Welcome to today’s asynchronous Project Browser Site Builder subcommittee meeting! We meet here every Tuesday at 8:00PM UTC (4:00PM ET) I will be starting threads with items that we would like feedback on. If you have something else you would like to discuss, please add another thread. Please say hi on this thread so we know who attended.

AmyJune (volkswagenchick) Hola! AMyJune here from Ohlone lands in Northern CA,USA
thejimbirch Hi
bsnodgrass (he/him) Howdy... Chicago Area, taking a break from painting to join you fine filks here
rkoller good evening
shaal Hello, Ofer Shaal, from Boca Raton FL :sunglasses:... Very late to the meeting
gaurav.kapoor Hi.

1️⃣ Card (Teaser) view on project browser initial results page. Do you agree that the six items listed are the appropriate fields to be displayed on the card.  Issueimagetitle (linked to d.o page in new window)short descriptionsecurity statuscategory and download/install icon/button(edited)

AmyJune (volkswagenchick) I question security status since so many module need to be reviewed…
thejimbirch Will non-security status modules be listed?  I thought they were filtered out,  In which case, I'd suggest removing it as every card would have it.
bsnodgrass (he/him) If only security coverage modules would be listed, I agree with @thejimbirch on the security status no showing.Not sure that has been determined yet however (edited)
leslieg I think that would depend on if you are using the default “Browse” criteria that has security coverage as a behibd the scenes filter or are using the Advanced criteria where you could choose not to use security coverage as a filter.
bsnodgrass (he/him) The default listing would likely include only security covered projects, however If additional filters/search criteria are applied this is an important distinction. and Why would we provide two different card displays?
leslieg That’s why we were going to always include it on the card view
leslieg Do folks agree with the other proposed fields? Anything we are missing?
rkoller hm one field might  be about the minimum supported drupal version (at the moment more relevant than in a year or two after eol of drupal7 and 8). at least in the search logs i took a brief look in it was a recurring search query.
shaal Perhaps a field for 'top 100` modules?A way to distinguish very popular modules
Nico Grienauer (he/him) I would call the image -> logo because we shoudl try to seperate that and make it clear, what there shoudl be displayed. a screenshot in this size makes no sense in my opinion. and on the detail page we can then have a “slider”/gallery with screenshots.
rkoller i agree that screenshots are probably too small in size for a teaser card. but in case you call it logo you have another problem. if you take a look at the current project page (https://www.drupal.org/project/project_module) only a small fraction of modules have an actual logo. the rest either a screenshot or nothing at all. that would require module maintainers to come up with an actual logo to keep appearance of the cards consistent. wouldnt it make sense to choose only fields for the teaser card 100% of all modules are able to provide and populate?
Ron Northcutt On the security status: I think we can simply show a badge if it has security coverage, or nothing if it doesn’t. Should keep it cleaner. We need to be able to filter by security status for sure.
Ron Northcutt As for images/logos, I suggest that we just that the first image in the series regardless of what it is, and show a generic Drupal logo if it is blank. Then, project owners can choose to update their projects with a logo, image, etc. and know that the first one is the one that will be shown.
Ron Northcutt I don’t think we can expect every module to have an image/logo
Nico Grienauer (he/him) I disagree. if there will be no “logo” or representable image of the module, there should be NO image at all at the overview. also no defualt druplicon. this is not needed and I think we should not have aots of placeholder images. and a screenshot etc is just not suitable over at the overview pages.
Ron Northcutt Well, if we have a design that includes images for some things and not others, then we will have blanks and gaps on many of the cards. Do you have an example of a card grid that demonstrates what you are talkinga bout?
Nico Grienauer (he/him) at one call, the person who did the figma designs said, that they check this and will make a design, so we can see how it looks.

2️⃣ A decision needs to be made on the “Recommended” text on the Project Browser page. Since we are not recommending modules in this phase, should another term like Suggested or Supported or ??? be used instead? Issue

leslieg Any thoughts on the text used? Feedback is that Recommended has a connotation that we do not intend
bsnodgrass (he/him) I like "highlighted" and if we were to also include Stable, active, and supported that further qualifies it.
leslieg Thanks @bsnodgrass (he/him), Do others have ideas/thoughts on the term we use?
Nico Grienauer (he/him) Supported I would not choose also not Recommended.highlighted could be a thing.I am not sure which modules will be viible there. if it shows them just from a search algorythm , it would be also fine to write this. e.g top “installed modules”, etc
Ron Northcutt FWIW - I think that “recommended” and “highlighted” both convey a sense that they are promoted in some way. On the other hand, “suggested” seems a bit less promotional.Then the question is “why are these suggested?“, and the answer (which could be in a tooltip or description) is that these suggestions are based on the current site and the configured filter options. Then we can have a link to change the suggestion settings
chrisfromredfin ^ Incidentally I love the way that @Jillian Chueka et al have proposed getting at the "advanced" filter set. In this way I think the recommended / top modules would really be a set of default filters... possibly.
Jillian Chueka What about "Top Used" or "Most Frequently Used"? We could revisit recommended, highlighted, suggested, etc. at a later date when we have more information?
chrisfromredfin Another thought... "Top Projects" (or Top Modules?). The idea is that there's a handful of things that are factors, like popularity... but also other things, like "stable" and "secure" etc

