal.org domain name - can we start using it on that SE site? What we do on d.o for support doesn't really have any bearing on that.

webchick’s picture

That'd be a question for http://drupal.com/trademark/apply and/or Dries, imo. Because it's pointing at a commercial web site whose direct profits are not exclusively used to foster the Drupal project, it doesn't qualify for an automatic license to the trademark. See http://drupal.com/trademark for more details.

klonos’s picture

-1 here too on going with stackexchange.com (or any other non-d.o way for what it matters).

...I am still bitter about how Experts Exchange evolved and how it "locked" users out by going the $$$ way once they "lured" enough people. Yes, there are others that do some certain things better and we don't plus we are too slow on improving things and all, but lets just do it. Lets fix things here instead of outsourcing them. I am confident that we'll benefit in multiple ways along the process.

cweagans’s picture

Experts Exchange != Stack Exchange in any way. Plus, all the content on stack exchange is under a creative commons license. Since SE doesn't own that content, they cannot relicense it and do what Experts Exchange did.

@klonos, if you want to build a better support system, then do! I think that if it were a possibility, it would have been done by now. It obviously hasn't, hence this issue.

klonos’s picture

@cweagans: if we all though that way, then I'm sure that most of the project/project_issue functionality wouldn't be here and that we would have moved over at github for code, rapidshare/whatever for releases, flickr for uploading all screenshots, Wikipedia for our documentation and so on :p

...if you want to build a better support system, then do!

If you want to drive your own CMS's community to a 3rd party outsourcer, then do!

The issue's title says: "Decide on a course of action..." so I think I am entitled to at least express my opinion without being told off for that. You need to cool down, I'll buy you beer ;)

giorgio79’s picture

This improvement project could be taken step by step.

First we could install Vote Up Down for comments
http://drupal.org/project/vote_up_down

That would be a major step forward. After that we could install User Points and User Badges, so for example whenever a comment is upvoted or downvoted a user gets or loses a point.
http://drupal.org/project/userpoints

Then for various actions, such as editing, or answering own questions users could get badges
http://drupal.org/project/user_badges

With these 3 modules drupal.org would be halfway there :)

Michelle’s picture

I think that if it were a possibility, it would have been done by now.

That's not much of an argument. The fact that something has not been done, yet, does not make it impossible. It's perfectly possible to build a kick-ass support system right here on drupal.org. We just lack people with available time. That's not exactly unheard of in our community. Changes on drupal.org often take years until it hits a critical moment where things come together and it gets done. "Death to subscribe comments" anyone? :P

Michelle

cweagans’s picture

@klonos, we're talking about one external service - a free service - that ensures the content created therein will remain free (in terms of licensing) forever. I wasn't trying to come off as upset. I'm saying that the reason we don't have a support tool like SE or similar on Drupal.org is because nobody has stepped up and built it. I DO want to drive support to a third party because their entire mission in life is to provide a great Q&A environment that's easy to use and useful. Our mission in life is not that - it's to provide a kick-ass CMS/framework thing (and related addons). We should focus on that.

@giorgio79, there are lots of options for what to do to improve support, but the reason that nothing has been built is because there has been no decision as to what needs to be built. We could clone stackexchange. Or we could improve the forums (in some unspecified way). Or we could force all support into IRC and make people comb through the logs for historical answers.

@Michelle, it's sort of hard to outsource issue subscriptions =P I'm not saying that the fact that something hasn't been done yet makes it an impossible task. I'm saying the fact that it hasn't been done yet is the reason for this issue. We just need to decide on something and then DO it so we can stop worrying about it. But there's not many people that have the clout to make such a decision: IMO, we need somebody like Dries or webchick to come to this issue and say "This is what we're doing. Let's get to building it. Go."

webchick’s picture

I don't know why a webchick or a Dries would need to "bless" someone who wanted to put time/effort in to creating a better support section for D.o. The steps are all outlined at http://drupal.org/node/1293124 and those steps apply to anyone with an idea for improving d.o. If someone wants to start step 1—creating an issue and generating an implementation plan, then getting community buy-in on that plan—that would be awesome!

