This time for real. Note that I left migrate in. There's nothing left to be said or done. My vision, my understanding of what Drupal is or what it should be now hopelessly differs not just in technical matters but also in process and the community as well. Time to step down. I will

  • see migrate through including core patches either containing or blocking migrate. I see very little potential for controversy here, the migrate architecture is almost done as it is in core, the changes while important are not substantial except the load plugin system. The migrate blocking core patches are very small.
  • see #2068325: [META] Convert entity SQL queries to the Entity Query API through. Zero controversy, there are others wanting this, esp the multilingual initative but if it's not done by Szeged, I will wrap these up.
  • not participate in any core issues not listed above. I have filed an infrastructure issue asking for the removal of my follows on core issues. I hope this happens soon.
  • likely stay around just not in a core developer capacity
  • help in transferring my knowledge (if there's any left that's not hopelessly outdated)

I have removed my elevated roles on drupal.org , groups.drupal.org and qa.drupal.org and vacated #drupal-contribute . I will stay around in small tematic channels, most importantly #drupal-migrate .

Thanks for the last ten years, it's been quite the ride.

Files: 
CommentFileSizeAuthor
#35 for_real.patch2.11 KBchx
PASSED: [[SimpleTest]]: [MySQL] 63,689 pass(es).
[ View ]
#30 for_real.patch2.12 KBchx
PASSED: [[SimpleTest]]: [MySQL] 63,286 pass(es).
[ View ]
#4 log-2008.txt98.68 KBchx
#4 log-2011.txt79.21 KBchx

Comments

sun’s picture

Title:Remove me from MAINTAINERS.txt» Remove chx from MAINTAINERS.txt
Assigned:Unassigned» Dries
Priority:Normal» Major
Status:Needs review» Reviewed & tested by the community

1) An emotional, destructive issue like this would normally won't fix, but

2) MAINTAINERS.txt should match reality, and even more so, respect availability and willingness of contributors, so

3) under the assumption that this issue is meant seriously, we're not playing chess, and the change represents @chx's actual decision to step back from core development, this patch can be committed.

As this is a major change, I'm assigning this issue to @Dries.

The pragmatic code/patch/peer review process ends below the line.

Personally, I'm very sad to see this; it actually scares me, to the point I'm starting to think about my own involvement. I'd love to won't fix this and call you instead to talk about what we can do to change the situation.

I already tried to voice and channel related community concerns on your blog very recently. However, your reply sounded like those are not fully in line with your concerns, so I'm not sure whether I'm the right person to talk to. Though again, I'd love to.

webchick’s picture

I won't change the status on the issue, but I really feel it's more properly categorized as "postponed (maintainer needs more info)." We should not simply accept + RTBC requests like this; it's imperative that we dig deeper, to find out more about what's going on, and see if there are ways we can resolve the situation. Especially in the case of an individual who has given so much of himself to the project for almost a decade, and has had such massive influence in the lives of so many people (including myself).

So, I really don't want to open a can of worms, nor do I want to reopen a painful wound for you, but I'm sure your friends would really appreciate some more context here. Or at least I know I would.

The only thing I've seen/heard in explanation is a pointer to #1341924: Make the WSCCI stack a singleton and a concern that we're headed in a direction where it's ok to design APIs in core without specific use cases in mind. Is that what this all boils down to, or is there more? Because at the moment, what I observe is that the context system patch is being actively and vigorously reviewed by a dozen or so people, with what seems like lots of constructive criticism being raised, and alternate approaches being explored. And in any case, it's in absolutely no danger of being committed blindly in its current form, so I'm afraid don't quite grok this concern.

So chx, what's up man? :( And how can we help?

chx’s picture

What's up? It is very hard to put it into words because more than once my exact words were taken as an affront when I really didn't mean one. So please try take the meaning of what's below. Still won't be too easy to draw a coherent picture.

tl;dr: I lost heart, direction and motivation but I do not seem to be alone.

So, I have been using Drupal 7 longer than most. 'Cos we started using it a year before it was released. I call building a site with D7 precision surgery because you just need to add a few lines of code to do something amazing. That's great. However, when something goes wrong, they go horribly wrong. To keep a semblance of speed, we have grown an intricate web of static and database caches that are extremely hard to flush in the correct order or to even figure out what to flush. Remember the joke: there are two hard things in computer science: cache invalidation, naming things, and off-by-one errors.

