Todays playlist is long songs. Everybody
If people actually let you get away with it... I mean, I expect some fundamentalists on Mastodon to demand a full image description in alt-text, no matter how, no matter what, full stop.
I almost always put my image descriptions into the post text body for two reasons. One, they contain explanations and other information neither available in the actual post text nor for sighted people in the image itself. Two, they're way too long. They exceed 1,500 characters multiple times, sometimes multiple dozen times. I don't have local character limits to worry about, but at least Mastodon, Misskey and all their forks chop alt-text over 1,500 characters off at the 1,500-character mark.
In addition, I always mention in alt-text where exactly this image description can be found. Not only in the post text body, but I add stuff like "above the image and hidden behind a content warning if you're viewing this post on Mastodon and right below this image if you're viewing this post on Friendica, Hubzilla or (streams)".
More recently, however, I started adding a short description of the image to the alt-text. It actually doesn't even begin to describe the actual contents of the image. It's only there to satisfy the "alt-text must describe the image or else" crowd. I may have alt-texts well beyond the 850-characters mark nonetheless.
And still, I think the only reason why nobody complains about the lack of image description in my alt-texts is because too many Mastodon users or entire instances have muted or blocked me for not acting "Mastodon-like" enough.
#Long #LongPost #CWLong #CWLongPost #FediMeta #FediverseMeta #CWFediMeta #CWFediverseMeta #AltText #ImageDescription #ImageDescriptions #ImageDescriptionMeta #CWImageDescriptionMeta Long Weil es zu fast allen Themen Leute gibt, die sich darber aufregen, da die nicht hinter einer CW versteckt werden. Ja, auch Katzenbilder (CW: cats).
Deswegen bin ich dafr, da Mastodon mal den umgekehrten Weg geht. Zur Abwechslung mal nicht dem ganzen Fediverse seine eigene Kultur aufzwingt, die geprgt ist durch technisch ungerechtfertigte Beschrnkungen, Neuerfinden von Rdern und Walled-Garden-Denken. Statt dessen mal was nicht nur technisch, sondern vor allem in seine Kultur bernimmt, was im Fediverse vor Mastodon der Standard war:
Content Warnings aus Filtern. Automatisch generiert fr jeden Leser individuell.
Das kann Mastodon auch, und zwar seit Version 4.0, also seit letztem Jahr. Nur nutzt das keiner, und zwar aus einer ganzen Reihe an Grnden:
Erschwerend kommt hinzu: Viele Projekte, die nach Mastodon als Alternativen zu Mastodon gestartet wurden, wurden nicht nur auf Mastodons Technik, sondern auch auf Mastodons Kultur zugeschnitten. Entweder haben die gar keine Filter, oder die haben bis heute keine Mglichkeit, per Filter CWs zu generieren.
#Long #LongPost #CWLong #CWLongPost #LangerPost #CWLangerPost #FediMeta #FediverseMeta #CWFediMeta #CWFediverseMeta #CW #CWs #ContentWarning #ContentWarnings #CWMeta #ContentWarningMeta Okay, here's a bit more. And I'm trying to write this from a Mastodon point of view, but as someone who has got experience outside of Mastodon.
If you're on Mastodon which you probably are, get used to the lingo.
This may come as a surprise, but the Fediverse is not only Mastodon. There are many other projects in the Fediverse which are connected to Mastodon. Some of them have actually already been around before Mastodon. And they have features that Mastodon doesn't have. So don't freak out if you see a post with over 500 characters. Or with italics. Or with bold type. Or with a
Double your content warnings with hashtags. Some projects in the Fediverse outside of Mastodon use text filters to automatically generate individual content warnings for those who need them. And Mastodon can do that, too, so even some Mastodon users do it.
If a post with content that you may deem sensitive has only got these hashtags but no actual content warning, this is not necessarily due to ill intent or neglect. Some Fediverse projects outside of Mastodon can't do content warnings the Mastodon way. Hubzilla can only do them in first posts, but not in replies. On Friendica, they're generally very difficult to do. So don't immediately call for moderation.
Actually describe your images in alt-text. And do only that. No hashtags in alt-text. No links in alt-text. No SEO keyword spamming in alt-text.
If there's text in your picture, transcribe it in alt-text word by word.
Don't forget: Different instances have different rules. So don't report someone on another instance for not adding a content warning that's mandatory on your instance but not on theirs or the like. And don't forget either: Different projects have different capabilities and different cultures of using them.
#Long #LongPost #CWLong #CWLongPost #FediMeta #FediverseMeta #CWFediMeta #CWFediverseMeta #Fedisplaining #CWFedisplaining #MastodonIsNotTheFediverse #NotOnlyMastodon #FediverseIsNotMastodon #Hashtag #Hashtags #AltText
Da gibt es leider keinen wie auch immer gearteten Konsens, auch wenn einige behaupten, den gbe es. Genau wie mit der Lnge und Ausfhrlichkeit von Bildbeschreibungen.
Die einen Blinden wollen Bildbeschreibungen gern kurz und prgnant. Die anderen wollen sie ausfhrlicher. Z. B. lassen sie sich nicht damit abspeisen, da du Elemente in einem Bild erwhnst, sondern sie wollen auch wissen, wie das alles jeweils aussieht.
Wenn du jegliche Bildbeschreibung in den Alt-Text packst, bringt das fr Blinde noch einen weiteren Nachteil mit sich: Screenreader knnen Alt-Text nicht navigieren. Das heit, die knnen nicht innerhalb des Alt-Text an eine bestimmte Stelle zurckspringen, sondern nur wieder ganz zum Anfang, um dann alles wieder von vorne durchzurattern.
Das ist einer von diversen Grnden, warum ich z. B. meine ausfhrlichen Bildbeschreibungen immer in den Post selbst packe. Ein anderer ist, weil sie meistens weit lnger als 1500 Zeichen werden, aber zumindest Mastodon, Misskey und die Forkeys schneiden Alt-Text an der 1500-Zeichen-Marke hart ab.
Allerdings poste ich im allgemeinen Bilder zu einem sehr obskuren Nischenthema. Die meisten Leute brauchen sehr, sehr viel Erklrung, um die Bilder zu verstehen. Und gerade meine eigenen Bilder sind manchmal ziemlich komplex, auch wenn ich versuche, das zu vermeiden.
Wie gut meine Beschreibungen ankommen, wei ich nicht. Ich habe bisher mehr Bilder beschrieben als Feedback dafr bekommen, und zwar inklusive Feedback von Leuten, denen ich ausdrcklich meine Bilderposts als Beispiele fr lange, detaillierte Bildbeschreibungen gezeigt habe. Ungefragtes Feedback hatte ich bisher nur einmal, auch wenn das positiv war.
Nicht jeder, der blind ist, wurde blind geboren. Wer irgendwann in seinem Leben erblindet ist, hat frher mal Farben gesehen, also auch einen Begriff von Farben.
Auerdem ntzen Bildbeschreibungen nicht nur Blinden, sondern auch Sehenden, die mieses Internet haben, wo die Bilder berhaupt nicht laden.
Ich wrde Farben immer erwhnen.
Die meisten, die es gibt, sind nur fr professionelle/kommerzielle statische Websites oder vielleicht noch professionelle/kommerzielle klassische HTML-Blogs. Die meisten, die Social Media behandeln, sind nur fr kommerzielle Social Media, also Facebook, Twitter, Instagram, LinkedIn usw. Und die paar, die tatschlich das Fediverse behandeln, sind nur fr ganz normales Vanilla-Mastodon und gehen felsenfest davon aus, da Posts nicht lnger sein knnen als 500 Zeichen.
Die Sonderflle "Zeichenlimit deutlich ber 500 Zeichen" und "Social-Media-Post mit Bild, das viel Beschreibung und/oder Erklrung braucht", sind noch nie in so einer Anleitung abgedeckt worden.
CC:
#Long #LongPost #CWLong #CWLongPost #LangerPost #CWLangerPost #AltText #Bildbeschreibung #Bildbeschreibungen #BiBesch "MIA" means "missing in action".
I don't know all the details, just so much: The last time Kainoa was seen contributing to the Firefish repository was in June or July. Around that time, I guess, was also the last time that Kainoa was seen taking care of firefish.social. Kainoa is still alive and well and has posted something not long ago. Could be busy, nobody knows.
Anyways, fast-forward to December, 2023/January, 2024. Due to literally no maintenance whatsoever because the only admin hasn't shown up in a while, firefish.social has become entirely unresponsive. Nothing works anymore.
Also, Firefish is badly lagging behind Misskey's development. Not only no new features, but also no bug fixes, no security fixes, no nothing.
So the Firefish community took matters into their own hands. Just to mention three new forks:
#Long #LongPost #CWLong #CWLongPost #FediMeta #FediverseMeta #CWFediMeta #CWFediverseMeta #Misskey #Forkey #Forkeys #Firefish #Iceshrimp #Sharkey #Catodon
Web interfaces could get switches like these, too, and ideally, admins could add their own switches, depending on the target audience of an instance. Ideally, apps could recognise these custom, instance-specific switches and add them to their UI.
On the posters' side, it'd still require discipline because they'd have to add appropriate hashtags, and these hashtags would have to be standardised, otherwise each filter would have to cover each trigger with half a dozen different hashtags per language or so. Dedicated apps could assist in this by adding checkboxes to the post editor which have the app automatically add appropriate standard hashtags for certain topics to the "hashtag line" at the end.
This could actually start in apps without breaking standards, and it could nudge Mastodon itself into adding it to its Web UI. In fact, I can see at least some Forkeys being quicker at introducing it.
The best part: This would move the Fediverse away from something Mastodon-specific and towards something more standard.
#Long #LongPost #CWLong #CWLongPost #FediMeta #FediverseMeta #CWFediMeta #CWFediverseMeta #CW #CWs #ContentWarning #ContentWarnings #CWMeta #ContentWarningMeta
Even if Mastodon won't introduce the creation of quote-posts, Misskey, the various Forkeys (Firefish, Iceshrimp, Sharkey, Catodon etc. etc.), Friendica, Hubzilla and (streams) can quote-post just about any Mastodon toot out there with no resistance whatsoever. They always could, even when Mastodon wasn't able to display quote-posts halfway appropriately.
#Long #LongPost #CWLong #CWLongPost #FediMeta #FediverseMeta #CWFediMeta #CWFediverseMeta To be fair, this doesn't apply to all the Fediverse. Mastodon, yes, but not the entire Fediverse.
I dare say there are two projects which handle this marvellously: from 2015 and a fork of a fork... by Hubzilla's own creator, commonly referred to as while technically being nameless. While both can "talk ActivityPub", they are based on different protocols which give them a feature that Bluesky tries hard to make everyone believe that they're working on inventing: .
First of all, Hubzilla and (streams) don't put your Fediverse identity directly into your account. Your account only grants you access. Your identity, your connections, your posts, your settings, your content etc. etc., all that resides in a , a kind of container. You can have multiple channels on one account, so you can also have fully separate identites on the same instance without having to log out and back in.
Nomadic identity itself can do two things. One, it takes care of moving a channel to another instance. And I mean actually moving it, not creating a dumb copy and leaving a dead account behind.
Two, and this is the actual killer feature, it can clone your channel to another instance. Again, not create a dumb copy. The clone is always synchronised with the original and vice versa. If one goes down, you've got the other readily at hand, so you're even safe from spontaneous shutdowns. And you can have more than one clone for each one of your channels.
Basically, moving is cloning light. Moving creates a synchronised clone, declares the clone the new main and the old main a clone, then deletes the old-main-gone-clone, and if there aren't any other channels on your old account, deletes that account.
Nomadic identity was invented in 2011 by who also created Hubzilla, now maintains (streams) and, in 2010, had created which was named Friendika back in 2011. Public Friendika nodes closing down had become a serious issue already, regardless of whether the shutdown was announced or came out of the blue. Early on, whenever this happened, all users on the node lost everything and had to start over from scratch.
Friendika eventually introduced moving from node to node by exporting and importing your account, but real resilience would have been different. Thus, nomadic identity was conceived. The Zot protocol was written around it because it couldn't be introduced to Friendika's DFRN protocol. In 2012, Friendica was forked and its backend re-written against Zot, becoming the Red Matrix, direct precursor of 2015's Hubzilla.
In a perfect Fediverse, everything would feature nomadic identity. But as it stands now, everything that isn't these two projects doesn't even understand nomadic identity and is easily confused by nomadic moves or someone posting from their clone instead of their main.
Unfortunately, nothing will change about this anytime soon. Mike Macgirvin has already proposed extensions for ActivityPub that'd introduce nomadic identity, albeit not in exactly the same fashion as on Hubzilla and (streams), but probably somewhat compatible. However, they have all been rejected.
#Long #LongPost #CWLong #CWLongPost #FediMeta #FediverseMeta #CWFediMeta #CWFediverseMeta #Hubzilla #Streams #(streams) #NomadicIdentity This might be a case of Chinese whispers.
What really increases accessibility when it comes to hashtags is how they're written if they consist of multiple words.
Always write multiple-word hashtags in "camelCase" (every word after the first one starts with a capital letter) or, better yet, "PascalCase" (every word starts with a capital letter). Screen readers used by blind or visually impaired users can't tell with certainty otherwise where a word ends and the next one starts.
Also, stop using Mastodon like Twitter and put all hashtags at the end of the post instead of into the text.
#Long #LongPost #CWLong #CWLongPost #Hashtag #Hashtags #Accessibility #A11y #camelCase #PascalCase
What do you mean would cause an instant ban The use of quote-posts for harassment or the use of quote-posts in general
I don't know if you've read the whole thread. I've already posted in it. If not: Lots of Fediverse projects have quote-posts regularly implemented. Two of them, Friendica and Hubzilla, have had them before Mastodon even existed, and it was Mastodon which federated itself with them and not the other way around. They call them "shared posts" and use them for the same purpose as Mastodon uses boosts: to forward posts.