cweagans’s picture

Okay, so here's the options that I've read in this (very long) thread:

Regarding the forums:
1) Close them
2) Leave them open

Regarding stackexchange:
1) Put a support.drupal.org CNAME on drupal.stackexchange.com
2) Point askdrupal.org to drupal.stackexchange.com
3) Don't point any domain/cname at drupal.stackexchange.com, but endorse it as an official support channel
4) Don't officially endorse any external support mechanisms on Drupal.org

Regarding drupal.org support improvements:
1) Build out all the stuff in Hedgemage's sandbox: http://drupal.org/sandbox/HedgeMage/1088372
2) Build a StackExchange clone (1:1 feature parity)

Are there any more options? If not, can we come to a decision about all of the above by November 15? This issue has already been open for almost three months. Let's make a decision and get it over with.

Michelle’s picture

@cweagans: Applying the logic of outsourcing stuff that no one is stepping up to do, we could have simply moved to an external tracking system for issues that already included subscription options. Seems like there was quite a stretch of time when it was mainly dww giving Project any love and it would have been faster and less work for him to use an existing issue tracking application / site.

To be clear, I'm not saying we should have done that. I'm just applying the logic in this issue to another "problem child". It could be applied to other areas as well. I'm sure there's a lot better places to host documentation than using the book module. Why not outsource all our docs hosting?

Yes, moving the burden of a support environment to SE is easy and it's there and all that. But it's one piece being removed from our current model of centralizing around d.o. And people like me who don't like the SE culture are forced to suck it up or no longer be part of that bit of the Drupal community that moves there. And once there's a precedent for giving up on making our own solution and using someone else's website, it's easy to point to that when the next one comes along with an idea to move something that's hard to do.

Michelle

moshe weitzman’s picture

If you have something novel to add, then please post a comment here. Otherwise, rehashing past arguments is unhelpful IMO. Let's wait for Dries to make a call. Consensus aint gonna happen.

webchick’s picture

Just a correction. #266.4 isn't "Pretend drupal.stackexchange.com doesn't exist". It's "Don't officially endorse any external support mechanisms on Drupal.org". (without running it through the DA's advertising program to bring in income in trade, at least)

cweagans’s picture

@webchick, Updated comment.

@moshe, this issue has been open for three months. Dries hasn't made a call. Technically, we really don't need him to make a call on it, since Drupal.org is a community run thing. If we reach consensus that we want to endorse some external support mechanism by slapping a CNAME over it, that's when Dries needs to get involved (in terms of trademark usage and in terms of "it's his domain, so he should approve it"). We can try to make a decision here.

sreynen’s picture

In related news, I recently requested to become maintainer of ArrayShift after finding out eaton was looking for a new maintainer. That's unlikely to become something usable on Drupal.org any time soon, but it should be at least moving in that direction again soon.

mechaflash’s picture

One fallback you can create in case of legal or otherwise dreadful situations between d.o and se relationship, make it so that the information database is shared between both sites, so there is a joint union in the holding and administration of the information. Also, have a way to migrate the data to and from d.o and se in case the relationship goes south.

Moving community support to the stack exchange system is the best thing you can do to grow the community and get people more involved. The Q&A system that SE has is great at rewarding those people with questions, and those with the experience to give answers. It will also speed up the community support area as many times, questions get addressed within the first few minutes after it is posted.

mototribe’s picture

I would love to see Stackexchange getting used as the official drupal forum.
I was a newbie to Drupal last Spring and have learned a ton since. I use SE almost daily and the quality and speed of responses is pretty awesome. I think mechaflash hit the nail on the head:

Moving community support to the stack exchange system is the best thing you can do to grow the community and get people more involved. The Q&A system that SE has is great at rewarding those people with questions, and those with the experience to give answers. It will also speed up the community support area as many times, questions get addressed within the first few minutes after it is posted.

