This same exact thing happened to the blind community, of which I'm a member of. Blind Twitter was pretty vibrant before Elon took over, but that changed almost completely around November/December of last year. The killer change for us was the death of third-party clients, on which blind people relied almost exclusively. The whole community is on Mastodon now, mostly concentrated around two instances, though there are plenty of people elsewhere.
Glad to hear a new home was found. My child and I began to analyze random public braille the other day out of random interest and found it was surprisingly easy to pick up a few letters, apparently being both alphabetic and potentially conceived as some kind of semi-derived state from regular roman shapes which provides a handy memory reference for we sighted people. I'd be very interested to hear what the best braille interfaces are. How long does it take to learn to use it for input? How do totally blind people handle the presumed need to switch between typing (on a regular keyboard, I would assume) and "input" (fingers on a braille generating device) modes at once? I suppose the staccato nature of this type of interaction harks back heavily to the old days of Unix, limited baud terminals and early micros.
Braille screens are prohibitively expensive. Most blind people on the internet either have enough sight to navigate a heavily vision-optimized interface (high-contrast, large text) or use a screen reader, or both.
As far as input goes, braille keyboards are more affordable, but many users can touch-type (and have their screen reader read back). Dictation is also a mature option at this point.
Blind people need 3rd-party apps because these apps can be more compatible with screen readers, or can be themed in an easier way allowing for blind-friendly presets. In general, official apps tend to be heavily obfuscated to prevent automated scraping and puppeting. Since blind users often need to basically puppet their devices by not using common interfaces, this puts them at odds with corporations who do not care about them and would ban them if allowed. The alternative is of course not to use big tech products but just because you have a disability shouldn't mean you are abandoning at least half the current web.
Understood. What do you feel the tipping point would be on price and features for an open source braille display? In particular, what is the required number of rows / columns for English to make it very useful?
All of the infosec people I follow have moved to Mastodon. As someone with primarily tech related interests, I'm currently finding Mastodon as good or better than Twitter at it's prime.
But they're specifically moving to Mastodon (infosec.exchange and defcon.social), which is not the same as moving to Lemmy, KBin, or PixelFed. For example, Lemmy federates with KBin, but not with Mastodon. Kbin federates both with Lemmy and Mastodon. Also, Mastodon has a Twitter-like UX, while Lemmy/Kbin have a Reddit-like UX. PixelFed has a Flickr-like UX.
It makes a difference, and insisting on calling them ActivityPub or Fediverse servers has strong GNU/Linux vibes, which we should probably avoid too.
You are making the same mistake. There is no Mastodon as a place or protocol. There's ActivityPub and a bunch of server/client software. Mastodon is one of those
People are allowed to be private and not want others to follow them by not saying exactly where they moved. I am allowed to voice my frustration about that.
Segregating communities is the trend now (post web 2.0 social platforms), I don't think that is over dramatizing things.
I'm more about tech people casually dropping they moved to Mastodon, us techies kinda get Mastodon is not a single place like Twitter so the only explanation for not saying "I moved from Twitter to such or such community" is not wanting people to know.
> us techies kinda get Mastodon is not a single place like Twitter so the only explanation for not saying "I moved from Twitter to such or such community" is not wanting people to know.
Eh? There are _some_ Mastodon instances which are to some a community distinct from Mastodon writ large, but that's not most of them. I moved from Twitter initially to mastodon.online (one of the giant ones, not really a community at all), then to mastodon.ie (the generic Ireland one; marginally a community, but only very marginally), and reserve the right to move further if appropriate. But I don't see why anyone would care about the distinction between them, really.
Generally, I would assume it's less not wanting people to know, and more that it's irrelevant detail. I think this is maybe a common point of confusion over Mastodon; for most people, for most purposes, the instance _simply does not matter_. There are some generally very small instances which are highly focused on specific communities (and may even require that user content revolves largely around those communities) but those are rare; I'd expect under 1% of the user ship. For most people Mastodon is the community, not their instance.
If someone says "I used to do business mostly by post and fax, but now do it mostly by email" you wouldn't go "ah, but are you using gmail or hotmail? It must be a conspiracy because you don't want to tell us!" They don't name the email provider because it is _irrelevant_.
