Revisit the modules enabled by the Standard profile

There's probably a pretty strong use case for: Media, Media Library, Content Moderation (maybe?), Layout builder (maybe?).

What even is Standard for these days, it feels a bit like a vague profile with no specific direction as to what it's supposed to contain. Should we have a separate profile for content focused sites that includes Media, LB, CM, etc? Should we have a profile that's designed as a blog/personal site starting point? There's some really great profiles in contrib, could the Project Browser initiative one day allow us to have a curated set of contrib profiles right there in the installer? If all or some of those happen, do we reach the point where the standard profile itself is removed?

(Mostly rhetorical questions, trying to spark some ideas).

Given #3158669-2: [Policy] By default deprecate non-experimental modules that are used by less 5% of sites before the next major version I think there is a definitely a strong case for adding Media to the standard profile - 32.5% of sites have it installed anyway. There is probably also case for Responsive Image with over 26% of sites having it already.

I also think there is a case for removing RDF and have worked on #3243121: Remove RDF module from the Standard profile for this.

> I think there is a definitely a strong case for adding Media to the standard profile

If Media isn't in the Standard Profile, doesn't that mean that the Article node type set up by Standard uses a plain image field, which is directing people down a site structure pattern that we shouldn't be recommending?

@joachim good point - to me that is an even stronger reason for adding Media to the standard profile, and reconfiguring the article image field to be a media entity.

Should we made a child issue to specifically convert the article node type to use media? I feel that this issue is more of a strategic thing that's clearly taking a while, whereas encouraging people to use image fields instead of media should be done a bit more urgently.

@joachim sure, this is a "plan" issue anyway, these don't usually get committed directly but work happens in child issues.

Adding some related.

The objective of the "Easy out of the Box" initiative is to make Layout Builder, Media, Media Library, and Claro feature-complete and part of Standard. So I think we should leave that out of scope here for now; it has an entire initiative.

For reference, here's the list of what's in Standard, and their usages as of 2020 from #3158669: [Policy] By default deprecate non-experimental modules that are used by less 5% of sites before the next major version.

  - node 89.25%
  - history 76.06%
  - block 89.69%
  - breakpoint 87.39%
  - ckeditor 87.1%
  - color 75.03%
  - config 86.68%
  - comment 68.67%
  - contextual 84.19%
  - contact 70.6%
  - menu_link_content 86.96%
  - datetime 86.21%
  - block_content 85.55%
  - editor 87.27%
  - help 80.8%
  - image 87.39%
  - menu_ui 86.82%
  - options 87.25%
  - path 87.3%
  - page_cache 92.83%
  - dynamic_page_cache 91.96%
  - big_pipe 51.27%
  - taxonomy 86.04%
  - dblog 82.1%
  - search 67.27%
  - shortcut 78.78%
  - toolbar 87.23%
  - field_ui 86.02%
  - file 100%
  - rdf 79.18%
  - views 83.42%
  - views_ui 83.42%
  - tour 71.87%
  - automated_cron 80.01%

If site owners have gone out of their way to turn a module off, that might be an indication that it doesn't belong in Standard (even if it does belong in core). A couple thoughts:

1. Tour's low numbers make sense because there honestly isn't a lot of Tour content in core. Tour should have been a huge UX win, but we never did get enough Tours written to make it a default part of the core admin experience. Not sure what the path forward is there. I still believe it's a great feature, but if there aren't enough Tours to make it an expected part of the UI, people won't use it. OTOH, it's also probably common for people to turn if off if they already know how to use their site. (Although note that Help is used on fully 10% more of sites than Tour is.)
2. Comment is quite low compared to most of Standard, which makes sense because commenting, while it is a common feature, is also not desirable on many websites. Comment still belongs in core, but maybe it doesn't belong in Standard anymore?
3. Shortcut seems like a good candidate to turn off in Standard and see what happens to its usage stats over time. I have a feeling this is one of those "It's just there but I don't use it" things like QuickEdit.
4. There are a couple Standard profile modules where a popular contrib module is often the preferred replacement. Once Project Browser gives better discoverability for those contrib modules, maybe the core equivalents could be at least dropped from Standard to start. In particular:
   • Search is also very low, especially for a feature that pretty much every site on the internet needs. 10 years ago it was common to replace Drupal's built-in search with a Google search. More recently, Search API is used on 100K D8+ sites according to its stats, which would mean like half of them.
   • Contact is quite low. Probably gets turned off on sites where Webform is used instead.

Usual caveats blahblah about how this is based on update.module data etc.

@joachim, re:

Should we made a child issue to specifically convert the article node type to use media? I feel that this issue is more of a strategic thing that's clearly taking a while, whereas encouraging people to use image fields instead of media should be done a bit more urgently.

I'd suggest looking at the #2825215: Media initiative: Roadmap and adding your thoughts there if they aren't already covered.

For easier scanning, here's the ones from #11 that are below 80%:

  - rdf 79.18%
  - shortcut 78.78%
  - history 76.06%
  - color 75.03%
  - tour 71.87%
  - contact 70.6%
  - comment 68.67%
  - search 67.27%
  - big_pipe 51.27%

I don't think that popularity should be the only factor, but I do think each of these should be evaluated for whether we think they're important to keep in Standard or in core at all. We've already done that for RDF in #2152459: [Policy] Deprecate RDF module and move it to contrib.