Hopefully a decision can be made soon.

Michelle’s picture

If it's working for you, great, use it. Doesn't mean we have to close down support here. Not everyone wants to use some other community to get support for Drupal.

Michelle

silverwing’s picture

Our current slogan is: "Come for the software, stay for the community." Are we changing it to "Come for the software, stay for the community, go somewhere else for support"?

If we go the SE/CNAME route, we're announcing that we don't have enough confidence in our software to give support with it. That to me sends a very negative message about what Drupal can (or can't) do.

I'd rather do nothing than outsource support.

mariomaric’s picture

Improving d.o is on the DA radar - http://groups.drupal.org/node/197558

Probably it will impact also the support mechanism we have here.

+1 for #275 :)

rfay’s picture

@silverwing

I'd rather do nothing than outsource support.

So then we can use use the message

Come for the community, stay for the software. Don't expect any support though.

Michelle’s picture

@rfay: Wow, do you realize how incredibly insensitive that is to the people who pour their time into helping in the forums? I know you're a nice guy so maybe you just didn't think that through but that really sounded awful. :(

Michelle

cweagans’s picture

It is a simple fact. ~10 people helping hundreds of people in the forums = not a very good chance of getting support. Forums are not a good support mechanism. They've been sort of working so far, but we need to deal with the rapid expansion of the Drupal community and we need to do it *now*. "Fixing" the forums is not the way to do this: we need a solid tool, focused on supporting users (like, for instance, StackExchange or ArrayShift).

There's no way of knowing what version of Drupal a certain question targets, if there's an accepted answer, or how many other times this question was asked elsewhere on the site. For that matter, there's not really any way of knowing if the OP is even asking a question! It could be a rant about how Drupal sucks.

For some reason, people are so attached to the forums that this will never happen. Forums have their place, but this is not it. I've given up on trying to move this forward, and I suspect many other people have done the same.

Michelle’s picture

There's a huge difference between acknowledging that we are overwhelmed with support requests and saying that people shouldn't expect any support. All I'm saying is that it would be nice to be considerate of the people who are working hard to do support in the forums and not just come in and piss on their efforts and declare that they aren't doing good enough therefore we're going to just shut them down.

I'd like to know how many of the people advocating shutting down the forums actually do support there on a regular basis. Do you even know what's going on? Or are you just repeating the "forums suck" litany? Because, frankly, I'm in the forums a lot and I see plenty of people getting plenty of support. Is it ideal? No, obviously. Is it worth keeping? I sure think so. That's not to say that no other avenues for support are allowed to exist and, certainly, no one is suggesting shutting SE down. But that doesn't mean we can't continue to work on our own efforts as well.

Michelle

cweagans’s picture

There's a huge difference between acknowledging that we are overwhelmed with support requests and saying that people shouldn't expect any support.

No. There's not. If the community keeps growing at the rate that it has for the past year or two (and/or if the rate of growth increases, as it has for the past year or two), then we're going to have a *really* big problem on our hands, wherein we have TONS of people on the forums asking for support, and still the same number of people giving support. We're overwhelmed now. How can we expect to keep this system working in the future after such projected growth?

All I'm saying is that it would be nice to be considerate of the people who are working hard to do support in the forums and not just come in and piss on their efforts and declare that they aren't doing good enough therefore we're going to just shut them down.

I'm certainly thankful for the people that do give support on the forums. The work that they do is amazing, and I don't think anyone is discrediting that. The problem is that, other than the people that are currently helping on the forums, nobody wants to help on the forums. Forums are old, kludgy means of communications, and frankly, as a community of web developers, we should strive for something better.

I'd like to know how many of the people advocating shutting down the forums actually do support there on a regular basis.