I should know Drupal 7. I have written or helped writing most of it. And yet, the intricacies of field API throw me off more than once.

Form API, which I probably should take most responsibility for is simply horrific by now. It's a hodgepodge of arbitrary callbacks, checks etc which we added to force the mess called HTML into order -- but, again, it should be seriously cleaned up.

Or say, menu API. Peter and I have started back in the summer a cleanup but that never went anywhere. It's nothing new. The issue has been there literally for years. No help. Might be usual, might not be.

Yet these aren't the questions being raised. These aren't the focus of Drupal 8. We are adding a context system which uses ->get('arbitrarystring') as its main API. We are adding a config system, which, again, I helped design, and I am not quite sure the complexity we created is the right answer. Being unsure is part of the problem and will be the hardest to put into words but I am trying.

I have reviewed the git log from (the arbitrary) Sep 2008-Nov 2008 and Sep 2011-Nov 2011. It's about the same time in the D7 and the D8 lifecycle, right? It's actually one more month since release for D8. One oddity is the second log is smaller. That, in itself, is strange, you could say that catch was only named at the end of September but still it's odd. Actually reading the logs, the first impression that comes out of the first period is that the release was full steam ahead, very important features getting added (#input_format, file revamp, the testing system becomes more solid, DST, beginnings of RDF, SQLIte, UI-wise better pwd checker, content admin screen ) and we had fun! Both committers are joking, smileys are abound. Compared to this, you'd be hard pressed to find a single great feature or a joke in the 2011 fall period. Yes -- I am a little self contadictory with wishing for cleanups and features at the same time but the aforementioned lack of speed / steam / whatever-you-call-it is more visible from this.

So when you say "but chx, you could go and help fixing all the problems you raised" consider the environment. While I have my doubts over what we have created noone else seems to be overly concerned or focusing on this. Could I drive a major cleanup? Perhaps. But, consider the still present 100 major bugs and 10+ critical issues which I am dedicated to help with and I am (this two weeks were busy but still I tried to lend a hand where I could) but it's simply a drag on everyone to see that mass. Consider what I have said about what's being added to D8 and I am not sure whether I wouldn't be playing Don Quixote.

chx’s picture

StatusFileSize
new79.21 KB
new98.68 KB

Here are the git logs.

chx’s picture

If you want to talk of different priorities, we actually have a uniquely clear contrast in http://drupal.org/node/1323210#comment-5168606 "we decided to solve the [scatteredness of context] problem" and http://drupal.org/node/1233232#comment-5301452 "that was never really the point". Apparently the we in "we decided" was just me.

good_man’s picture

I'm sorry don't want to open any kind of debate or long discussion here, I just wanted to share my thoughts, as a contributer but not in the top-contribution-circle (wish is a rare comment to find in deep issues like this).

While I have my doubts over what we have created noone else seems to be overly concerned or focusing on this.

Frankly, most of us do, but due to the fact of your deep involvment with D7, and the small number of contributers like you, nobody from outside have the guts to raise his sound like you, because simply (for me at least), #1 I don't have the enough info to do this massive cleanup, #2 don't have the deep involvment (and the knowledge of the old issues and debates) that allows me to know the ins and outs of different things, features, and architectures in core. The most of what I can do is to open a single-issue at a time to cleanup something, not to lead a big-but-necessary cleanup like yours.

So in short, I think many many people think like me, I'm tired of many things in Drupal that I can't even list here, and I wish to do a cleanup but only *with* the help of deeply involved contributer like you (i.e. don't have the requirements to lead it!). I think thats why you are thinking that you are sailing alone.

I hope that other top contributers (and initiatves leaders) give at least the same priority to the cleanup as to adding features, we don't want to stop adding cool features, but at least we need the same amount of work (or more) to do cleanups after adding these features (wish sounds very rational).

Hope my comment helps showing "what normal contributer thinks" of this issue.

chx’s picture

Well, then.

Let me answer webchick's question: #1366940: Start a Code Cleanup Initative . Pending on that, this one can be closed as closed (fixed).

chx’s picture

Status:Reviewed & tested by the community» Active

And so active for now.

