In reply to Chronocide: “@khurtwilliams@islandinthenet.com As someone who’…” …

Basically a wide swath of people that left mainstream social media due to being harassed and bullied, have cultivated an “alternate social media network” out of Mastodon. They don’t want the Fediverse to turn into Facebook basically, so a lot of the instances they’re on have chosen not to federate with Threads before it happens.

Nevermind the fact that anyone that disagreed with this stance but has their account on their server, now must move to find one that works. (This is the reason I use my blog as an instance; I can follow/block whatever I want.)

(cc

15 responses
  1. @starrwulfe You might not have realised it yet, but this is how the #Fediverse works. It's federated. So you choose an instance that corresponds with how you think an instance should be run (as best you can, as there may never be a 100% fit). Then if the situation changes, you move to another instance that is a closer match. That's why there's functionality to support movement. People voting with their feet keeps the whole thing healthy. That's not just about #Meta. #fediPact @khurtwilliams

    1. @tokyo_0
      Not sure if you’re being cynical towards me or not, but I’ve been here in the fediverse since 2017, starting with my own Hubzilla instance on a VPS, so I have a bit of experience. This whole “choose an instance” design needs to be updated since we’re no longer accommodating other gearheads like myself; this is the reason why “twitter refugees” and the like who just want to find a space to continue to microblog get flabbergasted with Mastodon and the like here in the #fediverse. It’s a pain point and not easily understood.

      I have several accounts and have my own instances even, including this one which is actually running WordPress as an #ActivityPub actor. I moved my main Mastodon instance from mastodon.cloud to vivaldi.net due to the whole preemptive Threads.net block.
      My complaint here is I was on an instance that I was happy with for 6 years until this. I shouldn’t be made to uproot my entire online presence without some sort of due diligence like that. I’m a very technical person, and I was already prepping to change anyway due to other factors before Threads was even a thing, but most people are not and I could imagine scenarios where a layperson trying to follow their friend suddenly can’t because they’re blocked. Moving instances is NOT easy. You have to set up a “forwarding” notice in the place you’re moving from, and also a “masquerading as” notice in the place you’re moving to for a while. Then depend on your followers to get the message and move accordingly…or wind up dropping you. Then at some point, go back to the old instance and shut everything down. It’s a big friction point that should be improved on (which is happening actually) and should only be used as a last resort by most.

      In this case, why not leave it in the hands of the users and let them set their own blocking/muting/visibility settings? Bluesky actually has a distinct advantage here with subscription based stackable moderation tools and I wish such a paradigm could make it as an ActivityPub FEP. However, the devs working on Friendica, the Misskey forks, and some of the newer Fediverse servers have incorporated their own strong and easy to understand granular blocking settings too, so all isn’t lost.

      However it’s really all up to the owner of the instance to implement and monitor this all, and no availability of better filtering tools will account for a tyrannical instance admin who doesn’t do their due diligence in talking to their users about a strategy that will help everyone on it.

      1. @starrwulfe The process of migration is fairly smooth on Mastodon (migrating followers, that is – content is a very different story). Hubzilla is even better but only if you consider moves within Hubzilla itself. A lot of things seem to go wrong with how Hubzilla interfaces with AP, but I don't think that's so much to do with migration as just the way AP has been bolted on to Hubzilla.

        1. @starrwulfe Having to move after such a long time is an upheaval, and you're right to say it could be improved a lot, but surely the admin of your instance gave some consideration to the decision. While it's an unfortunate that the decision didn't go your way, it sounds like that's the biggest root of your discontent. I wonder if you would have been so dissatisfied with the level of consideration they gave the decision if they'd decided to federate with Threads as you'd have preferred.

          1. That’s just it– In my case, there was 0 communication. Bans just pop up on the “moderated servers list” and any attempt to communicate to the admins goes to /dev/null and zilch. If they did nothing and just left things federate across, I have no problem muting, blocking, etc on my own. I prefer it that way. The internet is a public space and sometimes you drive through the ghetto to get to the good part of town. But don’t just wall off parts of the city without saying anything. And that kind of power being wielded in such a manner is tantamount to censorship. I vehemently oppose censorship in all its ugly forms.

            1. @starrwulfe That does sound frustrating. I've been on a couple of instances before now where the admins just don't respond. These days before moving to an instance I send a message to see if the admin is willing to interact with a mere mortal like me 😅 but I'm very fortunate that I recognised the problems with noncommunication on the past instances and moved before I ran into any issues I really needed support with.

              1. @starrwulfe There only seem to be a handful of instances that actually have responsive admins, too. It's a problem.

        2. @tokyo_0
          There’s FEP-ef61: Portable Objects being looked at to evaluate just that actually.

          I for one would like to figure out a way to combine the strengths of all protocols into the fedi myself…

          1. @starrwulfe IIRC, I think Mike Macgirvin, the developer of Streams and Hubzilla, has recently started working on adding Nomadic Identity to ActivityPub

          2. @starrwulfe There has been a lengthy discussion for about five years here: https://github.com/mastodon/mastodon/issues/12423

            The lack of progress in that thread is how MastodonContentMover came to exist.

            Recently I heard that Mike Macgirvin (who wrote the protocol Hubzilla is based on) was looking at a way to build content portability into ActivityPub (looking at his recent posts I think that may be connected wih the link you shared). I think that's going to be really tough, but am interested to see how it works out.

            1. @starrwulfe Just finished reading your link — it looks really promising. I hope they can iron out any risk of the duplication or conflict issues Hubzilla used to run into when sending updates from mirrored identities (replicated on multiple instances) across AP. When I was testing before, there was this issue where if you're hosted on multiple instances and you send an update to AP, it doesn't just get sent from your primary instance but from all of them. Then, here on the, … (cc @mikedev)

              1. @starrwulfe … Fediverse side something would happen with the multiple copies of essentially the same update, where whichever arrived first would be recognised, and the subsequent arrivals would be disregarded. What that seemed to create was some kind of race condition between the different Hubzilla instances hosting your profile, so that rather than the primary account showing updates here on AP, each update would only show up from the instance that happened to send it first. @mikedev

                1. @starrwulfe At least, that's what seemed to be happening. It was difficult to get much cooperation from the admin of the instance I was on or the dev who is handling Hubzilla now. I understand @mikedev had already moved away from the project by that time as well, and I don't know if he'd had any involvement in how Hubzilla became connected to AP (the identity portability within Hubzilla works so smoothly, and the AP plugin on Hubzilla really doesn't, so it's hard to imagine he could have).

                  1. @starrwulfe I wanted to try to help iron those things out but by that point there were so many things to address and there was so little interest from the dev there, I moved back to Mastodon and built MastodonContentMover instead as a workaround, in hope that the existence of a third-party tool would prompt the Mastodon developers to do something to add content portability within the system itself. I really hope what @mikedev is working on achieves that — it would make a really huge difference.

                    1. @tokyo_0
                      From what I know about @mikedev, portable identity is his Holy Grail. Especially since there’s an FEP for it now, he’s actively working on it and posts about it from time to time. Maybe you might consider working with him on the project since it aligns with your goals as well?
                      How it was initially written as you outlined earlier, would seem to create a mini-DDOS situation with multiple instances trying to “win a posting race” if that’s how it was designed. There’s likely more to it than what my imagination is conjuring up I’m sure. I’ll follow Mike’s feed and monitor the development. Hopefully it’s something that will catch on and can be incorporated upstream in the next year once developed.

Leave a Reply

Your email address will not be published. Required fields are marked *