Infosec twitter may be gone, but the infosec spam bots are still there. Just try searching for "Linux kernel" and try to find content by a real human. Before you manage to find one, you'll have to scan through hundreds if not thousands of low-quality bot posts about the latest linux kernel commit, linux kernel CVEs, or linux kernel mailing list posts.
Holy carp, that's handy! The only downside to moving to a new social media platform is curating a new set of people to follow and unfollow, which is more overhead than I'm willing to endure. But starting with a list like this means just having to unfollow people who take a "build a brand" approach via quantity rather than quality. So, thank you.
This is... fairly normal; possibly people forget because we have been in a period of relative stability for social networks, but this is how it usually works. It's basically how web forums reproduced, when those were the primary form of social media; forum starts to get a bit shit, people get annoyed, communities within that forum start decamping en masse for elsewhere (often a completely new forum). It's a fairly natural human behaviour; if the people you interact with go to X, you probably go to X, too.
I mean, what do you expect to happen? 20% to Mastodon, 20% to Threads, 20% to T2, 20% to Bluesky, 20% stay, to avoid being perceived as a 'hivemind'?
This is actually one reason I think it's rather unlikely that there will be a single Twitter-killer, with everyone from Twitter moving to Threads or whatever and then continuing on as normal. Threads will probably be the biggest (particularly if it manages to launch in the EU; that's a big problem for it right now), but I'd expect a year from now that there'll be Threads, a bunch of Mastodon instances (maybe more loosely federated than today, probably more community-oriented), maybe Bluesky (I don't personally _get_ it, but some people seem to like it) and a rump Twitter.
Many had automation that broke with the twitter api being rug pulled.
Others don't care for the way they systematically dismantled their trust and safety team.
And finally, some just don't like the direction the platform is heading.
Twitter has made it clear they don't care about their employees, their platform users or anything but frantically trying to extract a bit of money out of zombie company.
The tech community in general but especially the infosec community has strong representation from the LGBTQ community. It should be no surprise that Musk’s antics would drive that community off the site before others. The fact that they are more technically competent than other Twitter sub-communities (like sports or movies) means that the initial learning curve of Mastodon was not a blocker.
I think the reason is more practical. Mastodon is distributed, user-controlled, and easily managed. You can just choose to simply not interact with privacy-adverse systems rather than being forced to use a single centralized system.
But it is. Users chose infosec.exchange because it promised what there were after, but if the instance admins start misbehaving, it would be significantly easier to migrate again to another, or their own, Mastodon instance. The difficulty of search in federated spaces means users are more likely to centralize themselves, but the movement federation offers ensures no funny business on an admins part.
If most the infosec professionals consider the latest changes have made Twitter an inviable option, that alone speaks volumes of the quality of those changes.
Also, a highly changing environment (regarding security threats) makes strong networking a survival trait. You don't want to become isolated from a part of the community, and you can always keep both Twitter and Mastodon. In fact, keeping both accounts is a must, because many security companies and government institutions are still only on Twiller.
> As it stands now the "fediverse" resembles a collection of islands with various shifting bonds between coalitions where those from coalition A shun those from coalitions B and C.
Hrm, what makes you believe that? There are a _few_ largely disconnected instances, but, in particular, the vast majority of instances in terms of usership (excepting intentional islands like Truth Social and Gab, and that one Japanese one) do federate with the giant instances like mastodon.social and similar, tying the whole thing together. In general, if you're using a "normal" instance (ie either one of the big ones or one moderated in a similar way) you can see most other users and most other users can see you. It's only if you're on either a service which everyone wants to block (rare, mostly small Nazi-oriented ones or ones which tolerate cryptospammers) or a service who blocks practically everyone (overly trigger-happy; these tend to rapidly wither and die) that you're isolated.
> it would be far less damaging to the structure of the "fediverse" if that blocking was user-based instead of site-based
Both are options. Defederation, for most instances, is the nuclear option, not the first tool in the box.