Dries’s picture

I refuse to commit this patch until I had a chance to talk to chx at length. My desire is to get to the bottom of this, and to help work out a plan forward. More later.

beejeebus’s picture

yay for #9.

chx’s picture

An alternative would be ditching all initiatives. The initatives mean, those are the places where things happen, where the official attention is, anything outside of that is not too important and so you are not interested even worse disagree with them, good bye. This is why I want my own initiative. This might explain partiallly the apparently almost moribund state of Drupal 8 visible from those logs.

pounard’s picture

I agree for you to have your own initiative: the one you opened about code cleanup is more that what core needs right now. But you cannot ditch others. core is too big already for everyone to work on everything. There is a point where you have to trust others.

sun’s picture

This is the wrong issue to discuss this, but:

There is a point where you have to trust others.

It's not about trust. It's about disagreement.

Traditionally, a major API change to Drupal core required a major buy-in and wide agreement from contributors. If that level of agreement couldn't be achieved, the change did not make it in.

Typical resolutions have been to figure out what parts we can agree on and try to revamp the change to only do those, or to re-implement the idea from scratch in order to achieve an agreement.

If all fails, the change still wouldn't make it in.

merlinofchaos’s picture

Previously, changes to core more or less required the buy-in of people who had the free time to be able to sit on the issue queue and read issues and respond to everything. People without that kind of time or attention span were unable to participate and often left out of major changes to the API because they were simply drowned out by the people who had the time to participate.

The initiatives are structured in such a way as to try and open participation by not requiring constant monitoring of the issue queue. This has had the effect of making the people who held a lot of power through hyper-participation in the issue queue feel left out because the discussions are happening at a different pace and in different places and even though they get to contribute just as much as anyone else, they somehow feel like their contributions are being pushed out.

This argument looks like the other side of the coin that's been argued in the Drupal community for years: people feeling left out of contribution by being unable to participate in the existing channels. Only now, people who have been highly participatory are making the same arguments that they dismissed from others in the past.

pounard’s picture

Yes may be you're right, as the core grows, it becomes nearly impossible for any individual to be aware of everything that's happening, it's a difficult situation. There is still a problem, you cannot ask everyone to reveiw each patch, nor let them enough time to be able to do so, core is too big now.

Bojhan’s picture

I am sad to see this post, but I also understand how the past few months have lead to this.

I don't think this is the right place, to have a large discussion about core development. But we do need to have this conversation somewhere, I feel that the past few months we have been doing to much treatment of individual cases and symptoms. If you look at the larger trends, there is a significant number of heavily involved core contributors who are unhappy with the direction of the core development (as a community and as a codebase). We have seen blogposts from chx, sun, catch and significant decrease of all maintainers there involvement.

The reactions of @webchick and @Dries make me worried, this will again be treated by tackling the symptoms (getting the clean up rolling) and not about getting to the heart of the issue (why many maintainers are running out of motivation, and core has not moved significantly in the past two years). Lets not focus on finding a solution, but actually understanding the problem this time, and communicate around this clearly.

Honestly most of the responses around initiatives, remind me of D7UX and my reflections on it "I think its essential to recognize that with how the feedback process was handled during D7UX, we demotivated a crucial group: our core developers, who can do the heavy lifting required for some of the proposed changes. ". The exact same thing is happening with WSCCI it seems, and its because there is a communication/decision breakdown - this needs to be recognized and tackled (you dont solve this by more comments).

Darn it, now I am totally contradicting the first line of my second paragraph :) Anyways, chx please take your time and discuss this throughly with Dries, webchick and hopefully also us - I think many including me share, your concerns with the current direction of core development.

beejeebus’s picture

can we close this now? seems to have been superseded by the cleanup initiative?

Dries’s picture

Assigned:Dries» Unassigned
Status:Active» Closed (works as designed)
pounard’s picture

"works as designed"?

sun’s picture

Not that I wouldn't be happy to see this closed as bogus, but a little bit more transparency would be nice.

adamdicarlo’s picture

@sun, I agree.

@Dries, @chx?

Michelle’s picture

Transparent or not, I'm just glad he's staying. chx++ :)

Michelle

damien_vancouver’s picture

[ Edit: removed mis-post of http://drupal.org/node/1366940#comment-5628716 for brevity ].