They also both had quotes like the one above since their respective inception.
CC:
#Long #LongPost #CWLong #CWLongPost #FediMeta #FediverseMeta #CWFediMeta #CWFediverseMeta Have there ever been any specific promises how ease-of-use could be improved
I can't see anything that could be done without cutting features and betraying the Fediverse's principles except everything from Mastodon to Hajkey mimicking Twitter's default immediately-pre-Musk UI/UX. And maybe railroading newbies even harder by not even telling them that Mastodon is decentralised and taking instance selection away entirely and an AI that increases the mollycoddling by hiding Mastodon's decentrality and all contents coming in from the rest of the Fediverse until it decides the user is "ready".
CC:
#Long #LongPost #CWLong #CWLongPost #FediMeta #FediverseMeta #CWFediMeta #CWFediverseMeta Well, is writing about static websites and long-form blogs and not about the Fediverse, especially not about Mastodon.
On Mastodon very specifically, everything that describes and/or explains an image always goes into the alt-text. Unlike all other *blogging projects in the Fediverse, Mastodon with its tiny character limit doesn't leave you much of a choice. So people got used to it and see it as a set-in-stone Fediverse standard.
#Long #LongPost #CWLong #CWLongPost #FediMeta #FediverseMeta #CWFediMeta #CWFediverseMeta #AltText #ImageDescription #ImageDescriptions #Caption #Captions #ImageDescriptionMeta #CWImageDescriptionMeta Man darf sich aber nie der Illusion hingeben, da KI jedes, aber auch jedes Bild auf Anhieb perfekt akkurat und informativ beschreiben kann.
Ich will's mal so sagen: Einfache Katzenbilder bekommt KI beschrieben. Fr reine Text-Screenshots braucht es keine KI, da reicht OCR.
Aber wenn die Bilder a) hochkomplex sind, b) extrem nischiges Unbekanntes zeigen und c) gerade deshalb sehr detailliert beschrieben und umfassend erklrt werden mssen, damit das Publikum sie versteht, und wenn an die Beschreibungen hohe Ansprche gestellt werden, das wird KI nie so gut hinbekommen wie ein entsprechend kompetenter Mensch.
Generell sollte man KI-Beschreibungen nie aus Bequemlichkeit ungesehen copy-pasten, sondern immer gucken, ob das Geschriebene so stimmt, und gegebenenfalls hndisch nachbessern.
#Long #LongPost #CWLong #CWLongPost #LangerPost #CWLangerPost #KI #AltText #Bildbeschreibung #Bildbeschreibungen So you're finally no longer allowed to reply to a post that was written neither by a follower nor in response to a follower
Or, if you're on Friendica, Hubzilla or (streams), to a post that wasn't started by one of your followers because these three don't serve you single comments from within threads which didn't appear on your stream from the very beginning
Well, that makes showing Mastodon users the Fediverse outside of Mastodon much more difficult. Or that Mastodon's "rules" for how and where to describe images are not necessarily set in stone for the whole Fediverse because everything that isn't Mastodon doesn't have Mastodon's limitations.
Or that the first time the term "Metaverse" was used for an actual, real 3-D virtual world was not in 2021 by Zuckerberg for something he planned, but in 2007 for something that already existed and still exists.
I'd wonder if you're already a reply guy if you reply to a post in your local or federated timeline. But I don't have either.
CC:
#Long #LongPost #CWLong #CWLongPostBisher wei man wenig ber Long Covid bei Kindern und ob impfen dagegen hilft. Daten von ber einer Million Kindern und Jugendlichen zeigen nun moderaten Schutz durch die Impfung.
"The Metaverse" as in new 3-D virtual world projects of the 2020s in general
Or "The Metaverse" as in...
...as in "Zuckerberg's Metaverse" as in "Facebook's Metaverse" as in "Meta's Metaverse" as in Horizons
Can everyone please with a cherry on top stop referring to Meta's virtual worlds as "the Metaverse"
They're called "Horizons"!
Zuckerberg did not invent the Metaverse. Zuckerberg did not even invent the term Neal Stephenson did in 1991. A Second Life in-world expo in the year 2007 was called "Metaverse". And the OpenSim community has been using the word "Metaverse" on a daily base down to grid names before 2010, too.
Thanks from someone who has been in OpenSim and thus known and used the term "Metaverse" in 2020 already in behalf of everyone in Second Life, OpenSim and all other virtual worlds that already existed before 2022.
P.S.: Second Life did not shut down in 2008 nor in 2009 either.
#Meta #Horizons #HorizonWorlds #HorizonWorldsIsNotTheMetaverse #Metaverse #TheMetaverse #SecondLife #OpenSim #OpenSimulator #VirtualWorlds #Long #LongPost #CWLong #CWLongPost
Oh, and by the way, a finding of mine for those who still think Zuck was the first to slap "Metaverse" on an actual real 3-D world:
Something for those who think Mark Zuckerberg has invented the term "Metaverse" in 2021.
, the oldest grid (est. 2007), is referring to itself as "The Open Source Metaverse". Guess what It has taken over that slogan from OpenSim itself, and .
Metropolis, the first German grid (est. 2008), boldly used the full name "Metropolis Metaversum" until it was officially shut down in summer 2022. .
Since I started out in Metropolis on April 30th, 2020, I was in something called Metaverse almost a year and a half before Zuckerberg used that term.
Another fun fact: The virtual worlds/virtual reality news site has named its OpenSim category "" as early as 2010.
#OpenSim #OpenSimulator #Metaverse #VirtualWorlds #Metropolis #OSgrid
For the record: Yes, I know it was Neal Stephenson who coined that term in 1991 already.
#Meta #Horizons #HorizonWorlds #HorizonWorldsIsNotTheMetaverse #Metaverse #TheMetaverse #SecondLife #OpenSim #OpenSimulator #VirtualWorlds #Long #LongPost #CWLong #CWLongPost
Part of the secondary topic of this channel is showing a very-much-non-Mastodon view on the Fediverse to the Fedizens out there. Lots of fresh Twitter-to-Mastodon converts started following me in late 2022 and early 2023 because I was one of the few to explain the Fediverse outside Mastodon to Mastodon users.
This doesn't work anymore if my posts don't reach their recipients anymore.
And besides, triggering Mastodon users with non-Mastodon behaviour can have nasty side-effects. Yeah, some mute you. Some block you. Often even without also unfollowing you. Some complain to you about it and demand you don't use any features of your instance that vanilla Mastodon 3.x didn't support yet.
But then there are those who try to complain to your instance admin about you and try to have you sanctioned. Or they have their instance admins/mods complain to your instance admin about you. And this can escalate.
A Hubzilla admin would laugh that off. Most Hubzilla hubs don't even have rules, and the rules of any one Mastodon instance don't apply here either.
But in most cases, the complaints never reach their recipients. See, the average Mastodon user doesn't know anything about Hubzilla. So they try to complain the same ways as they would within Mastodon.
They use Mastodon's report feature, but Hubzilla doesn't have it implemented. They try to send your admin a DM, but finding the admin of your hub doesn't work like finding the admin of a Mastodon instance, and oftentimes, the admin's channel isn't linked anywhere at all. They publicly complain about you with a Mastodon admin hashtag, hoping that your admin will read it. But you can't follow hashtags on Hubzilla, and even if you could, why would a Hubzilla admin follow a Mastodon-specific hashtag
So the admin and the "moderation" of your hub doesn't react. Conclusion for a Mastodon user who only knows Mastodon: Your instance is basically unmoderated. And what does Mastodon do with an unmoderated instance whose users misbehave for any definition of that word Block it. Like, the whole instance.
One Hubzilla user misbehaving in the eyes of Mastodon users can cause the whole hub to be blocked on thousands of Mastodon instances, maybe including mastodon.social.
Now you could say I should flip the ActivityPub side of the Fediverse the bird and turn Pubcrawl off for good. If I did that, I could just as well delete the whole channel because it'd lose its very purpose.
#Long #LongPost #CWLong #CWLongPost #FediMeta #FediverseMeta #CWFediMeta #CWFediverseMeta The biggest cultural difference based on server software has to be between Mastodon on the one hand and the Mike Macgirvin creations Friendica, Hubzilla and (streams) on the other hand.
Not only do they have vastly different user bases, but they developed independently from one another. When the Mastodon culture developed, those who shaped it didn't even know the other side, so they couldn't adopt any of its culture.
Said other side's culture dates back to 2010 when Friendica was launched as Mistpark, and since that was almost six years before Mastodon, it couldn't be inspired by Mastodon's culture either.
Add to that that these respective cultures are greatly shaped by technical features and limitations or the lack thereof.
Mastodon's culture is largely built around its 500-character limit which is ample for your typical phone-wielding Mastodon user. Friendica, Hubzilla and (streams) don't have any defined character limit whatsoever, and its target audience is largely on desktop or laptop computers, often running Linux, with large screens and full-size hardware keyboards.
So it's the most normal thing in the world for them to write in one post as much as they want while Mastodon users debate whether threads are good, or you should always limit yourself to 500 characters or less.
Also, alt-text. Mastodon has many disabled users, including blind or visually-impaired users. And it has a dedicated alt-text field for each image. On top of that, it offers 1,500 characters for each alt-text which, in connection with the 500-character limit for toots, has people write detailed image descriptions and explanations and put them into the alt-text. That's often information that doesn't even belong in alt-text, but there's no room for it elsewhere.
Friendica, Hubzilla and (streams) have unlimited room, so putting stuff into alt-text because the post text is too limited seems ridiculous. But they don't have a vocal disabled community, so there's little interest in accessibility. And neither of them has a dedicated alt-text field. Alt-text is supported, but it has to be manually grafted into the image-embedding code in the post. And there's no official documentation for that, I think not even for Friendica which is the only one out of the three with actually useful end-user documentation.
It's similar with content warnings. On Mastodon, they're put into the repurposed summary field, and next to nobody knows that it's a repurposed summary field rather than invented for content warnings from scratch. So since Mastodon has a content warning field, writer-side content warnings are huge, but also cause for drama.
Mastodon 4.0 has introduced filters that can create reader-side content warnings, but hardly anyone uses them, even fewer people support them with keywords or hashtags, many don't even know this feature exists, and it's generally ignored because it's un-Mastodon.
The Friendica/Hubzilla/(streams) complex doesn't have a content warning field. Hubzilla and (streams) have a summary field labelled as such. Friendica doesn't even have that it uses a pair of BBcode tags for that.
And within their own ecosystem, they don't even need it. They've got the "NSFW app" instead, an over-one-decade-old, optional, simple-as-anything substring filter that automatically hides entire posts with all media and everything behind content warnings if it finds one of the entered keywords or hashtags.
So they can't understand Mastodon's commotion about content warnings, and Mastodon users can't understand why they don't add Mastodon-style content warnings.
And then there are all the things that were or are being debated on Mastodon and whether or not it should introduce them. Especially the second-wave Twitter refugees are often staunchly against them.
Full-text search and quote-tweets are being actively used on Twitter to track down and harass members of minorities who have fled to Mastodon. Of course, they don't want Mastodon to introduce either. That is, in the case of full-text search, Mastodon has found a solution, but one that doesn't really federate to the rest of the Fediverse.
Quotes and text formatting are seen as bad, too. Many don't know quotes because Twitter doesn't have them, and so they think quotes could be used as tools of harassment. And both are seen as making Mastodon feel less like Mastodon and more complicated.
Friendica and Hubzilla had all this before there was Mastodon, and (streams) inherited it from its long line of ancestors. Their users have gotten so used to having all this that they don't understand what the problems should be, also because they're so detached from Mastodon's culture. So they keep on using these features unashamedly, even around Mastodon users.
Some differences are rather simple. Take mentions, for example. Friendica has always used long names for mentions, as does its offspring. Mastodon users may find that freaky. Meanwhile, Friendica, Hubzilla or (streams) users may find Mastodon's mentions cryptic because they use the short name. But even they matter.
#Long #LongPost #CWLong #CWLongPost #FediMeta #FediverseMeta #CWFediMeta #CWFediverseMeta #FediverseCulture #Fediverse #Mastodon #Friendica #Hubzilla #Streams #(streams) #CharacterLimit #CharacterLimits #500Characters #CW #CWs #ContentWarning #ContentWarnings #ContentWarningMeta #AltText #ImageDescription #ImageDescriptions #ImageDescriptionMeta #CWImageDescriptionMeta #Quotes #QuotePost #QuotePosts #QuoteBoost #QuoteBoost #QuoteTweet #QuoteTweets #QuoteToot #QuoteToots #QuotedShares #TextFormatting #FullTextSearchScott M. Stolz I'm not talking about the occasional, moderate use of stuff that Hubzilla can do that Mastodon can't.
I'm not even only talking about using such stuff all the time and not in moderate amounts.
I'm talking about things like "content warnings" that read, "Summary of the post CW: long (number of characters), stuff-I-think-not-everyone-wants-to-read-about meta, more-stuff-I-think-not-everyone-wants-to-read-about meta." And then forgoing content warnings altogether in replies.
Or having eight and more hashtags at the bottom of a post, some of which look totally weird because they're the same as another one of the hashtags, but with "CW" at the beginning. (I use them to trigger filters. See below.)
Or image posts with alt-text for the images, but the alt-text wastes over 800 characters barely even describing the image and mentioning that the actual detailed image description is in the post itself. WTF, nobody puts image descriptions in the post, they belong into alt-text and nowhere else!
Worse yet: The post is hidden behind a content warning. And the content warning says something about tens of thousands of characters. And then you click/tap the content warning. And then an utter wall of text jumps at you, way longer than 100 toots could ever be, all in one enormous chunk.
Last but not least, my constant jabs at Mastodon, Mastodon users, especially Twitter refugees, Mastodon culture...
