The domain https://fediverse.observer redirects to sedo now.
On October 11th these two commits where made to the repo:
this will be my final release, glad to have contributed since 2011 to some fediverse stats, https://fedidb.com/ is a great resource with more interest
this software is no longer developed
Sad to have lost such a great resource.
Encouraging people to contribute only goes so far when this place is simultaneously discouraging people’s contributions. It is systemic, not limited to ML, and needs major efforts including enhanced technical capabilities to try to solve. Note that if ML ever were to be defederated, the absolute most toxic trolls would simply switch to using their alts on some other instance such as Lemmy.World, which they have announced they will do in advance, several years ago. They even indicated that they would make apps that automate the process of trying to send messages to someone, cycling through an array of alts until one is found that is not banned on that instance (or by that recipient?) - no truly does mean to simply push forward all the harder, for some.
But making a more firm stance would also send a signal about what is desired vs. not. Are calls for murder of all people in the Western World acceptable? Apparently, judging from their presence, yes.
Most places are defederated from HB - which was not always the case (I was on both StarTrek.Website and Discuss.Online prior to blocking them), but now that it is I don’t worry about them so much as ML that is still sending out content that turns people away.
I don’t know how to solve this problem, I only showed references in support of the fact that the problem lies there in front of us. Or at this point, behind us, since once a reputation is gained (for e.g. nostr, 4chan, etc.) it is exceedingly difficult to change.
On the whole, I think our reputation is closer to nostr and 4chan than it is to Bluesky. Institutionally at least, anything goes - including the worst CSAM. Something must be specifically blocked in order to filter it out (either that, or a new instance can include only trusted existing instances, but that runs into problems where new ones pop up all the time and will get excluded, so this approach seems not to be used by the vast majority of existing instances, at least the major ones).
Wonderful!
That team is so energetic and helpful, it’s a pleasure to see it! 😁
My understanding is the trolls exist because .ml admins allow it. If they try using .world, I expect the .world admins would ban. If they ban evade, I imagine a new safeguard would be added to counter them.
I don’t go to other social media, so my only knowledge of our reputation is from people repeating things here.
Hopefully “our reputation” can jump to a different comparison scale (I wouldn’t want to be compared to any of those options).
Okay, I’m going to push back on this. At no point have I ever heard CSAM being accepted or allowed. The only time I’ve ever heard of it coming up, everyone reacted negatively. The poster was blocked, all the posts were deleted immediately, and new safeguards were put into prevent it happening again.
Agreed :)
Edit: missed a section.
Perhaps this is an exceedingly poor way to describe it - I just meant that federation (like with ML, hexbear, lemmygrad, and any new instance) happens by default. Instances that host CSAM - like extremist far-right content - are quickly defederated from by the rest of the Threadiverse, but such requires effort on behalf of the instance admin. I thought, though perhaps I am wrong.
Anyway it could still be misleading even if correct, bc this is at the instance level, whereas someone still has to subscribe at the community level in order for content to begin coming in. So the potential for content - good, bad, or whatever - to spread is baked right into the ActivityPub Protocol from the start, by design. Just like what people keep saying about 4chan (though admittedly I never went there personally).
And people must have subscribed to hexbear and lemmygrad and ML communities on all of the major instances, therefore all of that content was already there for me to stumble upon as a new joiner myself, fresh from KBin where it had been severely curtailed. I - like all other new joiners - then bad a decision whether to flee this place permanently (which I almost did, and continue to revisit that thought every year or so as expectations keep not being met, making it all the harder to hold out much hope for the future) or to begin blocking people, one by one, and even making a (successful) petition to defederate discuss.online from hexbear.net.
All of which is a lot of effort, especially since it must continue to happen with each fresh new alt made to subvert bans. Each new joiner must then be subjected this again and again, like a hazing ritual that we who have been through such already now offer little to no help to new joiners to avoid. Except for your codeberg request that will help to change that - hooray!
Ultimately, it’s arguably fine for places like 4chan and nostr to exist (unless you take the position that the echo chamber then breeds toxic trolls that when they leave it, act to harm others, though keep in mind that in this analogy, the Threadiverse itself is 4chan and nostr, in that we allow anything and everything, especially on ML that we then host on all of our machines too - e.g. if ML were to ever go down, that content would continue to remain, foreve, just like happened with Lemm.ee), but what’s far less fine is to allow people to stumble upon it unawares, particularly when it is labeled oppositely from what it truly is. A place’s reputation should proceed it, not be entirely ignored and force each new joiner to discover it all over again on their own, as what happens with ML.
I have no source for this, but I think, now, first content/post/comment from a new instance has to be approved by an instance admin (could be wrong and misremembering though). If not, might be worth seeing if that is a option admins would want added. Thoughts?
On that note, it seems Pylova has finished writing up the block subscription code (yesterday) and it’ll be in the next update (unknown date).
It could remain forever, but I think most instances (or at least Piefed instance) are set up to delete old content after “x” amount of time. Alternatively, admins could just manually delete the old content, if they wanted.
Agreed. I think .zip’s approach where it has the default blocks and explains them is a good approach. If the subscribable blocks also make it into the on-boarding, I think there will be a large swath of new .zip users who almost never encounter the triad issues (assuming Pylova subscribable block list spreads back to Piefed and beyond). If that on boarding spreads throughout the Threadiverse, the triad could be dealt with, without major confrontations, which I think most admins don’t want to deal with.