Dave Reid’s picture

Wow, all those words and none of them related to this actual issue at hand. Good job.

/me claps.

arianek’s picture

Damien dude - wrong issue! I'm pretty sure that was supposed to be on #1366940: Start a Code Cleanup Initative

damien_vancouver’s picture

Sorry about that, drupal.org kept crashing and returning me a site offline message or no CSS as I was trying to post it. :(
Will repost there too then.

xjm’s picture

It also has nothing to do with #1366940: Start a Code Cleanup Initative.

arianek’s picture

Sure - I'd just been talking to Damien earlier and was pretty sure he posted on the wrong thread... ;)

chx’s picture

Title:Remove chx from MAINTAINERS.txt» Remove chx from MAINTAINERS.txt (except migrate)
Priority:Major» Normal
Issue summary:View changes
Status:Closed (works as designed)» Reviewed & tested by the community

This time for real. Note that I left migrate in. There's nothing left to be said or done. My vision, my understanding of what Drupal is or what it should be now hopelessly differs not just in technical matters but also in process and the community as well. Time to step down. I will

  • see migrate through including core patches either containing or blocking migrate. I see very little potential for controversy here, the migrate architecture is almost done as it is in core, the changes while important are not substantial except the load plugin system. The migrate blocking core patches are very small.
  • see the "convert to entity query meta" through. Zero controversy, there are others wanting this, esp the multilingual initative but if it's not done by Szeged, I will wrap these up.
  • not participate in any core issues not listed above. I have filed an infrastructure issue asking for the removal of my follows on core issues. I hope this happens soon.
  • likely stay around just not in a core developer capacity

I have removed my elevated roles on drupal.org , groups.drupal.org and qa.drupal.org and vacated #drupal-contribute . I will stay around in small tematic channels, most importantly #drupal-migrate .

Thanks for the last ten years, it's been quite the ride.

chx’s picture

Issue summary:View changes
StatusFileSize
new2.12 KB
PASSED: [[SimpleTest]]: [MySQL] 63,286 pass(es).
[ View ]
chx’s picture

Issue summary:View changes
jherencia’s picture

chx thank YOU for the incredible work you've done :).

alexpott’s picture

Status:Reviewed & tested by the community» Needs work
Issue tags:+Needs reroll

for_real.patch no longer applies.

error: patch failed: core/MAINTAINERS.txt:321
error: core/MAINTAINERS.txt: patch does not apply
chx’s picture

Status:Needs work» Needs review
StatusFileSize
new3.42 KB
chx’s picture

StatusFileSize
new2.11 KB
PASSED: [[SimpleTest]]: [MySQL] 63,689 pass(es).
[ View ]

Status:Needs review» Needs work

The last submitted patch, 35: for_real.patch, failed testing.

The last submitted patch, 35: for_real.patch, failed testing.

chx’s picture

Status:Needs work» Needs review

35: for_real.patch queued for re-testing.

chx’s picture

Status:Needs review» Reviewed & tested by the community
Dries’s picture

Status:Reviewed & tested by the community» Fixed

Committed to 8.x. Sad but we can always roll it back. Thanks for all your contributions, chx.

chx’s picture

Issue tags:-Needs reroll+sad chx
xmacinfo’s picture

I feel very sad too. I was thinking that chx and Drupal would always be linked together, but we would not have the same Drupal 8 as we have now.

Now that it is impossible to upgrade a Drupal site to Drupal 8, most developer will begin to think:

— “Do I migrate to Drupal 8, with the new migrate path, or do I migrate to another platform?”

Good luck chx.

neclimdul’s picture

um... I think you may have missed the "Except migrate" part of the title.

Thanks chx, I look forward to seeing you around the issue queue even if you aren't on the maintainers list.

benjy’s picture

Issue tags:-sad chx+sad s

@neclimdul, what makes you say that, seems to be there? http://drupalcode.org/project/drupal.git/blob/refs/heads/8.x:/core/MAINT...

benjy’s picture

Issue tags:-sad s+sad chx

Fixing tags. Not sure what the happened there.

neclimdul’s picture

I should hope so, that's what the title says.

Status:Fixed» Closed (fixed)

Automatically closed - issue fixed for 2 weeks with no activity.