Restore People/Favourites metaspaces

[justjanne]

Fixes #30861

In Short

This PR restores the "People" and "Favourites" metaspaces for the new room list.

Until this gets merged, I'll maintain

• a deployment
• docker images
  • v1.12.0 k8r.eu/justjanne/element-web:v1.12.0
  • v1.12.1 k8r.eu/justjanne/element-web:v1.12.1
  • develop k8r.eu/justjanne/element-web:develop
• packages for Ubuntu and Windows (TODO)

Rationale

Many group chat apps have a separate view just for DMs. Discord calls this "Friends" (vs "Servers"), Teams calls this "Chats" (vs "Teams"), Slack introduced the "DMs" view in 2023 (vs "Channels").

Many people rely on this feature, and so do I.

Checklist

• I have read through review guidelines and CONTRIBUTING.md.
• Tests written for new code (and old code if feasible).
• New or updated public/exported symbols have accurate TSDoc documentation.
• Linter and other CI checks pass.
• I have licensed the changes to Element by completing the Contributor License Agreement (CLA)

[t3chguy]

Code-wise looks fine but needs product sign off to continue

[t3chguy]

Actually, upon testing, this doesn't seem to return either of the setting pages necessary to enable those metaspaces, so this would only work for users which already have those metaspaces enabled.

[justjanne]

Actually, upon testing, this doesn't seem to return either of the setting pages necessary to enable those metaspaces, so this would only work for users which already have those metaspaces enabled.

Good point, should've thought of that. EDIT: Done

[CLAassistant]

All committers have signed the CLA.

[justjanne]

Are there any tests I'd have to update for this? I haven't checked whether the new room list changes also affected the E2E tests yet. EDIT: Found them.

[t3chguy]

LGTM if Product approves

[daniellekirkwood]

Hi Janne! Thanks for the contribution, I'm taking a look into this. Thanks for your patience with me!

[daniellekirkwood]

Hey,

Some thoughts...

• Slacks DMs view is more aligned to a future idea we have for the notification bell (currently in labs). They appear to show something slightly different from a room list because they assume that messages in DMs are more critical than messages from Channels. I don't think we can apply that same assumption to Element.
• The Teams "People" list is not a list of chats but a list of Contacts -- I don't think this is your idea for the meta-space? I would really like us to explore further the idea of having a directory, it'll be hard to do for Community use cases and servers like M.Org. Otherwise, Teams has the same pill selectors we do (channels = rooms & people = chats). They do have a room list sections concept though and that is something we'd like to do/look into. Designs are in progress.

We removed People and Favourites meta-spaces because it didn't make sense to have 2 filters so close together (ultimately the meta-space is a filter on your room list). We currently aren't planning on keeping any of the meta-spaces in our long-term vision so adding them back in isn't something we'd be looking to do right now.

[Conan-Kudo]

The problem is that your vision completely conflicts with how people want to use the app. Most of us mix among all three classes at the same time. The fliter mechanism makes the assumption we only want to operate on one type at a time, and that doesn't really line up with us at all.

[nie7321]

If you are drawing inspiration from Teams, I would point out that Teams still has their equivalent of the "Favorites" meta-space at the top.

[spaetz]

I don't often comment on this because it is not constructive or helpful, but I need to get this off my chest:

• Using MS Teams as an argument for improving UI is perhaps not helping the case, as MS teams is a UI desaster (sidenote: also your boss just said in a room that Element is trying to be WhatsApp/Signal and not Teams/Slack for now :-), so why resemble them?).
• I don't understand why we need to compare to any other product anyway: there was a UI change and the peanut gallery (us) revolted, we perceive it as a change for the worse. There is no need to argue what other products do, if people are that vocal about a change, it might not have been the smartest of all ideas, full stop.

It is your full right to ignore the peanut gallery and implement your vision, but this is a rare case of a community MR and simply ignoring it like that paints a very pessisimistic picture on your willingness to accept community contributions per se (there are not that many). Sorry, I will stop here.

[justjanne]

if people are that vocal about a change, it might not have been the smartest of all ideas, full stop

If you compare the old and new room list, you'll see that most of the changes were engineering-driven, as the old roomlist was hard to maintain and the new room list will provide a solid foundation for future changes.

I don't understand why we need to compare to any other product anyway

I originally made the comparison to other products, Danielle just replied to my comments. The reason I made those comparisons was to paint a picture of how people use Chat apps, and which requirements people have. I especially focused on the more business-oriented apps and needs that the new room list is unable to cover.

but this is a rare case of a community MR and simply ignoring it like that paints a very pessisimistic picture on your willingness to accept community contributions per se (there are not that many).

Note that "community PR", while technically accurate, isn't the entire story – I used to work at Element (even on Element Web itself), and I still work on other matrix clients nowadays, so I can understand why the new room list was necessary.

That's in fact why I'm proposing to port the most important features to the new room list, instead of just reverting to the old room list.

[edwardsdean]

will provide a solid foundation for future changes.

if its not ready then dont push it, its a horrible design and shouldnt have been pushed

[justjanne]

will provide a solid foundation for future changes.

if its not ready then dont push it, its a horrible design and shouldnt have been pushed

Note that I don't work at Element. In fact, I don't believe the new room list is quite ready yet either (there's a reason I've opened two PRs to reintroduce some of the old functionality).

But I'd prefer if you don't use my PRs as battleground for your complaints. I'd prefer if we could use this space to focus on a constructive discussion about the PR itself.

[Nemric]

Hi, I'm reading this PR comments and, I don't know what are the "strategic" constraints but, my thought is that each filter key (unread, people, invite, rooms, low priority, ... ) could be use as a "group by" key in UI ?
So we can manage our own room list with groups defined by selecting them in a list (I want groups for favorites, people, low priority as an example)
what do you think ?