I do think at some point it's somewhat likely that there'll be a Muskian 'free speech' shadow fediverse (ie the sort of place Mr Cat Turd would be proud to call home), which will be largely detached from the main one but will federate amongst itself. There's some indication this is kind of developing already (many of the instances that nearly everyone blocks do federate with _each other_). Lots of viable islands seems less likely, though.
1. Both high quality and high quantity posts instantly available
2. Reasonable API access, maybe not directly but they want their niche workflows to be supported
3. Clear and consistent moderation. Eliminate disruptive content while giving reasonable people a clear understanding of whether something will be removed before they post it.
Threads has made vague promises to these three but not yet delivered.
I for one welcome the decentralized internet that Elon is ushering in. Things were better when we had small disparate forums where people could develop their perspectives in community. As long as people are members of multiple communities we should have enough cross-pollination and course corrections to keep things mostly on track.
Mods and groupthink are always a threat but it turned out they were a bigger threat still when everyone was trying to be on one social network to rule them all.
This is sad indeed, but it was already infested with low quality trolls before the Elon Musk purchase. The entire Jonathon Scott clown-fiesta was utterly embarrassing.
I suspect your reasoning is also motivated, though I don’t know by what.
There are things I’ll give credit for. I suspect the “community notes” feature is overall beneficial (though I haven’t seen it much firsthand because I use a very outdated version of the application, which doesn’t support it.). Also, making the code for the recommendation algorithm publicly readable was very cool, and I appreciate it.
The way that verification was replaced with the current system, was, I think, rather poorly designed. That’s not to say that the critiques of the previous system were illegitimate. But the way the change was done threw the baby out with the bath water. Ideally, the system should work as follows: paying-users and associated features, should be independent of verification. Anyone should be able to pay a one-time (unless they change display name too much) fee to have their account verified, where this verification should be an actual verification, not just “did this person pay money”. (This might have to cost a fair chunk in order to be viable, due to labor costs of verifying, but I think it could still be reasonably cheap..)
There could be an option for paying subscribers to have some icon display by their name, like the checkmark does currently, but this should be optional, and should be distinct from the verification icon.
Also, there’s all the stupid stuff, like blocking links to competitors, and seemingly not having a competent rollback system for updates when they break stuff, etc.
>Also, making the code for the recommendation algorithm publicly readable was very cool, and I appreciate it.
The code for the recommendation algorithm WAS NOT made publicly available, but rather just the scaffolding for interacting with the still black box ML model. There was zero informational value there (well there was a few tags that were interesting, including one intended to measure sentiment against musk's tweets specifically) and it looked familiar to anyone who has done ML in python.
Basically it's like he posted the source code for the CRM, but not the database.
It’s not a Nazi dominated hellhole like everyone exaggerating is saying it is. It’s the exact same thing as it was. you can hate Elon Musk without automatically hating everything he owns because there’s a lot more people that work there than just him it’s pretty shortsighted to condemn all those people for the actions of one billionaire
Ok, sure, I agree with the part of that first statement before the "like everyone exaggerating is saying it is". (I'm not contradicting the "like everyone [...] it is", just not endorsing it.)
I don't hate Elon Musk, and I have a generally positive (but not particularly well-informed) opinion of SpaceX and Tesla (though I'm skeptical of "Boring Company", though I'm also not informed about that one either.).
I'm just of the opinion that he made quite a few errors in running twitter (but also some upsides compared to previous status quo, not all downsides).
translation: al-qaeda had a twitter account before the mediocre-man with a middle-life crisis accidentally bought it. Now the nazis have accounts there.
The Saudis were already an owner of Twitter before Elon Musk. In fact they were the largest.
But if you need another reason here’s a dead simple one. All that Saudi money invested in Twitter and musk, at least isn’t going somewhere else to fuel more nefarious things.
I think everyone in the (us) could serve to take a breath, and turn things down and notch or two. We all care about our country and want the best things for it. We just disagree about how to get there and I'm willing to de-escalate when I can in hopes that reciprocity will follow and civility can regain a foothold
this is a throwaway and will remain so, but before the troubles people were already complaining about twitter being infested with nazis, it's just that then whatever had been done to get rid of them was undone, thus my comment
Hard for me to know. I used to browse some people anonymously but can't do that anymore. Not making an account to read some physics professor's tweets...