As I said in #169, I don't spend time in forums, because they are a waste of time. If there is a particularly complicated topic of discussion, the comments can be quite lengthy, and you have to read all of them in order to find a solution (again, there's no way to flag something as an accepted answer). And for every answer you do find, you have to try it, and if it doesn't work, continue reading till you find another.

All of that aside, the number one reason I don't spend time in the forums is because people like webchick, davereid, merlinofchaos, et al don't hang out there. They are the people that I think of when I think of the Drupal community, and they are usually on IRC. I tried getting support through the forums when I first started using Drupal and it was unhelpful at best. Questions would sit unanswered for weeks, and when they did get an answer, it was often wrong.

Because, frankly, I'm in the forums a lot and I see plenty of people getting plenty of support. Is it ideal? No, obviously. Is it worth keeping? I sure think so. That's not to say that no other avenues for support are allowed to exist and, certainly, no one is suggesting shutting SE down.

Again, I'm coming back to the community growth rate. We cannot afford to just keep the status quo. We need in find new, better ways of supporting our users. The people that have been tirelessly giving support on the forums are amazing, and I'm so glad we have people like that in our community, but we simply cannot expect that model to work when the community doubles or triples in size over the next few years.

I think nobody is suggesting the SE be shut down because 1) it's awesome, and 2) it's a valuable, efficient support resource where answers are cataloged and easily searchable.

But that doesn't mean we can't continue to work on our own efforts as well.

This thread is a prime example of why we should outsource our support. We've been arguing about how to solve the problem of how to improve support mechanisms on Drupal.org for 5 months, and to my knowledge, nobody has spent any time on building an alternative to Stack Exchange meant for deployment on Drupal.org. If we're going to do it, then let's do it. If, as chx said in the original post, we're going to take a few years to debate and argue the finer points of a support system on Drupal.org, then let's use StackExchange.

As my dad would say, "Shit, or get off the pot".

Michelle’s picture

Well, I can't speak for anyone else but I've been busting my butt working on a better forum for Drupal that, hopefully, will be able to be used on here some day. If folks decide we can't support our own users and shut the place down before I have something usable, well, so be it. Not checking the forums for people I can help will free up more time for me, I guess. All the others that help in the forums will have a lot more free time, too. Of course, all the people who currently get help in the forum would be screwed but, hey, it sounds like SE doesn't have any growth problems. Since they have infinite help resources over there, maybe we should outsource all our support. Hell, let's send the issue queues over there, too. Then we won't have to deal with any pesky users here on drupal.org.

Michelle

rfay’s picture

We've actually already had this whole conversation in this thread. And I've probably already been inflammatory and been called out for it.

But the simple fact is: It's not that we don't have great people already offering support, but that we don't have a structure that actually scales to the need we have. We have absolutely awesome people busting their [any body or other part you want here] to try to handle support. But we don't have a mechanism that deals with it efficiently or that engages enough people in it.

Drupal Answers actually provides the structure that engages people in the process and can more efficiently in the short and the long term answer immediate questions and then provide easy access to them through web search down the line. I'd love it if we copied it or exceeded it. But instead we just keep having the conversation about how awful it is to say that the wonderful people doing their best are not adequate to the huge task we face as a community.

The problem is structure, not the great people who are currently offering support. And of course I would never say anything bad about the people, as I treasure them. But you'll find that I already got in trouble in this thread, for almost the same reason. Same problem still exists. Not going away.

cweagans’s picture

If anyone is interested, we're discussing this in IRC *right now*. #drupal-contribute.

Michelle’s picture

@rfay: Yes, they have a nice technical setup. And, as I'm likely to crosspost with cweagans about, I have no problems letting people know it exists. But I don't think this is either-or. We don't have do destroy what we have here. Both can exist. Let people choose which they prefer rather than forcing them to use what some people think is best for them.