#Long #LongPost #CWLong #CWLongPost #FediMeta #FediverseMeta #CWFediMeta #CWFediverseMeta You can't "quote-toot" tweets on Mastodon. In fact, you can't "quote-toot" toots on Mastodon. You can't "quote-toot" anything on Mastodon.
So unless you're on Misskey or one of its forks or on Friendica or one of its descendants, or unless you want to quote something outside the Fediverse, screenshots are the way to go.
That's also because next to everyone on Mastodon is on a phone. And copy-pasting text from e.g. the Twitter app to whatever Mastodon app you use and copy-pasting the URL of e.g. the tweet into that Mastodon app is very inconvenient because copy-paste is a pain on a phone with no separate pointing device and no hardware keyboard.
Add to this that Mastodon can't only not "quote-toot", it can't quote either.
#Long #LongPost #CWLong #CWLongPost #FediMeta #FediverseMeta #CWFediMeta #CWFediverseMeta
Use as many as you need. But don't abuse them. For example, there are people who slap #Mastodon on posts that aren't even about Mastodon or #Fediverse on posts that aren't about the Fediverse in general, just to try and increase their range. This is bad behaviour, this is what's counterproductive, and it's actually often being sanctioned.
Still, there shouldn't be an upper limit. Some Fediverse users have filters that are triggered by hashtags and either remove posts from their timelines or put them behind automatically generated reader-side content warnings.
On Friendica, Hubzilla and (streams), this is actually the default, and it has been since even before Mastodon existed, and neither of them has a dedicated "content warning" text field. Also, at least on the latter two projects, there is no means whatsoever to add a Mastodon-style content warning to a comment like this.
Thus, at least some users, especially from these three projects (although (streams) isn't a project), add hashtags for the purpose of triggering filters. As you'll see at the bottom of this comment, it'll cause a whole lot of hashtags for the standards of someone who's used to Twitter. But they're deeply ingrained in the culture of certain non-Mastodon projects. And they're useful for at least some Mastodon users.
So any reply you may get that tells you that nobody in the whole Fediverse can possibly want more than four or eight hashtags is simply false. Some don't want more than that. Others depend on them.
#Long #LongPost #CWLong #CWLongPost #FediMeta #FediverseMeta #CWFediMeta #CWFediverseMeta #Hashtag #Hashtags #HashtagMeta #CWHashtagMeta Okay, this one is going to be long. Don't let the length and the text formatting disturb you. That's how we do things over here on Hubzilla, and I'm glad Mastodon can at least display them now.
So let's see...
First of all, it's unusual to have multiple description of an image in the alt-text. I'd say it's redundant, but I myself give short descriptions in the alt-text and then much longer ones than yours in the post text body which the alt-text also references.
Pick only one. For technically speaking, alt-text is not a synonym for a short description. Alt-text is the whole thing you create when pasting the AI output into that text field for images. It's what stands in for the image when you can't see the image for whichever reason, and it's all of it.
And don't add "alt-text" to alt-text.
Next, please excuse the pun, but it looks like you put blind faith into AI, and you copy-paste its output into your alt-texts unread, not to mention unredacted.
In the description of the fourth image you've linked, it really goes to show that whatever AI you use (ChatGPT for detailed image descriptions) doesn't even recognise what's in the lower picture. It doesn't get the picture, so it doesn't get the meme, so it fails to explain the meme.
I would definitely have added the information that the lower picture shows Christopher Lloyd as Doctor Emmett L. Brown in the final showdown scene of the 1985 film Back To The Future, trying to reconnect the lightning conductor cable which is about to channel the lightning bolt that shall damage the clock on Hill Valley's town hall into the DeLorean time machine with Marty McFly inside and use the energy to send him back into his time.
That is, in practice, I would have chosen a different wording and added another bit more of information. But without it, the image description is half-useless because it doesn't mention what the meme references and hints at.
Yes, it's easier and more convenient to believe in perfect fire-and-forget recognition, identification, description and explanation of any image imaginable by AI. But if you really want your image descriptions to be informative and, above all, correct, read them. And redact them. Or, better yet, write them yourself.
#Long #LongPost #CWLong #CWLongPost #FediMeta #FediverseMeta #CWFediMeta #CWFediverseMeta #AltText #ImageDescription #ImageDescriptions #ImageDescriptionMeta #CWImageDescriptionMeta #AIScott M. Stolz
Good luck getting Mastodon to implement any standard they haven't invented themselves.
This will end in
This won't change much about many public, open-registration Friendica/Hubzilla instances not providing a link to the account/channel of the admin. Or any admin contact at all. All because the admin didn't bother to edit that page/hasn't gotten to it yet.
And it changes nothing about Mastodon users/admins trying to reach Friendica/Hubzilla/(streams) admins the exact same way as they always try to reach other Mastodon admins. And failing because it doesn't work like on Mastodon. Even less in a mobile app that's geared towards Mastodon and Mastodon only.
Don't forget: On Hubzilla, everyone's on a Linux PC using a standard browser. On Mastodon, everyone's on a phone using a Mastodon app.
#Long #LongPost #CWLong #CWLongPost #FediMeta #FediverseMeta #CWFediMeta #CWFediverseMeta If done the Friendica/Hubzilla/(streams) way, that'd be disturbing to Mastodon users again.
I think nobody on Mastodon knows this, but these three projects have a Facebook-like/Tumblr-like/blog-like one-post-many-comments thread structure as opposed to Mastodon's Twitter-like many-single-posts-tied-together thread structure.
This means the thread starter owns the whole thread. This also means the thread starter can moderate their own thread and even delete comments from their own thread. Even if they weren't appointed moderators by the admin. AFAIK, the deletion is actually forwarded to the source.
So a Mastodon user may e.g. troll around in a thread started on Hubzilla or attack the thread starter. A bit later, that Mastodon user will discover that their toots in reply to that Hubzilla posts are gone. Maybe the thread starter even mentions they've moderated their own thread and deleted comments.
The short form: Regular users of Friendica, Hubzilla and (streams) can delete Mastodon toots. In fact, posts from anywhere. Only toots/posts in reply to their own posts, but yes, they can. And they could before Mastodon even existed.
If this started happening often enough, and word started spreading around especially Mastodon, it'd be perceived as invasive and disruptive by many. It might cause such a big uproar that even more Mastodon users would call for defederating Friendica, Hubzilla and (streams) completely.
As for actual, traditional moderation, I'm not saying it's bad per se.
I'm saying that many Mastodon admins really don't know anything about the Fediverse outside Mastodon.
They think everything else that does *blogging is basically Mastodon with a different UI. Like an alternative mobile app for Mastodon itself. Or it's Mastodon + feature x on top. But otherwise like Mastodon.
And then they act accordingly. They try to interact with Friendica, Hubzilla and (streams) admins the same way as they interact with other Mastodon admins. And they neither know nor realise that these three projects are too different from Mastodon for these methods to even work!
Again:
You can't report a Hubzilla user to a Hubzilla admin using the report system because Hubzilla does not have Mastodon's report system!
You can't expect Hubzilla admins to follow hashtags for Mastodon instance admins either. Not only because they aren't Mastodon admins, and they probably don't even know that hashtag. But because you can't follow hashtags on Hubzilla!
And so forth.
So what appears to be a Hubzilla admin's laziness or carelessness or ignorance are actually technical differences between Mastodon and Hubzilla that Mastodon admins are unaware of.
This has nothing to with moderation by admins/mods only being bad. This has everything to do with Hubzilla being vastly different from Mastodon. And with Mastodon admins/mods not knowing jack shit about Hubzilla.
(Good thing this is so long that nobody will read it. Otherwise it'll empower those who say that Friendica, Hubzilla and (streams) must be redesigned to be exactly like Mastodon or face full defederation otherwise for being too different.)
#Long #LongPost #CWLong #CWLongPost #FediMeta #FediverseMeta #CWFediMeta #CWFediverseMeta gave me to think.
Could it be that countless Friendica, Hubzilla and (streams) users are blocked on countless mostly Mastodon instances by the admins because reporting users the Mastodon doesn't work on these projects
So there's a user who doesn't fully act according to the Mastodon community standards. That user's posts appear on some Mastodon instance.
The wrongdoing: For example, what's perceived as hashtag abuse see the linked thread. Or no Mastodon-style content warning where Mastodon culture would demand one*. Or something like that.
What does the admin do Use the report system to report that user to the admins and moderators of their own home instance.
Problem: That particular user isn't on Mastodon. Not on anything that was modelled after Mastodon either. That user is on Friendica or Hubzilla or (streams). Correct me if I'm wrong, but AFAIK, neither has Mastodon's report system implemented.
The report never reaches the admin of that instance. And the instance doesn't have any more staff.
Well, then they could write directly to the admin of that instance. If only the Fediverse contact of the instance admin was available on the instance frontpage. Or anywhere on the instance Web interface.
Even if they could, they might get the idea that they could catch the admin's attention by mentioning them in a public post. Spoiler: Doesn't work with Friendica accounts, Hubzilla channels and (streams) channels.
Oh, and at least Hubzilla and (streams) allow you to restrict from whom you receive direct messages. Regardless of whether or not that's a good idea, it's possible to make it so that DMs from random Mastodon users no longer end up in your stream. Worse yet: These Mastodon users don't even know that their DMs don't reach the recipient.
Okay, last resort, complaints about that user can be posted publicly under the hashtag #MastoAdmin. Should reach lots of admins, right
Yes, but almost exclusively Mastodon admins. It's MastoAdmin, after all. Why should an admin of, say, a Friendica node or a Hubzilla hub follow that hashtag Neither of them is Mastodon, and neither of them has anything to do with Mastodon. They didn't even federate with Mastodon, Mastodon federated with them.
Oh, and besides, to my best knowledge, they can't even follow hashtags in the first place. Or is Hubzilla the only one out of the three that doesn't have that feature yet
Anyways, the warning with the #MastoAdmin hashtag doesn't reach them either.
So whatever you try to let some Friendica or Hubzilla or (streams) admin know that a user on their instance "misbehaves", the admin doesn't react and "moderate" that user.
Conclusion for your typical Mastodon admin: That instance is unmoderated. From the point of view of people who only know Mastodon beyond the name, the admin must ignore all reports.
We can be glad if this leads only to blocking the "misbehaving" user on lots of Mastodon instances and not to what's standard for unmoderated or undermoderated instances on Mastodon: blocking the whole instance.
*Footnote: Neither of the three projects mentioned here has a "Content Warning" field. Hubzilla and (streams) have a "Summary" field which is the same thing, but especially newbies and those who are hardly in touch with the ActivityPub side of the Fediverse don't know it's the same. Also, that field is only available for posts (= first posts) and not for comments (= replies which are something entirely different on these projects). Friendica doesn't even have that a pair of BBcode tags is needed for a Mastodon-style content warning, and AFAIK, this isn't documented anywhere.
#Long #LongPost #CWLong #CWLongPost #FediMeta #FediverseMeta #CWFediMeta #CWFediverseMeta #BlockingMeta #BlocklistMeta #CWBlocklistMeta
But what if the offending user is on an instance
a) of something that is not Mastodon, that wasn't even inspired by Mastodon because it's older than Mastodon, and that doesn't have Mastodon's report system implemented
b) whose admin doesn't follow MastoAdmin because they aren't Mastodon admins
c) whose admin doesn't follow MastoAdmin because that project doesn't have following hashtags implemented
So whatever you do the Mastodon way never reaches the admin in the first place
#Long #LongPost #CWLong #CWLongPost #Defederation #FediMeta #FediverseMeta #CWFediMeta #CWFediverseMeta Just out of curiosity:
Let's suppose you come across someone whom you find misbehaving. You try reporting them to their instance moderation.
But they aren't on Mastodon. They're on Friendica or Hubzilla or (streams). And you can't for the life of you figure out how to get into contact with the admin of that specific instance via Mastodon. And yes, it's a public instance.
What do you do Defederate the whole instance because its management doesn't appear to be up to Mastodon's community standards Demand it be fediblocked altogether for that reason
Or what if there's only an e-mail address, but no link to a Fediverse account If that's a problem for Mastodon instance admins, I might single-handedly be responsible for hub.netzgemeinde.eu being defederated by dozens or hundreds of Mastodon and *key instances for "non-Mastodon-standard" behaviour.
(Hashtags below mostly for filter-triggering purposes. Text filters are the CW standard where I come from and have been since at least 2012.)
#Long #LongPost #CWLong #CWLongPost #Defederation #FediMeta #FediverseMeta #CWFediMeta #CWFediverseMeta Ich mache das anders. Ich warne im Summary (Mastodon: Content Warning) vor allem, was jemandem auf Mastodon unangenehm sein kann, und die eigentlichen Content Warnings dopple ich noch einmal mit jeweils einem und mehreren Hashtags fr die, die Filter verwenden.
Das ist auch ein bichen eine Demonstration dessen, was dabei rauskommen kann.
Zeichen habe ich selbst mehr als reichlich zu verbrennen, will sagen, praktisch unbegrenzt viele. Daran soll's nicht scheitern. Und wenn der Post lnger als 500 Zeichen wird, gibt's dafr eine Content Warning und die Hashtags #Long, #LongPost, #CWLong und #CWLongPost und bei deutschsprachigen Posts - wobei ich die Postsprache nicht einstellen kann - zustzlich #LangerPost und #CWLangerPost.
Bei Kommentaren gibt's dann nur die Hashtags, weil Mastodon-CWs nicht gehen.
CC: He may have a point, yes. There may be more people in the Fediverse who think like him, yes. He even sounds like he has come across one of my posts.
But if I did what he asked for, all my regular posts would grow to thousands upon thousands of characters and take hours to write because I'd have to break everything down to the very basics. Both the primary and the secondary topic of this channel are niche within a niche, maybe within yet another niche for a non-technical audience.
