… but I wanted to make sure everyone here had a chance to see it. I’m happy that I’ve been able to announce our list of finalist candidate stacks.
Once we’ve got them spun up - later this week, if all goes well - we’d be grateful for your feedback. I’ll have docs available with the URLs as soon as we’re ready to go live, and I’ll be setting up feedback channels on each of the servers, as well as a thread here. (And you can always email me, obviously.)
Ok, here we go! As promised, I am very happy to be able to point invested Mozillians at the test instances of our candidate synchronous messaging stacks.
On each of these servers, we’ll be taking feedback in the #synchronicity channels on each of those services, as well as in this thread. And, of course, you can always email me - mhoye@mozilla.com - directly.
Let’s give this stuff a shot, see how it works for us!
I got the following error on every platform when trying to connect with a FxA
" requires you to setup additional security measures for your account, such as enabling multi-factor authentication (MFA) or using a safer authentication method (such as a Firefox Account login). You will not be able to login until this is done."
nukeador
(Rubén Martín [❌ taking a break from Mozilla])
11
Merging topics for visibility and context. @mhoye my understanding is that we have a month to test them and provide feedback right? So until October 12th?
I’m going to pin this thread because it seems pretty important.
Thoughts so far:
Mattermost
Lacks ability to mute/ignore an individual user. I think this is a must for public services like this. Even IRC has this (but Slack doesn’t either). Threads don’t collapse. It almost feels like it adds more clutter to have it this way than if they didn’t exist at all.
Pretty similar to Slack from a UI standpoint
Riot
No custom reactions? I know that’s super unimportant. Still a bit rough around the edges. Creates a private channel as soon as you choose to open a conversation with somebody. Not a huge fan of that but it could very well be a non-problem for 99% of other people. Same type threads as Mattermost.
Matrix developers are already engaged with us. Seems pretty granular with permissions and such I feel like this will give people the most flexibility with clients. I haven’t looked too far into clients for the other two, but Matrix does have a number of other clients available if Riot isn’t some people want to use.
Rocket.Chat
No native dark mode yet Differentiation between threads and discussions. It seems like an interesting concept, but I’m not sure what the intended use-cases for each are. I’ll have to play around with each later.
I’ll update this post a few times as I use each of them more.
Reactions are in a bit of a weird state in Matrix. The spec has been proposed, which does allow completely arbitrary text as reactions, but only Riot iOS and RiotX (Android) support picking emoji other than the 8 shown in Riot web.
At a technical level Slack does this too - using wee-slack I get a DM window open the moment someone who hasn’t sent me a DM yet.
1 Like
nukeador
(Rubén Martín [❌ taking a break from Mozilla])
17
One thing that I consider super important is content moderation and flagging, specially in a place we want to be welcoming and safe for everyone and following our CPG.
I failed to find how this works for the 3 candidates.
How can I flag a comment?
How does moderation work?
Is there a community-moderation system? (to avoid bottlenecks on a few mods)
How content can be deleted for everyone?
Is there a way to block a channel if there is a flood attack? (a lot of people posting nonsense)
Is there a way to take actions on a malicious account?
/cc @Emma_Irwin1 since I think this is super important for a channel with a lot of live comments.
Matrix project lead here; We have an extensive moderation guide up at https://matrix.org/docs/guides/moderation which explains the options available on Riot/Matrix. We’re doing a lot of work in this space currently (the guide had a fairly major update yesterday); this is a massive area of focus for us.
Edit: to answer the specific questions:
You can flag a comment by selecting ‘report’ (although this is only present currently on Riot/iOS & Riot/Android - we’re just adding it into Riot/Web.). It reports the message ID into the server DB; the server admin can then decide what to do with it.
For moderation in general, see the guide.
Community moderation is supported
Content can be deleted for everyone using the ‘Remove’ button, which now also supports bulk removals
You can make rooms invite only, or mute-by-default to mitigate flooding attacks. There are also much more sophisticated plugins and bots and tools too as per the mod guide (e.g. nsfw-filtering bots)
You can deactivate malicious accounts as per the guide.
Signed up yesterday, for the betas and will help as much as i can
nukeador
(Rubén Martín [❌ taking a break from Mozilla])
20
Thanks, this is helpful. I see you are considering “Using reactions to filter content” as a way for community to help directly on the moderation, that’s nice.
nukeador
(Rubén Martín [❌ taking a break from Mozilla])
21
Am I the only one with issues using the mobile version and SSO?
Lacks ability to mute/ignore an individual user. I think this is a must for public services like this. Even IRC has this (but Slack doesn’t either).
Pretty similar to Slack from a UI standpoint
Differentiation between threads and discussions. It seems like an interesting concept, but I’m not sure what the intended use-cases for each are. I’ll have to play around with each later.