Michelle

cweagans’s picture

Status: Postponed (maintainer needs more info) » Needs review

Okay, after hashing this out in IRC for a little while, here's what we (myself, Michelle) think should happen:

1) Add a block to http://drupal.org/forum saying something to the effect of "Some support can be found in the forums, and you're welcome to use it as such, but you may find quicker response and more condensed information here: link to SE"
2) Add a link to SE on http://drupal.org/support
3) Point askdrupal.org at the Drupal StackExchange site

Changing status to needs review for community input. If somebody could assign this to Dries, that'd be awesome. I previously submitted a trademark license request regarding #3, but the law firm denied it without stating a reason and did not respond to follow-up questions. SE is a valuable community resource, and I don't see why it would be a problem.

webchick’s picture

I personally don't want to forward a crap-ton of drupal.org's traffic to a commercial entity without some kind of monetary compensation in return, which we can then use to reinvest in our own community's tools. I also don't really understand why that's necessary, given everyone's point above that people looking for help will simply use a search engine to get it, and receive both d.o and stackexchange links in return.

However, I also think this entire conversation needs to be revisited in light of the fact that the DA is planning on making significant contributions to Drupal.org this year. It'd be great to hear from the people actively providing support on Drupal.org as to features that would make their lives easier. Probably in a separate issue, though. (we might already have that somewhere in this monster thread; I can't remember.)

The issue is already assigned to Dries, though I know he's shortly on his way to Australia so he might not be able to get to it for a week or so.

VM’s picture

It's a bit misleading to state that ~10 are people providing support on d.o. forums. Perhaps that number is meant to reflect those who are also d.o. site maintainers and providing support.

It seems to me (I have no statistical proof) that there are far fewer 0 reply threads today than there were years ago. Considering the growth of drupal over the past years, that seems like a pretty big deal to me.

Based on what I've been witnessing by checking the first page of each forum (almost every day) users seem to be helping other users, which is what I thought the intent of the support forums were to begin with.

kiamlaluno’s picture

I am personally contrary to removing forums from Drupal.org, simply for the fact that Stack Exchange sites are not forums. That is the error done from who starts using Drupal Answers, and didn't use any other Stack Exchange site before; that is evident when you find "answers" such as:

Did you find a solution for your problem?

Stack Exchange is very rigid on what can be acceptable as answer, and as question: Subjective or too localized questions are not welcome. In fact, the FAQ used by any Stack Exchange site (including Drupal Answers) reports which subjective questions are not acceptable, and too localized questions are closed, which means (for who doesn't know Stack Exchange) that the question cannot be anymore answered, and in the future it could be deleted (users with the right reputation, such as 15000 or 20000, can vote to delete closed questions), which means the question and its answers would be visible only to moderators, and users with at least a reputation of 10000.
While this can be the reason of the Stack Exchange success, I think that we need a place where users can ask a question, even if this is subjective, or too localized. I think that is part of what makes Drupal.org friendly: We don't say, "this cannot be asked," "this can be asked," except (maybe) when the question is clearly off-topic for Drupal.org. I have seen questions on our forums about a problem with a Drupal site being answered saying that the problem is a server setting, and I like that in our forums who asks a question can get an answer without to be said the question is off-topic, when he is asking something about a Drupal site.

@Michelle Using a Stack Exchange site as issue queue doesn't work.

  • On Drupal Answers, when I was moderator, I added the questions about bug reports as off-topic, whatever the bug is about a module, or Drupal.org. This means that who asks a question about something that is caused a bug is asking an off-topic question there (except in the case the question is about how to write a work-around for a module bug). If somebody asks a question why it is not possible to do something on Drupal.org, when that is caused by code that need to be re.written, that person is asking an off-topic question. The part of the users using Drupal Answers, and who is also a user on Drupal.org, was clear about that.
  • In general, Stack Exchange is not useful for reporting bug reports; the only site where bug reports are done is in the meta sites, where you can report a bug about Stack Exchange. Stack Exchange sites are about questions, not reports. If you would ask a question about a bug, that question would be useful until the bug is not resolved, which could happen right two days after I ask the question; after, the question could probably have an historical meaning, if it has any meaning.