Essentially, I couldn't post anything anymore without long-post content warnings for Mastodon users who are disturbed by everything longer than 500 characters.
And my image descriptions for moderately complex in-world images, now sitting at 35,000 to 40,000 characters for the first image in a post and not counting the preamble, might easily grow beyond 50,000 or 60,000 characters. The preamble would grow well over 5,000 characters. Even longer if I took people's attention spans into consideration and explained certain things again and again.
This wouldn't even really help people outside my niche with understanding my posts. Even fewer of them would read them in the first place, and even more of them would mute or block me to get rid of those monstrous posts in their timelines.
And no, I can't explain the implications of the current scheduled OSgrid downtime in one simple sentence and in plain, everyday language to someone who doesn't even understand what a server is and knows exactly nothing about 3-D virtual worlds. Not in a way they'd fully understand.
Some of ya'll need to step back from the hyper-specific niche topics you post about and take a second to think about explaining your topics to normies, like ELI5 once in a while. Sometimes I'm scrolling my feed and I'm like wtf is all this this about.
Maybe I just curated a bunch of people with hard to follow niche interests.
Am I the only one
It seems like people aren't posting to have a conversation, It's just monologuing. I prolly fall into this once in a while as well.
#Long #LongPost #CWLong #CWLongPost
The latter is mostly because the vast majority of Lemmy users didn't come from Mastodon but from Reddit. They weren't told much about the existence of a Fediverse, only that there's a thing called Lemmy which is many copies of Reddit before its enshittification, and these are connected with each other. That's all that many know. Just like many Mastodon users think the Fediverse is only Mastodon.
It doesn't help that Lemmy barely gets any interaction from other projects. Mastodon is huge, and Mastodon users should be all over Lemmy. But many Mastodon users have never heard about Lemmy. Those who have may find it too inconvenient to follow a Lemmy community because that involves using the account search and copy-pasting. Don't forget that the huge majority of Mastodon users is on phones. And those who do manage to follow Lemmy communities say that the interaction between Mastodon and Lemmy is too limiting.
For the record: I do have Lemmy followers.
/kbin made bidirectional *blogging-style following possible only by bolting microblogging onto a Reddit clone. Lemmy is a more purist Reddit clone, it doesn't support domestic *blogging, so users of *blogging projects can't follow Lemmy users in the traditional sense.
As for lemmy.ml, that instance doesn't matter that much anymore. Even lemmy.world has been surpassed as the biggest instance.
Pixelfed could become big if Instagram was enshittified so tremendously that everyone except the biggest attention whores ("But muh followers, but muh fame") will start looking for alternatives. The advantage of Pixelfed for Instagram users over Mastodon for Twitter users is that Pixelfed allows direct imports of Instagram accounts with all content.
I'm not so sure about PeerTube, not only because that'd require gigantic amounts of hard drive space, but also because many users are on YouTube for the money, and PeerTube won't pay them a penny. If they moved to PeerTube, they'd lose a source of income. Also, fewer YouTube users have ever heard of PeerTube than users have heard of Mastodon.
BookWyrm would be easier, but I can't see right now how Goodreads could be enshittified enough to cause a mass migration. Maybe, however, BookWyrm becomes interesting for people who don't even know Goodreads and its whole concept, and they find out about BookWyrm before they find out about Goodreads.
Friendica tried to take a chance long ago, back in the early 2010s. It even tried to facilitate the transition of whole social circles from Facebook by federating with Facebook by means of a cross-poster. It didn't work out. People didn't want to leave their "friends" behind, not to mention that the average Facebook user was even less technologically adept than the average Twitter user a good decade later. Even trying to mimic Facebook's UI didn't help.
Maybe it was for the better. Typical server hardware that Friendica ran on back in the day could barely handle over 130 accounts on one node. The notoriously power-hungry Facebook connector cut a dozen or two out of this number. Many public Friendica nodes with the Facebook connector on closed their registrations at a bit over 100 accounts. I think not even the biggest root servers would have given you a four-digit capacity.
It would simply have been impossible to accommodate a flood of Facebook refugees on Friendica. Even if Friendica users with Facebook contacts had started their own private nodes, most of them would have needed multiple nodes to even have space for a fraction of their Facebook "friends".
As for Hubzilla, it'll first need a lot of polish. And then I can't see from where people would come flooding to Hubzilla. Facebook refugees would rather pile onto Friendica, the traditional more-powerful-than-Diaspora* Facebook "clone", or maybe (streams), the Fediverse champion in permission control.
Hubzilla could be something for companies, for organisations, for political offices, for journalists, for scientists etc. I could even see modern and progressive left-wing parties use it Pirate Parties, anyone They wouldn't have to worry about hub capacities because they could either use specialised hubs, e.g. for journalists, or they'd run their own hubs anyway, just like they run their own Mastodon instances now. In fact, if that thicket of instances run by German public broadcasters was Hubzilla instead of Mastodon, everyone could go nomadic without having to use general-purpose hubs for their clones.
But getting them from something as dead-simple as to the Leatherman of Fediverse projects is difficult, to say the least. Even from Facebook.
And private persons will only really tackle Hubzilla and stick with it if they're geeks enough.
#Long #LongPost #CWLong #CWLongPost #FediMeta #FediverseMeta #CWFediMeta #CWFediverseMeta #Lemmy #kbin #/kbin #Pixelfed #PeerTube #BookWyrm #Friendica #Hubzilla #FediverseMigration How about projects that use ActivityPub as a secondary protocol ((streams)) or as an optional add-on (Friendica, Hubzilla) while using something different with different features as the main protocol
I mean, I could write a lot about Hubzilla, and I occasionally contribute to the . But most of what could be written about Hubzilla would be about its inner workings (channels, permission control, no limits, theming etc.), stuff that doesn't federate in the traditional sense at all (articles, wikis, websites, the file space, CalDAV, CardDAV etc.) and stuff that only works on Hubzilla's main protocol Zot (nomadic identity).
While this is all part of what makes Hubzilla Hubzilla, none of it has got that much to do with ActivityPub, if anything.
#Long #LongPost #CWLong #CWLongPost #FediMeta #FediverseMeta #CWFediMeta #CWFediverseMeta #Hubzilla
One is that Mastodon users, of course, expect everyone to add alt-text to all their images. That's easy for Mastodon users: Images are attached to toots, and you get a nifty text field into which you put your alt-text.
This is different on Friendica, Hubzilla and (streams). Not only do you embed images much like in blog or forum posts. No, you don't have an alt-text field either. You embed these images using automatically generated BBcode, again, much like in forum posts. If you want alt-text, you have to edit that embedding code and graft the alt-text into it. Manually.
There is no official documentation on this whatsoever for either of these projects. The knowledge about this spreads by word-of-mouth only.
So users of these projects risk being ostracised for something they don't know a) that they have to do, b) that they can do and c) how they could possibly do.
The other issue is "image description == alt-text, full stop, end of discussion". The assumption that "alt-text" and "image description" are always mutually fully synonymous with no exception.
This comes from Mastodon's post length limitation. 500 characters. These are for the actual post, for mentions, for hashtags and for the content warning which actually counts into the post length. Doesn't leave you much room to describe an image, does it
But in alt-text, you have 1,500 characters. Per image. Always. No matter how much you want to toot otherwise.
With such a "high" limit, Mastodon users often don't simply add an alt-text of a maximum of 150-200 characters as defined in those many "how do I write alt-text for my static webpage with thousands of characters for explaining stuff" guides. No, they go well into detail. They give in-depth descriptions. They explain stuff. After all, they've got 1,500 characters of room.
Other Mastodon users love that, and they may do it themselves. Yet other Mastodon users see that the aforementioned Mastodon users love it and thus do it themselves.
With image descriptions of 700, 800, 1,000 or more characters, Mastodon users don't even question putting the descriptions into the alt-text. It's simply technologically impossible to do otherwise. So it doesn't even come to their minds. Image description == alt-text.
Okay, now here comes the catch.
First of all, you must never describe or explain something in alt-text that can't be accessed from anywhere else in the image.
That's because some people can't access alt-text, e.g. if they have a physical disability that prevents them from using a computer mouse, so they have to do everything with the keyboard, so they can't hover a mouse cursor over an image which is necessary to see alt-text in a desktop browser.
If they can't access alt-text, they can't access the information in it. If that information is only available in the alt-text but neither in the post text nor in the image, it's lost to them. Such information always goes into the post text.
Mastodon users are likely to be highly irritated upon reading that. How are you supposed to put that in the post text in the Fediverse! Mastodon doesn't give you enough characters! You can only put a detailed image description into the alt-text!
On Mastodon maybe. But only on vanilla Mastodon. Okay, and on Threads if it supports alt-text.
But everything that is not vanilla Mastodon gives you either more characters in the post text than in alt-text or no limits for either.
Even Misskey gives you 1,500 characters in alt-text and 3,000 in the post which is 2,500 more than on Mastodon. That's a thread of at least six posts rolled into one.
The other Forkeys give you 3,000 by default. Pleroma, Akkoma and their other forks give you 5,000 by default.
Friendica, Hubzilla and (streams) don't have any defined limits. Not for the post text and theoretically not for the alt-text either which, as I've explained above, is grafted into the post text and not put into a dedicated text field.
That's plenty of room for description and explanation, depending on how much you go into detail. In fact, you aren't even bound to a 1,500-character limit anymore!
I still expect Mastodon users to be highly irritated whenever I describe images. That's because my alt-text usually only contains a brief image description that doesn't actually describe anything. In addition, it says that the actual, fully detailed image description is in the post which is hidden behind a Hubzilla-style summary/Mastodon-style content warning combination.
I expect someone to try and educate me about all image descriptions always only belonging into alt-text because that's how it's done on Mastodon, yada yada yada. Regardless of the image description of even only one of the images being over 25 times longer than Mastodon's alt-text, so it wouldn't even fit in there.
Mastodon doesn't do it like this, so it's wrong everywhere else as well. Even though Mastodon doesn't do it like this because it can't, and everything else can, and Mastodon can happily display what everything else can.
#Long #LongPost #CWLong #CWLongPost #FediMeta #FediverseMeta #CWFediMeta #CWFediverseMeta #AltText #ImageDescription #ImageDescriptions #ImageDescriptionMeta #CWImageDescriptionMeta
I almost always put my image descriptions into the post text body for two reasons. One, they contain explanations and other information neither available in the actual post text nor for sighted people in the image itself. Two, they're way too long. They exceed 1,500 characters multiple times, sometimes multiple dozen times. I don't have local character limits to worry about, but at least Mastodon, Misskey and all their forks chop alt-text over 1,500 characters off at the 1,500-character mark.
In addition, I always mention in alt-text where exactly this image description can be found. Not only in the post text body, but I add stuff like "above the image and hidden behind a content warning if you're viewing this post on Mastodon and right below this image if you're viewing this post on Friendica, Hubzilla or (streams)".
More recently, however, I started adding a short description of the image to the alt-text. It actually doesn't even begin to describe the actual contents of the image. It's only there to satisfy the "alt-text must describe the image or else" crowd. I may have alt-texts well beyond the 850-characters mark nonetheless.
And still, I think the only reason why nobody complains about the lack of image description in my alt-texts is because too many Mastodon users or entire instances have muted or blocked me for not acting "Mastodon-like" enough.
#Long #LongPost #CWLong #CWLongPost #FediMeta #FediverseMeta #CWFediMeta #CWFediverseMeta #AltText #ImageDescription #ImageDescriptions #ImageDescriptionMeta #CWImageDescriptionMeta Long Weil es zu fast allen Themen Leute gibt, die sich darber aufregen, da die nicht hinter einer CW versteckt werden. Ja, auch Katzenbilder (CW: cats).
Deswegen bin ich dafr, da Mastodon mal den umgekehrten Weg geht. Zur Abwechslung mal nicht dem ganzen Fediverse seine eigene Kultur aufzwingt, die geprgt ist durch technisch ungerechtfertigte Beschrnkungen, Neuerfinden von Rdern und Walled-Garden-Denken. Statt dessen mal was nicht nur technisch, sondern vor allem in seine Kultur bernimmt, was im Fediverse vor Mastodon der Standard war:
Content Warnings aus Filtern. Automatisch generiert fr jeden Leser individuell.
Das kann Mastodon auch, und zwar seit Version 4.0, also seit letztem Jahr. Nur nutzt das keiner, und zwar aus einer ganzen Reihe an Grnden:
- Die Urgesteine und die Nutzer der ersten drei Migrationswellen (Februar 2022, November 2022, Mrz 2023) sind im Kopf immer noch auf anderen Mastodon-Versionen und haben von diesem Feature noch nichts gehrt.
- Newbies sind noch nicht so weit.
- Twitter hat sowas auch nicht, also kennt man das gar nicht. Vor allem nicht, wenn man Mastodon stur weiter wie Twitter benutzt.
- Fr technisch Unbedarfte sind die Filter viel zu kompliziert zu bedienen.
- Fr Mobilnutzer, also geschtzte 98% der heutigen Mastodon-Nutzer, sind sie unbedienbar. Ich glaube, keine einzige Smartphone-App hat Filtersteuerung eingebaut, also braucht es die Weboberflche. Und da gehen die meisten Mobilnutzer nie hin.
- Was ntzt es, Filter zu haben, wenn die Leute zu faul sind, Keywords oder Hashtags in ihre Posts einzubauen, die die Filter auslsen wrden
Erschwerend kommt hinzu: Viele Projekte, die nach Mastodon als Alternativen zu Mastodon gestartet wurden, wurden nicht nur auf Mastodons Technik, sondern auch auf Mastodons Kultur zugeschnitten. Entweder haben die gar keine Filter, oder die haben bis heute keine Mglichkeit, per Filter CWs zu generieren.