3️⃣ We talked about the card sorting exercise we did during the BoF at DrupalCon Europe. We would like to repeat that exercise with site builders and folks newer to Drupal. Who would like to be part of this small focused group? Time commitment would be a couple of hours to start

AmyJune (volkswagenchick) @leslieg I have gather a few names and emails for folks who’d like to be interviewed. What is the best way of sharing that info? (edited)
leslieg Email them to me for now and we’ll create issues for each as the interview is scheduled/completed
bsnodgrass (he/him) I would like to repeat that not only with site builders but also maintainers (since we need their involvement)And yes I will volunteer to help with that process. (edited)
rkoller i agree with bob i wouldn't limit the perspectives to site builders and people new to drupal. more perspectives should be seen and heard. and i could help as well
leslieg I certainly agree with getting maintainers involved in the process. The concern was that developers may categorize things in a manner that would be meaningful to them, but not necessarily to the site builder audience. Do we bring maintainers in after we get a list of proposed categories or when?
AmyJune (volkswagenchick) I think bring the SBs first, and then maintainers.
bsnodgrass (he/him) Or maybe try to recruit 3 focus groups... New to Drupal, Experienced Site Builders, Maintainers.
leslieg @bsnodgrass (he/him) and @rkoller what do you think about that approach. Others please weigh in as well
rkoller you could to open card sorts with site buildes and new folks first. then see if you get a rough direction category wise and then you can either iterate with a closed card sort with sitebuilders and new folks or extend the scope to maintainers as well.
rkoller then you get an idea which categories work and which dont.
rkoller another idea might be the following. perform a certain number of open card sorts with site builders and new folks in a remote non workshop setting like leslie suggested in 4️⃣. alongside analyse the search logs and based on both results come up with suggestions for categories. and test those in closed card sorts. first with site builders and new folks and then extend things to maintainers and others willing as well
AmyJune (volkswagenchick) Leslie - I emailed you the list of volunteers

4️⃣ Identify top 25, 50 and maybe 100 modules to use for the card sort. @tim, @Andrew Olson (andrewozone he/him)do you have a list of the most popular/used contributed modules that we should focus on?

bsnodgrass (he/him) Are you talking about the Card sort exercise like we did at DrupalCon? The number we did was appropriate for the time we allowed.If it was expanded for use by a focus group, I think the lists we already have can be used for that... I'm not sure what the average number of modules used on most sites would be (I'm thinking about 50-75) Maybe the right number to consider would be more than 100.
rkoller but i wouldnt pick 100% of the modules for the sort from the short head instead at least 20% from the long tail as well to get some diversity and variety of types.
leslieg Yes, like the one we did at DrupalCon, but possibly give participants directions and a spreadsheet so they could do step 1 of the card sort, then we explain step 2 and they can do that. We would then compile the results. It wouldn’t have to be done on a call. We could give them a day or so to send us their input.
rkoller i like the idea of asynchronous
tim From that spreadsheet I was working on located[#3240314]There are 19 that at least 3 websites had recommended so far.Admin ToolbarPathautoWebformParagraphsDevelRedirectTokenctoolsENTITY BROWSERField GroupGoogle AnalyticsRabbit HoleShieldSimple xml sitemapBackup and MigrateEnvironment IndicatorMetatagTwig Tweak
tim I don't feel this is a complete list. Maybe a good starting point.   I am sure there are critical ones that are not on this list that should be.There are another 33 modules that at least 2 people have suggested.
Ron Northcutt So - to be sure I understand - we are looking for a representative list of “top modules” that we are hoping the algorithm will surface? So, we use this list to help validate the filtering options so we can tune it to get as close as possible?
chrisfromredfin Yes to this ^^ and ALSO a list to provide people as a way to do a card sort exercise.

5️⃣ an issue was created in regards to including core modules in the Project Browser results. Chris and I discussed and thought the core modules would surface as contributed modules that depend on the core modules appear in the results. For example - a user is looking for page layouts. The layout builder module would not be in the results but contributed modules in the Layout Builder ecosystem would surface and would have the core Layout Builder module as a dependency. Do folks agree with that thinking?