I think that we can list Drupal Answers as resources about Drupal, but I don't think we should remove our forums and rely uniquely on Drupal Answers. After all, if somebody has a Wordpress forum about Drupal, we accept that forum on Drupal Planet, as long as the requirements for being included in Drupal Planet are met.

@webchick From this post, I read that Stack Exchange was a Platinum sponsor for Drupal Camp Indy. Doesn't that count as sponsoring Drupal activities?

Michelle’s picture

@Michelle Using a Stack Exchange site as issue queue doesn't work.

I was being sarcastic. There is this idea that moving to SE will magically solve our overloaded support issue so I suggested, facetiously, to move all support there, including support requests in the issue queues.

SE is good as a question-based pseudo-wiki documentation. That doesn't cover everything we need. Which is why I keep on saying (snarky comments aside) that can't be our only support area. Your post illustrates that quite well, better than I've managed. Hopefully it's clear now.

Michelle

mototribe’s picture

@cweagan: I like your proposal:

1) Add a block to http://drupal.org/forum saying something to the effect of "Some support can be found in the forums, and you're welcome to use it as such, but you may find quicker response and more condensed information here: link to SE"
2) Add a link to SE on http://drupal.org/support
3) Point askdrupal.org at the Drupal StackExchange site

In my business, before we have to make a big change we always try to test it on a small scale. If we offer an alternative support option with (IMHO) a much better user experience we let the users decide which option they like best. If it doesn't work we pull the links and haven't lost much.

I can even envision having "Question & Answer" links on module pages, for example, this link would show all SE questions that reference the views module: http://drupal.stackexchange.com/search?tab=votes&q=url%3a%22*project%2fv...

Why not give it a try. Of course it would be awesome if the Drupal Association can pull it off and build a better support site that is totally integrated with d.o. Until then let's use the tools that are available today to help us grow the community and take really good care of the beginners and newbies.

webchick’s picture

"@webchick From this post, I read that Stack Exchange was a Platinum sponsor for Drupal Camp Indy. Doesn't that count as sponsoring Drupal activities?"

Sure, but Acquia, Palantir, Phase 2, Lullabot, etc. are all companies that both make investments in camps, AND far more direct sponsorship of the Drupal project itself through porting modules, fixing bugs and adding features, providing employee time to contribute directly to Drupal, etc. And yet we still require them to pony up funding in order to do commercial advertisement on Drupal.org, and this money gets reinvested into things like DrupalCon, drupal.org improvements, and other things that benefit the entire Drupal community.

So why would we not apply this same guideline to all commercial entities, and throw out an opportunity to reinvest in our community? It doesn't make any sense to me.

mototribe’s picture

@webchick: on the other hand, if DA were to built a SE site it would probably cost tens of thousands of dollars to get a basic version built.

If we were to pay an outside company to run a Q&A support site it would probably cost hundreds of dollars a month. For example, http://getsatisfaction.com is one of the big players in that field.
If we look at it purely from a "business standpoint" we probably gain more using a free service like SE rather than doing it inhouse or paying for an outside service.

Personally, I'd rather see us use a Drupal solution too but it would would be nearly impossible to keep up with the SE site, those guys invested millions in their sites and are continuing to improve all the time. That's why their sites are so good and are so popular.

Maybe an understanding can be reached with Stackexchange that if we were to build an integrated Drupal solution we are free to import all data? I also wonder what kind of deal Ubuntu has with Stackexchange?

Michelle’s picture

on the other hand, if DA were to built a SE site it would probably cost tens of thousands of dollars to get a basic version built.

Not really, mostly it's a matter of time. I'm trying to build something that can be used for better support but it's not going to be any time soon, probably a good year or so, before it's ready. Plus, d.o needs to be upgraded to D7. So it's definitely a long term thing but it's not like nothing will ever be done if we don't go to SE and it's not that expensive.

Michelle

Pages