#Long #LongPost #CWLong #CWLongPost #LangerPost #CWLangerPost #FediMeta #FediverseMeta #CWFediMeta #CWFediverseMeta #CW #CWs #ContentWarning #ContentWarnings #CWMeta #ContentWarningMeta Okay, here's a bit more. And I'm trying to write this from a Mastodon point of view, but as someone who has got experience outside of Mastodon.
If you're on Mastodon which you probably are, get used to the lingo.
- (Twitter) tweet = (Mastodon) toot or post
- (Twitter) retweet = (Mastodon) boost
- (Twitter) like = (Mastodon) fave
This may come as a surprise, but the Fediverse is not only Mastodon. There are many other projects in the Fediverse which are connected to Mastodon. Some of them have actually already been around before Mastodon. And they have features that Mastodon doesn't have. So don't freak out if you see a post with over 500 characters. Or with italics. Or with bold type. Or with a
code block. Or using the content warning for a summary. Or with a quote. Or with a "quote-tweet". Outside of Mastodon, all this is standard and perfectly normal. Get used to it.Use content warnings (CW) to make browsing the #Fediverse safer for all.
Double your content warnings with hashtags. Some projects in the Fediverse outside of Mastodon use text filters to automatically generate individual content warnings for those who need them. And Mastodon can do that, too, so even some Mastodon users do it.
If a post with content that you may deem sensitive has only got these hashtags but no actual content warning, this is not necessarily due to ill intent or neglect. Some Fediverse projects outside of Mastodon can't do content warnings the Mastodon way. Hubzilla can only do them in first posts, but not in replies. On Friendica, they're generally very difficult to do. So don't immediately call for moderation.
Use #AltText on *all* your pictures! Visually impaired people like memes too.
Actually describe your images in alt-text. And do only that. No hashtags in alt-text. No links in alt-text. No SEO keyword spamming in alt-text.
If there's text in your picture, transcribe it in alt-text word by word.
You can mute, block, and/or report people, depending on the severity of their actions. That said, most people here are pretty cool, so give 'em a chance unless they're being a willful ass-hat.
Don't forget: Different instances have different rules. So don't report someone on another instance for not adding a content warning that's mandatory on your instance but not on theirs or the like. And don't forget either: Different projects have different capabilities and different cultures of using them.
#Long #LongPost #CWLong #CWLongPost #FediMeta #FediverseMeta #CWFediMeta #CWFediverseMeta #Fedisplaining #CWFedisplaining #MastodonIsNotTheFediverse #NotOnlyMastodon #FediverseIsNotMastodon #Hashtag #Hashtags #AltText
Anfangs schrieb ich lange Aufstze wie in der Schule. Dann meldete sich mal ein Blinder und meinte, allzu ausfhrliche Beschreibungen, seien gar nicht so erwnscht: Eigentlich klar: Whrend die Sehenden das mit einem Blick erfassen, mssen Blinde den ganzen Text sich vorlesen lassen. Am schwierigsten finde ich zu erkennen, was einem Blinden ntzt uns was nicht.
Da gibt es leider keinen wie auch immer gearteten Konsens, auch wenn einige behaupten, den gbe es. Genau wie mit der Lnge und Ausfhrlichkeit von Bildbeschreibungen.
Die einen Blinden wollen Bildbeschreibungen gern kurz und prgnant. Die anderen wollen sie ausfhrlicher. Z. B. lassen sie sich nicht damit abspeisen, da du Elemente in einem Bild erwhnst, sondern sie wollen auch wissen, wie das alles jeweils aussieht.
Wenn du jegliche Bildbeschreibung in den Alt-Text packst, bringt das fr Blinde noch einen weiteren Nachteil mit sich: Screenreader knnen Alt-Text nicht navigieren. Das heit, die knnen nicht innerhalb des Alt-Text an eine bestimmte Stelle zurckspringen, sondern nur wieder ganz zum Anfang, um dann alles wieder von vorne durchzurattern.
Das ist einer von diversen Grnden, warum ich z. B. meine ausfhrlichen Bildbeschreibungen immer in den Post selbst packe. Ein anderer ist, weil sie meistens weit lnger als 1500 Zeichen werden, aber zumindest Mastodon, Misskey und die Forkeys schneiden Alt-Text an der 1500-Zeichen-Marke hart ab.
Allerdings poste ich im allgemeinen Bilder zu einem sehr obskuren Nischenthema. Die meisten Leute brauchen sehr, sehr viel Erklrung, um die Bilder zu verstehen. Und gerade meine eigenen Bilder sind manchmal ziemlich komplex, auch wenn ich versuche, das zu vermeiden.
Wie gut meine Beschreibungen ankommen, wei ich nicht. Ich habe bisher mehr Bilder beschrieben als Feedback dafr bekommen, und zwar inklusive Feedback von Leuten, denen ich ausdrcklich meine Bilderposts als Beispiele fr lange, detaillierte Bildbeschreibungen gezeigt habe. Ungefragtes Feedback hatte ich bisher nur einmal, auch wenn das positiv war.
Wer noch nie "rot" gesehen hat fngt sich mit rotem Kleid wohl eher wenig an.
Nicht jeder, der blind ist, wurde blind geboren. Wer irgendwann in seinem Leben erblindet ist, hat frher mal Farben gesehen, also auch einen Begriff von Farben.
Auerdem ntzen Bildbeschreibungen nicht nur Blinden, sondern auch Sehenden, die mieses Internet haben, wo die Bilder berhaupt nicht laden.
Ich wrde Farben immer erwhnen.
Gibt es diesbezglich eine Anleitung/FAQ
Die meisten, die es gibt, sind nur fr professionelle/kommerzielle statische Websites oder vielleicht noch professionelle/kommerzielle klassische HTML-Blogs. Die meisten, die Social Media behandeln, sind nur fr kommerzielle Social Media, also Facebook, Twitter, Instagram, LinkedIn usw. Und die paar, die tatschlich das Fediverse behandeln, sind nur fr ganz normales Vanilla-Mastodon und gehen felsenfest davon aus, da Posts nicht lnger sein knnen als 500 Zeichen.
Die Sonderflle "Zeichenlimit deutlich ber 500 Zeichen" und "Social-Media-Post mit Bild, das viel Beschreibung und/oder Erklrung braucht", sind noch nie in so einer Anleitung abgedeckt worden.
CC:
#Long #LongPost #CWLong #CWLongPost #LangerPost #CWLangerPost #AltText #Bildbeschreibung #Bildbeschreibungen #BiBesch "MIA" means "missing in action".
I don't know all the details, just so much: The last time Kainoa was seen contributing to the Firefish repository was in June or July. Around that time, I guess, was also the last time that Kainoa was seen taking care of firefish.social. Kainoa is still alive and well and has posted something not long ago. Could be busy, nobody knows.
Anyways, fast-forward to December, 2023/January, 2024. Due to literally no maintenance whatsoever because the only admin hasn't shown up in a while, firefish.social has become entirely unresponsive. Nothing works anymore.
Also, Firefish is badly lagging behind Misskey's development. Not only no new features, but also no bug fixes, no security fixes, no nothing.
So the Firefish community took matters into their own hands. Just to mention three new forks:
- Iceshrimp was forked from Firefish, but pretty much immediately rebased to Misskey because Firefish was hopelessly outdated. It is currently being rewritten in C#.
- Sharkey was forked directly from Misskey, probably because Firefish's codebase would have needed too much maintenance. But some Firefish features shall be ported to Sharkey.
- Catodon was forked from Misskey as well, but with a different focus. Whereas Sharkey's goal is to become a Firefish replacement that has never been Firefish in the first place, having features that Misskey doesn't have while still keeping that typical Japanese style, Catodon is being tailored for Western tastes and simplified to Mastodon's level. Whether that only affects the UI or typical Misskey/Forkey features being cut, I don't know.
#Long #LongPost #CWLong #CWLongPost #FediMeta #FediverseMeta #CWFediMeta #CWFediverseMeta #Misskey #Forkey #Forkeys #Firefish #Iceshrimp #Sharkey #Catodon
Todays playlist is long songs. Everybody likes sagas:
This could work, and it could be a first step away from the oft-criticised "content warning field" and towards reader-side text filters which, by the way, have been in the Fediverse for longer than Mastodon itself.Web interfaces could get switches like these, too, and ideally, admins could add their own switches, depending on the target audience of an instance. Ideally, apps could recognise these custom, instance-specific switches and add them to their UI.
On the posters' side, it'd still require discipline because they'd have to add appropriate hashtags, and these hashtags would have to be standardised, otherwise each filter would have to cover each trigger with half a dozen different hashtags per language or so. Dedicated apps could assist in this by adding checkboxes to the post editor which have the app automatically add appropriate standard hashtags for certain topics to the "hashtag line" at the end.
This could actually start in apps without breaking standards, and it could nudge Mastodon itself into adding it to its Web UI. In fact, I can see at least some Forkeys being quicker at introducing it.
The best part: This would move the Fediverse away from something Mastodon-specific and towards something more standard.
#Long #LongPost #CWLong #CWLongPost #FediMeta #FediverseMeta #CWFediMeta #CWFediverseMeta #CW #CWs #ContentWarning #ContentWarnings #CWMeta #ContentWarningMeta
To be honest, I'm even feeling like maybe this shouldn't be a feature, just so the Twitter-refugees don't use it, because then they will think that that is somehow what's protecting them from harassment, and not the fact that this isn't Twitter.
Even if Mastodon won't introduce the creation of quote-posts, Misskey, the various Forkeys (Firefish, Iceshrimp, Sharkey, Catodon etc. etc.), Friendica, Hubzilla and (streams) can quote-post just about any Mastodon toot out there with no resistance whatsoever. They always could, even when Mastodon wasn't able to display quote-posts halfway appropriately.
#Long #LongPost #CWLong #CWLongPost #FediMeta #FediverseMeta #CWFediMeta #CWFediverseMeta To be fair, this doesn't apply to all the Fediverse. Mastodon, yes, but not the entire Fediverse.
I dare say there are two projects which handle this marvellously: from 2015 and a fork of a fork... by Hubzilla's own creator, commonly referred to as while technically being nameless. While both can "talk ActivityPub", they are based on different protocols which give them a feature that Bluesky tries hard to make everyone believe that they're working on inventing: .
First of all, Hubzilla and (streams) don't put your Fediverse identity directly into your account. Your account only grants you access. Your identity, your connections, your posts, your settings, your content etc. etc., all that resides in a , a kind of container. You can have multiple channels on one account, so you can also have fully separate identites on the same instance without having to log out and back in.
Nomadic identity itself can do two things. One, it takes care of moving a channel to another instance. And I mean actually moving it, not creating a dumb copy and leaving a dead account behind.
Two, and this is the actual killer feature, it can clone your channel to another instance. Again, not create a dumb copy. The clone is always synchronised with the original and vice versa. If one goes down, you've got the other readily at hand, so you're even safe from spontaneous shutdowns. And you can have more than one clone for each one of your channels.
Basically, moving is cloning light. Moving creates a synchronised clone, declares the clone the new main and the old main a clone, then deletes the old-main-gone-clone, and if there aren't any other channels on your old account, deletes that account.
Nomadic identity was invented in 2011 by who also created Hubzilla, now maintains (streams) and, in 2010, had created which was named Friendika back in 2011. Public Friendika nodes closing down had become a serious issue already, regardless of whether the shutdown was announced or came out of the blue. Early on, whenever this happened, all users on the node lost everything and had to start over from scratch.
Friendika eventually introduced moving from node to node by exporting and importing your account, but real resilience would have been different. Thus, nomadic identity was conceived. The Zot protocol was written around it because it couldn't be introduced to Friendika's DFRN protocol. In 2012, Friendica was forked and its backend re-written against Zot, becoming the Red Matrix, direct precursor of 2015's Hubzilla.
In a perfect Fediverse, everything would feature nomadic identity. But as it stands now, everything that isn't these two projects doesn't even understand nomadic identity and is easily confused by nomadic moves or someone posting from their clone instead of their main.
Unfortunately, nothing will change about this anytime soon. Mike Macgirvin has already proposed extensions for ActivityPub that'd introduce nomadic identity, albeit not in exactly the same fashion as on Hubzilla and (streams), but probably somewhat compatible. However, they have all been rejected.
#Long #LongPost #CWLong #CWLongPost #FediMeta #FediverseMeta #CWFediMeta #CWFediverseMeta #Hubzilla #Streams #(streams) #NomadicIdentity This might be a case of Chinese whispers.
What really increases accessibility when it comes to hashtags is how they're written if they consist of multiple words.
Always write multiple-word hashtags in "camelCase" (every word after the first one starts with a capital letter) or, better yet, "PascalCase" (every word starts with a capital letter). Screen readers used by blind or visually impaired users can't tell with certainty otherwise where a word ends and the next one starts.
Also, stop using Mastodon like Twitter and put all hashtags at the end of the post instead of into the text.
#Long #LongPost #CWLong #CWLongPost #Hashtag #Hashtags #Accessibility #A11y #camelCase #PascalCase
I consider this a reasonable compromise to be making with the people who are concerned about Quote-Responses, most of which seems to just be Twitter users, who are used to in environment which actively encourages using the feature, which is common on many websites, as a method of harassment, which is encouraged on Twitter but would be grounds for an instant ban on mainstream Fedi.
What do you mean would cause an instant ban The use of quote-posts for harassment or the use of quote-posts in general
I don't know if you've read the whole thread. I've already posted in it. If not: Lots of Fediverse projects have quote-posts regularly implemented. Two of them, Friendica and Hubzilla, have had them before Mastodon even existed, and it was Mastodon which federated itself with them and not the other way around. They call them "shared posts" and use them for the same purpose as Mastodon uses boosts: to forward posts.
They also both had quotes like the one above since their respective inception.
