Clarity on stance with regards to Threads

I double checked and we did not have a rule in place. I took the liberty of applying the limit since we hadn't done that yet (though of course if anyone feels like I've overstepped, please let me know, it's easily reversible). We currently have 3 accounts on threads.net stored on the server with a total of 25 followers across them (most of which are for the instagram founder guy). These accounts will not be affected by the limit, and searching for accounts won't be affected, but posts from threads will no longer show up in the federated timeline (limit really doesn't do much).

@Sam Whited

Thanks. I can't imagine this was overstepping, given that it's what the current policy says should happen.

@Sam Whited @Dynamic Note that @Erik Moeller's toot is also doing the rounds on the fedi now, and boosted already 281 times: https://social.coop/@eloquence/111588877096843391

@Sam Whited

What do you mean when you say "these accounts won't be affected by the limit"? Here's what I see if I try to follow Mosseri on social.coop now:

I'm assuming this means that any posts of his that get boosted also won't show up in my timeline until I follow him. If that's not the actual behavior, that would be great to know!

@Erik Moeller oh yah, I forgot about the warning. You can still show the profile and follow it though. I'm not sure how boosts are handled and I blocked the domain on my account so I can't test it. I could unblock it for a bit and then re-block it if someone wants to check.

@Sam Whited Yeah, that's the big difference to suspending. I'm around for the next hour or so, ping me on fedi if you want to do a quick boost test :-)

@Sam Whited I just saw a post from a Threads user I didn't follow, boosted into my home timeline by another user I followed. So a server-side "Limit" definitely does not stop that.

There's a bug for this, though it's not clear yet if the Mastodon folks themselves see it as a bug: https://github.com/mastodon/mastodon/issues/26301

@Erik Moeller

Thanks for sharing this. Not happy to hear it.

This seems like an interesting development: https://mastodon.social/@dansup/111617703110836835

I'd really like to see us turn on authorized fetch if we haven't already (especially in light of threads); I wonder if anyone on the tech team could weigh in on whether the extra authentication load for us would cause any problems? I suspect it would be fine, but they'd have a better view into how close to capacity the system is running than I would.

EDIT: See also this thread about authorized_fetch where discussion should probably happen: https://www.loomio.com/d/Ry5oCGg4/enable-authorized_fetch

Hi folks, I really appreciate that this conversation is happening, especially because I don't have capacity right now to dive into the details myself. Thank you!

May I ask that once things have been finalized, someone create a guide, even if it's just a few bullet points, for what the policy is, and how we can find and follow people on Threads if we want to?

Good idea @J. Nathan Matias it seems like a good page to have on wiki.social.coop that summarizes and points to the relevant polls and discussion?

I'm noticing that social.coop is listed on the sites supporting the threads block: https://fedidb.org/current-events/anti-meta-fedi-pact

Since we voted to limit and not block I guess we should ask for it to be removed?

@Ed Summers

I agree that we shouldn't be on that list. I wonder how we got there.

Is there a list of criteria that would cause us to suspend rather than limit? I would imagine that it may become necessary to suspend, but having clear guidelines around that would be nice.

@Jay

I haven't been involved in community working group (CWG) moderation activities personally, but my understanding is that our current culture is to err on the side of trusting moderator judgment rather than trying to formulate one-size-fits-all rules.

The proposal that was approved stated that the CWG would moderate Threads at the limit level and would otherwise treat with the same moderation approach as they would with other instances. So I would be very surprised if there is an enumerated list of criteria.