bsnodgrass (he/him) Yes, if only things like Core Layout Builder and Core Views could be selected as the primary ecosystem... but I've said that before! :grin:
mandclu So if a Site Builder wanted to create structured layouts, they would have to find a module that depends on Layout Builder in the project browser?
leslieg Guessing they would either search for layout or select a category related to layouts and many layout builder modules would be returned. Does that not sound right?
mandclu My personal take is that it would be better to have core modules in there as well.Hypothetical scenario: A small business owner keeps hearing from a colleague in the local chamber of commerce meetings about how great Drupal is, so he decides to give it a try. After playing with it for a couple of hours he's pretty impressed, but he wishes he had more control over the layout of certain pages. He asks his colleague and she tells him to check out Layout Builder. He goes into the project browser, but can only find projects that depend on Layout Builder, but not Layout Builder itself.The above sounds to me like it would be confusing.
thejimbirch I am also thinking of functionality like Book, Forum, Content Moderation/Workflows, Tours, and Media.
chrisfromredfin I'm growing more convinced; however, core subsystems are not "projects" on drupal.org, so they do not come through the API. They are "components" of the core project (and not ALL core components are modules - i.e. Documentation / Meetings). So to present this we will have a technical challenge of merging core modules (discovered) with data from the API (retrieved).
chrisfromredfin And how would we present things like "Layout Discovery" - I suppose we would then have to show that one, even though it's one of those "API modules" - like "install this if something tells you to"
mandclu It may be that this is technically too challenging in the initial phase, but a site builder who is new to Drupal won't know the difference between a core and a contrib module, so IMHO the desired end state should include both.You make a valid point about API modules, but the same is true of a number of contrib modules.
chrisfromredfin Oh yeah also very true ^ ctools, ACL, etc.

6️⃣ We have discussed a filter for “Use case”. The idea of creating a curated “Case Study” for each specific “use case” and then adding a list of modules that would be used for that use case. The “Case Study” would need to be vetted to make sure it is representative of the best modules to use. How would “Drupal Recipes” fit into this thinking @bsnodgrass (he/him)  “Case Study” would be a filter, in addition to Category. Do folks feel this would be a valuable addition?  Remember site builders or those new to Drupal may be looking to add a piece of functionality and would have no idea which categories to search to get all the modules they might need

leslieg Any thoughts on this?
bsnodgrass (he/him) I have started to wade through all of the notes I've collected through our "Drupal Recipes" and I do think there of plenty "use case" things that could be identified.I could see "use case" being a very important categorization for many things Drupal. i.e. Project Browser, documentation, examples, Case Studies. Drupal Recipes,  etc.
bsnodgrass (he/him) Early on, as we were looking at Drupal Recipes, I was hoping to use the current project categories or ecosystem as a way to catalog the many things people could come up with, but neither was a great fit so I plugged the gap with tags.
leslieg How do we take the next step on this?  Is there someplace you recommend that we look at as a starting point or should we have a zoom call to discuss?
rkoller i agree that recipes would be helpful and a good fit for project browser. but the editorial amount of work is huge. especially that many to all modules get covered and mentioned/used in recipes might herculean task. cuz from a user perspective using the project browser i would expect to find at least one recipe for every module i am interested in ;)
bsnodgrass (he/him) I was thinking of putting together a working list of "use case" terms and descriptions from Drupal Recipes notes... we could post that and have people submit resources (URLs) we could reference? (edited)
leslieg We were thinking that this would be added after the initial launch with text search and categories as the filters for the initial launch. We would identify what “use cases” to start with and then have someone create a “Case Study” that a set of community folks (a couple site builders, a couple maintainers, …) would review before it gets added to the “Use Case” filter.
bsnodgrass (he/him) it would definitely be a later thing, it is a big job to get enough content to provide value.
bsnodgrass (he/him) If I put an initial list together, what would we want to have in the Term description? As a {role}, I would like to do {task], to {objective}.Does something like that make sense? What else? (edited)
bsnodgrass (he/him) (Thread #6 was from yesterday)...7️⃣ Does anyone have thoughts on some designation for "use case" we discussed yesterday in the Site Builders meeting.@leslieg should I create an issue for this on D.O. and assign myself to start it out? (edited)
leslieg yes @bsnodgrass (he/him) I thought there was an issue already but not finding it in a quick scan
bsnodgrass (he/him) I will do that later today
bsnodgrass (he/him) @leslieg set this up as a child issue for the site building committee. #3246006: Define filter for "Use Case"#comment-14270153

Participants:

AmyJune (volkswagenchick), thejimbirch, bsnodgrass (he/him), rkoller, shaal, gaurav.kapoor, leslieg, Nico Grienauer (he/him), Ron Northcutt, chrisfromredfin, Jillian Chueka, tim, mandclu

Comments

NadiaFaucon created an issue. See original summary.

NadiaFaucon’s picture

NadiaFaucon commented
Status: Active » Needs review
leslieg’s picture

leslieg commented
Assigned: Unassigned » leslieg

leslieg credited bostonjillian.

leslieg credited bsnodgrass.

leslieg credited chrisfromredfin.

leslieg credited gaurav.kapoor.

leslieg credited rkoller.

leslieg credited rlnorthcutt.

leslieg credited shaal.

leslieg credited thejimbirch.

leslieg credited volkswagenchick.

leslieg’s picture

leslieg commented

leslieg credited lets4.

leslieg credited mandclu.

leslieg’s picture

leslieg commented
leslieg’s picture

leslieg commented
leslieg’s picture

leslieg commented
Assigned: leslieg » Unassigned
Status: Needs review » Fixed

Status: Fixed » Closed (fixed)

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