CC:
#Long #LongPost #CWLong #CWLongPost #FediMeta #FediverseMeta #CWFediMeta #CWFediverseMeta Have there ever been any specific promises how ease-of-use could be improved
I can't see anything that could be done without cutting features and betraying the Fediverse's principles except everything from Mastodon to Hajkey mimicking Twitter's default immediately-pre-Musk UI/UX. And maybe railroading newbies even harder by not even telling them that Mastodon is decentralised and taking instance selection away entirely and an AI that increases the mollycoddling by hiding Mastodon's decentrality and all contents coming in from the rest of the Fediverse until it decides the user is "ready".
CC:
#Long #LongPost #CWLong #CWLongPost #FediMeta #FediverseMeta #CWFediMeta #CWFediverseMeta Well, is writing about static websites and long-form blogs and not about the Fediverse, especially not about Mastodon.
On Mastodon very specifically, everything that describes and/or explains an image always goes into the alt-text. Unlike all other *blogging projects in the Fediverse, Mastodon with its tiny character limit doesn't leave you much of a choice. So people got used to it and see it as a set-in-stone Fediverse standard.
#Long #LongPost #CWLong #CWLongPost #FediMeta #FediverseMeta #CWFediMeta #CWFediverseMeta #AltText #ImageDescription #ImageDescriptions #Caption #Captions #ImageDescriptionMeta #CWImageDescriptionMeta Man darf sich aber nie der Illusion hingeben, da KI jedes, aber auch jedes Bild auf Anhieb perfekt akkurat und informativ beschreiben kann.
Ich will's mal so sagen: Einfache Katzenbilder bekommt KI beschrieben. Fr reine Text-Screenshots braucht es keine KI, da reicht OCR.
Aber wenn die Bilder a) hochkomplex sind, b) extrem nischiges Unbekanntes zeigen und c) gerade deshalb sehr detailliert beschrieben und umfassend erklrt werden mssen, damit das Publikum sie versteht, und wenn an die Beschreibungen hohe Ansprche gestellt werden, das wird KI nie so gut hinbekommen wie ein entsprechend kompetenter Mensch.
Generell sollte man KI-Beschreibungen nie aus Bequemlichkeit ungesehen copy-pasten, sondern immer gucken, ob das Geschriebene so stimmt, und gegebenenfalls hndisch nachbessern.
#Long #LongPost #CWLong #CWLongPost #LangerPost #CWLangerPost #KI #AltText #Bildbeschreibung #Bildbeschreibungen So you're finally no longer allowed to reply to a post that was written neither by a follower nor in response to a follower
Or, if you're on Friendica, Hubzilla or (streams), to a post that wasn't started by one of your followers because these three don't serve you single comments from within threads which didn't appear on your stream from the very beginning
Well, that makes showing Mastodon users the Fediverse outside of Mastodon much more difficult. Or that Mastodon's "rules" for how and where to describe images are not necessarily set in stone for the whole Fediverse because everything that isn't Mastodon doesn't have Mastodon's limitations.
Or that the first time the term "Metaverse" was used for an actual, real 3-D virtual world was not in 2021 by Zuckerberg for something he planned, but in 2007 for something that already existed and still exists.
I'd wonder if you're already a reply guy if you reply to a post in your local or federated timeline. But I don't have either.
CC:
#Long #LongPost #CWLong #CWLongPostBisher wei man wenig ber Long Covid bei Kindern und ob impfen dagegen hilft. Daten von ber einer Million Kindern und Jugendlichen zeigen nun moderaten Schutz durch die Impfung.
"The Metaverse" as in new 3-D virtual world projects of the 2020s in general
Or "The Metaverse" as in...
"Metaverse" and "The Metaverse" are registered trademarks of Meta Platforms, Inc. All rights reserved.
...as in "Zuckerberg's Metaverse" as in "Facebook's Metaverse" as in "Meta's Metaverse" as in Horizons
Stop calling it "The Metaverse" already!
Can everyone please with a cherry on top stop referring to Meta's virtual worlds as "the Metaverse"
They're called "Horizons"!
Zuckerberg did not invent the Metaverse. Zuckerberg did not even invent the term Neal Stephenson did in 1991. A Second Life in-world expo in the year 2007 was called "Metaverse". And the OpenSim community has been using the word "Metaverse" on a daily base down to grid names before 2010, too.
Thanks from someone who has been in OpenSim and thus known and used the term "Metaverse" in 2020 already in behalf of everyone in Second Life, OpenSim and all other virtual worlds that already existed before 2022.
P.S.: Second Life did not shut down in 2008 nor in 2009 either.
#Meta #Horizons #HorizonWorlds #HorizonWorldsIsNotTheMetaverse #Metaverse #TheMetaverse #SecondLife #OpenSim #OpenSimulator #VirtualWorlds #Long #LongPost #CWLong #CWLongPost
Oh, and by the way, a finding of mine for those who still think Zuck was the first to slap "Metaverse" on an actual real 3-D world:
"Metaverse". Used in OpenSim since 2007.
Something for those who think Mark Zuckerberg has invented the term "Metaverse" in 2021.
, the oldest grid (est. 2007), is referring to itself as "The Open Source Metaverse". Guess what It has taken over that slogan from OpenSim itself, and .
Metropolis, the first German grid (est. 2008), boldly used the full name "Metropolis Metaversum" until it was officially shut down in summer 2022. .
Since I started out in Metropolis on April 30th, 2020, I was in something called Metaverse almost a year and a half before Zuckerberg used that term.
Another fun fact: The virtual worlds/virtual reality news site has named its OpenSim category "" as early as 2010.
#OpenSim #OpenSimulator #Metaverse #VirtualWorlds #Metropolis #OSgrid
For the record: Yes, I know it was Neal Stephenson who coined that term in 1991 already.
#Meta #Horizons #HorizonWorlds #HorizonWorldsIsNotTheMetaverse #Metaverse #TheMetaverse #SecondLife #OpenSim #OpenSimulator #VirtualWorlds #Long #LongPost #CWLong #CWLongPost
More accountability: Ontario launches team to crack down on bad LTC actors
The 10-person team will be an "effective deterrent and tool" when escalated enforcement is required, the province said.
-termCare
More accountability: Ontario launches team to crack down on bad LTC actors
The 10-person team will be an "effective deterrent and tool" when escalated enforcement is required, the province said.
-termCare
More accountability: Ontario launches team to crack down on bad LTC actors
The 10-person team will be an "effective deterrent and tool" when escalated enforcement is required, the province said.
-termCare
More accountability: Ontario launches team to crack down on bad LTC actors
The 10-person team will be an "effective deterrent and tool" when escalated enforcement is required, the province said.
-termCare
Part of the secondary topic of this channel is showing a very-much-non-Mastodon view on the Fediverse to the Fedizens out there. Lots of fresh Twitter-to-Mastodon converts started following me in late 2022 and early 2023 because I was one of the few to explain the Fediverse outside Mastodon to Mastodon users.
This doesn't work anymore if my posts don't reach their recipients anymore.
And besides, triggering Mastodon users with non-Mastodon behaviour can have nasty side-effects. Yeah, some mute you. Some block you. Often even without also unfollowing you. Some complain to you about it and demand you don't use any features of your instance that vanilla Mastodon 3.x didn't support yet.
But then there are those who try to complain to your instance admin about you and try to have you sanctioned. Or they have their instance admins/mods complain to your instance admin about you. And this can escalate.
A Hubzilla admin would laugh that off. Most Hubzilla hubs don't even have rules, and the rules of any one Mastodon instance don't apply here either.
But in most cases, the complaints never reach their recipients. See, the average Mastodon user doesn't know anything about Hubzilla. So they try to complain the same ways as they would within Mastodon.
They use Mastodon's report feature, but Hubzilla doesn't have it implemented. They try to send your admin a DM, but finding the admin of your hub doesn't work like finding the admin of a Mastodon instance, and oftentimes, the admin's channel isn't linked anywhere at all. They publicly complain about you with a Mastodon admin hashtag, hoping that your admin will read it. But you can't follow hashtags on Hubzilla, and even if you could, why would a Hubzilla admin follow a Mastodon-specific hashtag
So the admin and the "moderation" of your hub doesn't react. Conclusion for a Mastodon user who only knows Mastodon: Your instance is basically unmoderated. And what does Mastodon do with an unmoderated instance whose users misbehave for any definition of that word Block it. Like, the whole instance.
One Hubzilla user misbehaving in the eyes of Mastodon users can cause the whole hub to be blocked on thousands of Mastodon instances, maybe including mastodon.social.
Now you could say I should flip the ActivityPub side of the Fediverse the bird and turn Pubcrawl off for good. If I did that, I could just as well delete the whole channel because it'd lose its very purpose.
#Long #LongPost #CWLong #CWLongPost #FediMeta #FediverseMeta #CWFediMeta #CWFediverseMeta The biggest cultural difference based on server software has to be between Mastodon on the one hand and the Mike Macgirvin creations Friendica, Hubzilla and (streams) on the other hand.
Not only do they have vastly different user bases, but they developed independently from one another. When the Mastodon culture developed, those who shaped it didn't even know the other side, so they couldn't adopt any of its culture.
Said other side's culture dates back to 2010 when Friendica was launched as Mistpark, and since that was almost six years before Mastodon, it couldn't be inspired by Mastodon's culture either.
Add to that that these respective cultures are greatly shaped by technical features and limitations or the lack thereof.
Mastodon's culture is largely built around its 500-character limit which is ample for your typical phone-wielding Mastodon user. Friendica, Hubzilla and (streams) don't have any defined character limit whatsoever, and its target audience is largely on desktop or laptop computers, often running Linux, with large screens and full-size hardware keyboards.
So it's the most normal thing in the world for them to write in one post as much as they want while Mastodon users debate whether threads are good, or you should always limit yourself to 500 characters or less.
Also, alt-text. Mastodon has many disabled users, including blind or visually-impaired users. And it has a dedicated alt-text field for each image. On top of that, it offers 1,500 characters for each alt-text which, in connection with the 500-character limit for toots, has people write detailed image descriptions and explanations and put them into the alt-text. That's often information that doesn't even belong in alt-text, but there's no room for it elsewhere.
Friendica, Hubzilla and (streams) have unlimited room, so putting stuff into alt-text because the post text is too limited seems ridiculous. But they don't have a vocal disabled community, so there's little interest in accessibility. And neither of them has a dedicated alt-text field. Alt-text is supported, but it has to be manually grafted into the image-embedding code in the post. And there's no official documentation for that, I think not even for Friendica which is the only one out of the three with actually useful end-user documentation.
It's similar with content warnings. On Mastodon, they're put into the repurposed summary field, and next to nobody knows that it's a repurposed summary field rather than invented for content warnings from scratch. So since Mastodon has a content warning field, writer-side content warnings are huge, but also cause for drama.
Mastodon 4.0 has introduced filters that can create reader-side content warnings, but hardly anyone uses them, even fewer people support them with keywords or hashtags, many don't even know this feature exists, and it's generally ignored because it's un-Mastodon.
The Friendica/Hubzilla/(streams) complex doesn't have a content warning field. Hubzilla and (streams) have a summary field labelled as such. Friendica doesn't even have that it uses a pair of BBcode tags for that.
And within their own ecosystem, they don't even need it. They've got the "NSFW app" instead, an over-one-decade-old, optional, simple-as-anything substring filter that automatically hides entire posts with all media and everything behind content warnings if it finds one of the entered keywords or hashtags.
So they can't understand Mastodon's commotion about content warnings, and Mastodon users can't understand why they don't add Mastodon-style content warnings.
And then there are all the things that were or are being debated on Mastodon and whether or not it should introduce them. Especially the second-wave Twitter refugees are often staunchly against them.
Full-text search and quote-tweets are being actively used on Twitter to track down and harass members of minorities who have fled to Mastodon. Of course, they don't want Mastodon to introduce either. That is, in the case of full-text search, Mastodon has found a solution, but one that doesn't really federate to the rest of the Fediverse.
Quotes and text formatting are seen as bad, too. Many don't know quotes because Twitter doesn't have them, and so they think quotes could be used as tools of harassment. And both are seen as making Mastodon feel less like Mastodon and more complicated.
Friendica and Hubzilla had all this before there was Mastodon, and (streams) inherited it from its long line of ancestors. Their users have gotten so used to having all this that they don't understand what the problems should be, also because they're so detached from Mastodon's culture. So they keep on using these features unashamedly, even around Mastodon users.
Some differences are rather simple. Take mentions, for example. Friendica has always used long names for mentions, as does its offspring. Mastodon users may find that freaky. Meanwhile, Friendica, Hubzilla or (streams) users may find Mastodon's mentions cryptic because they use the short name. But even they matter.
#Long #LongPost #CWLong #CWLongPost #FediMeta #FediverseMeta #CWFediMeta #CWFediverseMeta #FediverseCulture #Fediverse #Mastodon #Friendica #Hubzilla #Streams #(streams) #CharacterLimit #CharacterLimits #500Characters #CW #CWs #ContentWarning #ContentWarnings #ContentWarningMeta #AltText #ImageDescription #ImageDescriptions #ImageDescriptionMeta #CWImageDescriptionMeta #Quotes #QuotePost #QuotePosts #QuoteBoost #QuoteBoost #QuoteTweet #QuoteTweets #QuoteToot #QuoteToots #QuotedShares #TextFormatting #FullTextSearchScott M. Stolz I'm not talking about the occasional, moderate use of stuff that Hubzilla can do that Mastodon can't.
I'm not even only talking about using such stuff all the time and not in moderate amounts.
I'm talking about things like "content warnings" that read, "Summary of the post CW: long (number of characters), stuff-I-think-not-everyone-wants-to-read-about meta, more-stuff-I-think-not-everyone-wants-to-read-about meta." And then forgoing content warnings altogether in replies.
Or having eight and more hashtags at the bottom of a post, some of which look totally weird because they're the same as another one of the hashtags, but with "CW" at the beginning. (I use them to trigger filters. See below.)
Or image posts with alt-text for the images, but the alt-text wastes over 800 characters barely even describing the image and mentioning that the actual detailed image description is in the post itself. WTF, nobody puts image descriptions in the post, they belong into alt-text and nowhere else!
Worse yet: The post is hidden behind a content warning. And the content warning says something about tens of thousands of characters. And then you click/tap the content warning. And then an utter wall of text jumps at you, way longer than 100 toots could ever be, all in one enormous chunk.
Last but not least, my constant jabs at Mastodon, Mastodon users, especially Twitter refugees, Mastodon culture...
#Long #LongPost #CWLong #CWLongPost #FediMeta #FediverseMeta #CWFediMeta #CWFediverseMeta You can't "quote-toot" tweets on Mastodon. In fact, you can't "quote-toot" toots on Mastodon. You can't "quote-toot" anything on Mastodon.
So unless you're on Misskey or one of its forks or on Friendica or one of its descendants, or unless you want to quote something outside the Fediverse, screenshots are the way to go.
That's also because next to everyone on Mastodon is on a phone. And copy-pasting text from e.g. the Twitter app to whatever Mastodon app you use and copy-pasting the URL of e.g. the tweet into that Mastodon app is very inconvenient because copy-paste is a pain on a phone with no separate pointing device and no hardware keyboard.
Add to this that Mastodon can't only not "quote-toot", it can't quote either.
#Long #LongPost #CWLong #CWLongPost #FediMeta #FediverseMeta #CWFediMeta #CWFediverseMeta
Maria Ryabushkina
Russian Model
Winner Best Stringkini
Use as many as you need. But don't abuse them. For example, there are people who slap #Mastodon on posts that aren't even about Mastodon or #Fediverse on posts that aren't about the Fediverse in general, just to try and increase their range. This is bad behaviour, this is what's counterproductive, and it's actually often being sanctioned.
Still, there shouldn't be an upper limit. Some Fediverse users have filters that are triggered by hashtags and either remove posts from their timelines or put them behind automatically generated reader-side content warnings.
On Friendica, Hubzilla and (streams), this is actually the default, and it has been since even before Mastodon existed, and neither of them has a dedicated "content warning" text field. Also, at least on the latter two projects, there is no means whatsoever to add a Mastodon-style content warning to a comment like this.
Thus, at least some users, especially from these three projects (although (streams) isn't a project), add hashtags for the purpose of triggering filters. As you'll see at the bottom of this comment, it'll cause a whole lot of hashtags for the standards of someone who's used to Twitter. But they're deeply ingrained in the culture of certain non-Mastodon projects. And they're useful for at least some Mastodon users.
So any reply you may get that tells you that nobody in the whole Fediverse can possibly want more than four or eight hashtags is simply false. Some don't want more than that. Others depend on them.
#Long #LongPost #CWLong #CWLongPost #FediMeta #FediverseMeta #CWFediMeta #CWFediverseMeta #Hashtag #Hashtags #HashtagMeta #CWHashtagMeta Okay, this one is going to be long. Don't let the length and the text formatting disturb you. That's how we do things over here on Hubzilla, and I'm glad Mastodon can at least display them now.
So let's see...
First of all, it's unusual to have multiple description of an image in the alt-text. I'd say it's redundant, but I myself give short descriptions in the alt-text and then much longer ones than yours in the post text body which the alt-text also references.
Pick only one. For technically speaking, alt-text is not a synonym for a short description. Alt-text is the whole thing you create when pasting the AI output into that text field for images. It's what stands in for the image when you can't see the image for whichever reason, and it's all of it.
And don't add "alt-text" to alt-text.
Next, please excuse the pun, but it looks like you put blind faith into AI, and you copy-paste its output into your alt-texts unread, not to mention unredacted.
In the description of the fourth image you've linked, it really goes to show that whatever AI you use (ChatGPT for detailed image descriptions) doesn't even recognise what's in the lower picture. It doesn't get the picture, so it doesn't get the meme, so it fails to explain the meme.
I would definitely have added the information that the lower picture shows Christopher Lloyd as Doctor Emmett L. Brown in the final showdown scene of the 1985 film Back To The Future, trying to reconnect the lightning conductor cable which is about to channel the lightning bolt that shall damage the clock on Hill Valley's town hall into the DeLorean time machine with Marty McFly inside and use the energy to send him back into his time.
That is, in practice, I would have chosen a different wording and added another bit more of information. But without it, the image description is half-useless because it doesn't mention what the meme references and hints at.
Yes, it's easier and more convenient to believe in perfect fire-and-forget recognition, identification, description and explanation of any image imaginable by AI. But if you really want your image descriptions to be informative and, above all, correct, read them. And redact them. Or, better yet, write them yourself.
#Long #LongPost #CWLong #CWLongPost #FediMeta #FediverseMeta #CWFediMeta #CWFediverseMeta #AltText #ImageDescription #ImageDescriptions #ImageDescriptionMeta #CWImageDescriptionMeta #AIScott M. Stolz
For example, it might be useful to implement the following:
1. A standard way to announce to other projects who the administrators or moderators are.
2. A standard way to report a post via ActivityPub that automatically gets delivered to the correct people.
3. A moderation system for handling reports on each platform.
4. Bounce messages if a message cannot be delivered.
It would still be up to the projects to implement this, but developers are more likely to implement it if there is a standard.
Good luck getting Mastodon to implement any standard they haven't invented themselves.
This will end in
- Mastodon trying to push their own standard upon everyone else, Mike Macgirvin having it in dev alreeady, but Mario Vavti staunchly refusing it because he doesn't let Mastodon push anything upon him
- Mike Macgirvin making his own proposal, the best one of them all, which is rejected by everyone from the W3C to Eugen Rochko
- maybe, if they get their act together, the W3C making a proposal which Mike Macgirvin either rejects for being ridiculously weak or implements within 15 minutes, which Mario Vavti rejects because it'd be next to impossible to weave into Hubzilla and Zot6, and which Eugen Rochko rejects because Mastodon already has something that covers this, and everyone else must adopt this or else
For those concerned about privacy, if you are in a country that does not require you to disclose who runs the website, you could just put "Administrator" as the name of the Administrator. Even though you did not give a person or organization's name, there is still a listed Administrator for the instance.
This won't change much about many public, open-registration Friendica/Hubzilla instances not providing a link to the account/channel of the admin. Or any admin contact at all. All because the admin didn't bother to edit that page/hasn't gotten to it yet.
And it changes nothing about Mastodon users/admins trying to reach Friendica/Hubzilla/(streams) admins the exact same way as they always try to reach other Mastodon admins. And failing because it doesn't work like on Mastodon. Even less in a mobile app that's geared towards Mastodon and Mastodon only.
Don't forget: On Hubzilla, everyone's on a Linux PC using a standard browser. On Mastodon, everyone's on a phone using a Mastodon app.
#Long #LongPost #CWLong #CWLongPost #FediMeta #FediverseMeta #CWFediMeta #CWFediverseMeta If done the Friendica/Hubzilla/(streams) way, that'd be disturbing to Mastodon users again.
I think nobody on Mastodon knows this, but these three projects have a Facebook-like/Tumblr-like/blog-like one-post-many-comments thread structure as opposed to Mastodon's Twitter-like many-single-posts-tied-together thread structure.
This means the thread starter owns the whole thread. This also means the thread starter can moderate their own thread and even delete comments from their own thread. Even if they weren't appointed moderators by the admin. AFAIK, the deletion is actually forwarded to the source.
So a Mastodon user may e.g. troll around in a thread started on Hubzilla or attack the thread starter. A bit later, that Mastodon user will discover that their toots in reply to that Hubzilla posts are gone. Maybe the thread starter even mentions they've moderated their own thread and deleted comments.
The short form: Regular users of Friendica, Hubzilla and (streams) can delete Mastodon toots. In fact, posts from anywhere. Only toots/posts in reply to their own posts, but yes, they can. And they could before Mastodon even existed.
If this started happening often enough, and word started spreading around especially Mastodon, it'd be perceived as invasive and disruptive by many. It might cause such a big uproar that even more Mastodon users would call for defederating Friendica, Hubzilla and (streams) completely.
As for actual, traditional moderation, I'm not saying it's bad per se.
I'm saying that many Mastodon admins really don't know anything about the Fediverse outside Mastodon.
They think everything else that does *blogging is basically Mastodon with a different UI. Like an alternative mobile app for Mastodon itself. Or it's Mastodon + feature x on top. But otherwise like Mastodon.
And then they act accordingly. They try to interact with Friendica, Hubzilla and (streams) admins the same way as they interact with other Mastodon admins. And they neither know nor realise that these three projects are too different from Mastodon for these methods to even work!
Again:
You can't report a Hubzilla user to a Hubzilla admin using the report system because Hubzilla does not have Mastodon's report system!
You can't expect Hubzilla admins to follow hashtags for Mastodon instance admins either. Not only because they aren't Mastodon admins, and they probably don't even know that hashtag. But because you can't follow hashtags on Hubzilla!
And so forth.
So what appears to be a Hubzilla admin's laziness or carelessness or ignorance are actually technical differences between Mastodon and Hubzilla that Mastodon admins are unaware of.
This has nothing to with moderation by admins/mods only being bad. This has everything to do with Hubzilla being vastly different from Mastodon. And with Mastodon admins/mods not knowing jack shit about Hubzilla.
(Good thing this is so long that nobody will read it. Otherwise it'll empower those who say that Friendica, Hubzilla and (streams) must be redesigned to be exactly like Mastodon or face full defederation otherwise for being too different.)
#Long #LongPost #CWLong #CWLongPost #FediMeta #FediverseMeta #CWFediMeta #CWFediverseMeta gave me to think.
Could it be that countless Friendica, Hubzilla and (streams) users are blocked on countless mostly Mastodon instances by the admins because reporting users the Mastodon doesn't work on these projects
So there's a user who doesn't fully act according to the Mastodon community standards. That user's posts appear on some Mastodon instance.
The wrongdoing: For example, what's perceived as hashtag abuse see the linked thread. Or no Mastodon-style content warning where Mastodon culture would demand one*. Or something like that.
What does the admin do Use the report system to report that user to the admins and moderators of their own home instance.
Problem: That particular user isn't on Mastodon. Not on anything that was modelled after Mastodon either. That user is on Friendica or Hubzilla or (streams). Correct me if I'm wrong, but AFAIK, neither has Mastodon's report system implemented.
The report never reaches the admin of that instance. And the instance doesn't have any more staff.
Well, then they could write directly to the admin of that instance. If only the Fediverse contact of the instance admin was available on the instance frontpage. Or anywhere on the instance Web interface.
Even if they could, they might get the idea that they could catch the admin's attention by mentioning them in a public post. Spoiler: Doesn't work with Friendica accounts, Hubzilla channels and (streams) channels.
Oh, and at least Hubzilla and (streams) allow you to restrict from whom you receive direct messages. Regardless of whether or not that's a good idea, it's possible to make it so that DMs from random Mastodon users no longer end up in your stream. Worse yet: These Mastodon users don't even know that their DMs don't reach the recipient.
Okay, last resort, complaints about that user can be posted publicly under the hashtag #MastoAdmin. Should reach lots of admins, right
Yes, but almost exclusively Mastodon admins. It's MastoAdmin, after all. Why should an admin of, say, a Friendica node or a Hubzilla hub follow that hashtag Neither of them is Mastodon, and neither of them has anything to do with Mastodon. They didn't even federate with Mastodon, Mastodon federated with them.
Oh, and besides, to my best knowledge, they can't even follow hashtags in the first place. Or is Hubzilla the only one out of the three that doesn't have that feature yet
Anyways, the warning with the #MastoAdmin hashtag doesn't reach them either.
So whatever you try to let some Friendica or Hubzilla or (streams) admin know that a user on their instance "misbehaves", the admin doesn't react and "moderate" that user.
Conclusion for your typical Mastodon admin: That instance is unmoderated. From the point of view of people who only know Mastodon beyond the name, the admin must ignore all reports.
We can be glad if this leads only to blocking the "misbehaving" user on lots of Mastodon instances and not to what's standard for unmoderated or undermoderated instances on Mastodon: blocking the whole instance.
*Footnote: Neither of the three projects mentioned here has a "Content Warning" field. Hubzilla and (streams) have a "Summary" field which is the same thing, but especially newbies and those who are hardly in touch with the ActivityPub side of the Fediverse don't know it's the same. Also, that field is only available for posts (= first posts) and not for comments (= replies which are something entirely different on these projects). Friendica doesn't even have that a pair of BBcode tags is needed for a Mastodon-style content warning, and AFAIK, this isn't documented anywhere.
#Long #LongPost #CWLong #CWLongPost #FediMeta #FediverseMeta #CWFediMeta #CWFediverseMeta #BlockingMeta #BlocklistMeta #CWBlocklistMeta
I report the behavior through the report system and do my own justice. Maybe post a link to the account and the offending content with MastoAdmin and Fediblock tags and let people do their own research.
But what if the offending user is on an instance
a) of something that is not Mastodon, that wasn't even inspired by Mastodon because it's older than Mastodon, and that doesn't have Mastodon's report system implemented
b) whose admin doesn't follow MastoAdmin because they aren't Mastodon admins
c) whose admin doesn't follow MastoAdmin because that project doesn't have following hashtags implemented
So whatever you do the Mastodon way never reaches the admin in the first place
#Long #LongPost #CWLong #CWLongPost #Defederation #FediMeta #FediverseMeta #CWFediMeta #CWFediverseMeta Just out of curiosity:
Let's suppose you come across someone whom you find misbehaving. You try reporting them to their instance moderation.
But they aren't on Mastodon. They're on Friendica or Hubzilla or (streams). And you can't for the life of you figure out how to get into contact with the admin of that specific instance via Mastodon. And yes, it's a public instance.
What do you do Defederate the whole instance because its management doesn't appear to be up to Mastodon's community standards Demand it be fediblocked altogether for that reason
Or what if there's only an e-mail address, but no link to a Fediverse account If that's a problem for Mastodon instance admins, I might single-handedly be responsible for hub.netzgemeinde.eu being defederated by dozens or hundreds of Mastodon and *key instances for "non-Mastodon-standard" behaviour.
(Hashtags below mostly for filter-triggering purposes. Text filters are the CW standard where I come from and have been since at least 2012.)
#Long #LongPost #CWLong #CWLongPost #Defederation #FediMeta #FediverseMeta #CWFediMeta #CWFediverseMeta Ich mache das anders. Ich warne im Summary (Mastodon: Content Warning) vor allem, was jemandem auf Mastodon unangenehm sein kann, und die eigentlichen Content Warnings dopple ich noch einmal mit jeweils einem und mehreren Hashtags fr die, die Filter verwenden.
Das ist auch ein bichen eine Demonstration dessen, was dabei rauskommen kann.
Zeichen habe ich selbst mehr als reichlich zu verbrennen, will sagen, praktisch unbegrenzt viele. Daran soll's nicht scheitern. Und wenn der Post lnger als 500 Zeichen wird, gibt's dafr eine Content Warning und die Hashtags #Long, #LongPost, #CWLong und #CWLongPost und bei deutschsprachigen Posts - wobei ich die Postsprache nicht einstellen kann - zustzlich #LangerPost und #CWLangerPost.
Bei Kommentaren gibt's dann nur die Hashtags, weil Mastodon-CWs nicht gehen.
CC: He may have a point, yes. There may be more people in the Fediverse who think like him, yes. He even sounds like he has come across one of my posts.
But if I did what he asked for, all my regular posts would grow to thousands upon thousands of characters and take hours to write because I'd have to break everything down to the very basics. Both the primary and the secondary topic of this channel are niche within a niche, maybe within yet another niche for a non-technical audience.
Essentially, I couldn't post anything anymore without long-post content warnings for Mastodon users who are disturbed by everything longer than 500 characters.
And my image descriptions for moderately complex in-world images, now sitting at 35,000 to 40,000 characters for the first image in a post and not counting the preamble, might easily grow beyond 50,000 or 60,000 characters. The preamble would grow well over 5,000 characters. Even longer if I took people's attention spans into consideration and explained certain things again and again.
This wouldn't even really help people outside my niche with understanding my posts. Even fewer of them would read them in the first place, and even more of them would mute or block me to get rid of those monstrous posts in their timelines.
And no, I can't explain the implications of the current scheduled OSgrid downtime in one simple sentence and in plain, everyday language to someone who doesn't even understand what a server is and knows exactly nothing about 3-D virtual worlds. Not in a way they'd fully understand.
Some of ya'll need to step back from the hyper-specific niche topics you post about and take a second to think about explaining your topics to normies, like ELI5 once in a while. Sometimes I'm scrolling my feed and I'm like wtf is all this this about.
Maybe I just curated a bunch of people with hard to follow niche interests.
Am I the only one
It seems like people aren't posting to have a conversation, It's just monologuing. I prolly fall into this once in a while as well.
#Long #LongPost #CWLong #CWLongPost
Well, Lemmy is pretty much established right now and growing as well, despite being something different. Right now, this seems the biggest competitor to Mastodon in a way. But the limitation of Lemmy accounts not being able to follow other accounts (and only communities instead) makes it a bit of an outsider. Like, you can follow users from Friendica, Mastodon, even Hubzilla I think, and see their activity (comment, post), but they cannot see yours.
This also leads to the creation of a more specific culture, with people that are also unaware of other platforms and capabilities (albeit less pronounced, as the devs did a better job of keeping the whole ecosystem decentralized). For example, some users are surprised to hear that you can see their upvotes on other platforms.
The latter is mostly because the vast majority of Lemmy users didn't come from Mastodon but from Reddit. They weren't told much about the existence of a Fediverse, only that there's a thing called Lemmy which is many copies of Reddit before its enshittification, and these are connected with each other. That's all that many know. Just like many Mastodon users think the Fediverse is only Mastodon.
It doesn't help that Lemmy barely gets any interaction from other projects. Mastodon is huge, and Mastodon users should be all over Lemmy. But many Mastodon users have never heard about Lemmy. Those who have may find it too inconvenient to follow a Lemmy community because that involves using the account search and copy-pasting. Don't forget that the huge majority of Mastodon users is on phones. And those who do manage to follow Lemmy communities say that the interaction between Mastodon and Lemmy is too limiting.
For the record: I do have Lemmy followers.
Kbin and Mbin could have brought a solution to this issue, as they do support following users, even those from *blogging platforms, but because the projects are younger and less stable - and more so at the time of the Reddit migration - they failed to gain the required traction (i.e. more servers, user numbers more spread out across them) until now. There is still activity on these, they are still growing (people are joining them mostly because they are dissatisfied with the political leanings of the Lemmy devs, as well as their moderation policy on .ml which is subjective to say the least), but you can clearly see a bigger culture formed around Lemmy as of now.
/kbin made bidirectional *blogging-style following possible only by bolting microblogging onto a Reddit clone. Lemmy is a more purist Reddit clone, it doesn't support domestic *blogging, so users of *blogging projects can't follow Lemmy users in the traditional sense.
As for lemmy.ml, that instance doesn't matter that much anymore. Even lemmy.world has been surpassed as the biggest instance.
There seems to be something similar happening to Pixelfed, with user numbers growing month after month, and I am sure something similar will happen to Peertube when YouTube will flop badly again and PeerTube will be mature enough, or with Bookwyrm, Friendica, Hubzilla etc.
Pixelfed could become big if Instagram was enshittified so tremendously that everyone except the biggest attention whores ("But muh followers, but muh fame") will start looking for alternatives. The advantage of Pixelfed for Instagram users over Mastodon for Twitter users is that Pixelfed allows direct imports of Instagram accounts with all content.
I'm not so sure about PeerTube, not only because that'd require gigantic amounts of hard drive space, but also because many users are on YouTube for the money, and PeerTube won't pay them a penny. If they moved to PeerTube, they'd lose a source of income. Also, fewer YouTube users have ever heard of PeerTube than users have heard of Mastodon.
BookWyrm would be easier, but I can't see right now how Goodreads could be enshittified enough to cause a mass migration. Maybe, however, BookWyrm becomes interesting for people who don't even know Goodreads and its whole concept, and they find out about BookWyrm before they find out about Goodreads.
Friendica tried to take a chance long ago, back in the early 2010s. It even tried to facilitate the transition of whole social circles from Facebook by federating with Facebook by means of a cross-poster. It didn't work out. People didn't want to leave their "friends" behind, not to mention that the average Facebook user was even less technologically adept than the average Twitter user a good decade later. Even trying to mimic Facebook's UI didn't help.
Maybe it was for the better. Typical server hardware that Friendica ran on back in the day could barely handle over 130 accounts on one node. The notoriously power-hungry Facebook connector cut a dozen or two out of this number. Many public Friendica nodes with the Facebook connector on closed their registrations at a bit over 100 accounts. I think not even the biggest root servers would have given you a four-digit capacity.
It would simply have been impossible to accommodate a flood of Facebook refugees on Friendica. Even if Friendica users with Facebook contacts had started their own private nodes, most of them would have needed multiple nodes to even have space for a fraction of their Facebook "friends".
As for Hubzilla, it'll first need a lot of polish. And then I can't see from where people would come flooding to Hubzilla. Facebook refugees would rather pile onto Friendica, the traditional more-powerful-than-Diaspora* Facebook "clone", or maybe (streams), the Fediverse champion in permission control.
Hubzilla could be something for companies, for organisations, for political offices, for journalists, for scientists etc. I could even see modern and progressive left-wing parties use it Pirate Parties, anyone They wouldn't have to worry about hub capacities because they could either use specialised hubs, e.g. for journalists, or they'd run their own hubs anyway, just like they run their own Mastodon instances now. In fact, if that thicket of instances run by German public broadcasters was Hubzilla instead of Mastodon, everyone could go nomadic without having to use general-purpose hubs for their clones.
But getting them from something as dead-simple as to the Leatherman of Fediverse projects is difficult, to say the least. Even from Facebook.
And private persons will only really tackle Hubzilla and stick with it if they're geeks enough.
#Long #LongPost #CWLong #CWLongPost #FediMeta #FediverseMeta #CWFediMeta #CWFediverseMeta #Lemmy #kbin #/kbin #Pixelfed #PeerTube #BookWyrm #Friendica #Hubzilla #FediverseMigration How about projects that use ActivityPub as a secondary protocol ((streams)) or as an optional add-on (Friendica, Hubzilla) while using something different with different features as the main protocol
I mean, I could write a lot about Hubzilla, and I occasionally contribute to the . But most of what could be written about Hubzilla would be about its inner workings (channels, permission control, no limits, theming etc.), stuff that doesn't federate in the traditional sense at all (articles, wikis, websites, the file space, CalDAV, CardDAV etc.) and stuff that only works on Hubzilla's main protocol Zot (nomadic identity).
While this is all part of what makes Hubzilla Hubzilla, none of it has got that much to do with ActivityPub, if anything.
#Long #LongPost #CWLong #CWLongPost #FediMeta #FediverseMeta #CWFediMeta #CWFediverseMeta #Hubzilla
Enchanting Chinese Beauty with Long Hair in the Futuristic Starfleet
Checkout more promptden.com
-fi -hair -beauty -intel
Nothing per se. The issues I've mentioned are different.One is that Mastodon users, of course, expect everyone to add alt-text to all their images. That's easy for Mastodon users: Images are attached to toots, and you get a nifty text field into which you put your alt-text.
This is different on Friendica, Hubzilla and (streams). Not only do you embed images much like in blog or forum posts. No, you don't have an alt-text field either. You embed these images using automatically generated BBcode, again, much like in forum posts. If you want alt-text, you have to edit that embedding code and graft the alt-text into it. Manually.
There is no official documentation on this whatsoever for either of these projects. The knowledge about this spreads by word-of-mouth only.
So users of these projects risk being ostracised for something they don't know a) that they have to do, b) that they can do and c) how they could possibly do.
The other issue is "image description == alt-text, full stop, end of discussion". The assumption that "alt-text" and "image description" are always mutually fully synonymous with no exception.
This comes from Mastodon's post length limitation. 500 characters. These are for the actual post, for mentions, for hashtags and for the content warning which actually counts into the post length. Doesn't leave you much room to describe an image, does it
But in alt-text, you have 1,500 characters. Per image. Always. No matter how much you want to toot otherwise.
With such a "high" limit, Mastodon users often don't simply add an alt-text of a maximum of 150-200 characters as defined in those many "how do I write alt-text for my static webpage with thousands of characters for explaining stuff" guides. No, they go well into detail. They give in-depth descriptions. They explain stuff. After all, they've got 1,500 characters of room.
Other Mastodon users love that, and they may do it themselves. Yet other Mastodon users see that the aforementioned Mastodon users love it and thus do it themselves.
With image descriptions of 700, 800, 1,000 or more characters, Mastodon users don't even question putting the descriptions into the alt-text. It's simply technologically impossible to do otherwise. So it doesn't even come to their minds. Image description == alt-text.
Okay, now here comes the catch.
First of all, you must never describe or explain something in alt-text that can't be accessed from anywhere else in the image.
That's because some people can't access alt-text, e.g. if they have a physical disability that prevents them from using a computer mouse, so they have to do everything with the keyboard, so they can't hover a mouse cursor over an image which is necessary to see alt-text in a desktop browser.
If they can't access alt-text, they can't access the information in it. If that information is only available in the alt-text but neither in the post text nor in the image, it's lost to them. Such information always goes into the post text.
Mastodon users are likely to be highly irritated upon reading that. How are you supposed to put that in the post text in the Fediverse! Mastodon doesn't give you enough characters! You can only put a detailed image description into the alt-text!
On Mastodon maybe. But only on vanilla Mastodon. Okay, and on Threads if it supports alt-text.
But everything that is not vanilla Mastodon gives you either more characters in the post text than in alt-text or no limits for either.
Even Misskey gives you 1,500 characters in alt-text and 3,000 in the post which is 2,500 more than on Mastodon. That's a thread of at least six posts rolled into one.
The other Forkeys give you 3,000 by default. Pleroma, Akkoma and their other forks give you 5,000 by default.
Friendica, Hubzilla and (streams) don't have any defined limits. Not for the post text and theoretically not for the alt-text either which, as I've explained above, is grafted into the post text and not put into a dedicated text field.
That's plenty of room for description and explanation, depending on how much you go into detail. In fact, you aren't even bound to a 1,500-character limit anymore!
I still expect Mastodon users to be highly irritated whenever I describe images. That's because my alt-text usually only contains a brief image description that doesn't actually describe anything. In addition, it says that the actual, fully detailed image description is in the post which is hidden behind a Hubzilla-style summary/Mastodon-style content warning combination.
I expect someone to try and educate me about all image descriptions always only belonging into alt-text because that's how it's done on Mastodon, yada yada yada. Regardless of the image description of even only one of the images being over 25 times longer than Mastodon's alt-text, so it wouldn't even fit in there.
Mastodon doesn't do it like this, so it's wrong everywhere else as well. Even though Mastodon doesn't do it like this because it can't, and everything else can, and Mastodon can happily display what everything else can.
#Long #LongPost #CWLong #CWLongPost #FediMeta #FediverseMeta #CWFediMeta #CWFediverseMeta #AltText #ImageDescription #ImageDescriptions #ImageDescriptionMeta #CWImageDescriptionMeta
Last updated: 2024-01-23
All text and images © eLoc8.is-fabulous.com