LLMs aren't omniscient, and they will never be.
If I make a picture on a sim in an OpenSim-based grid (that's a 3-D virtual world) which has only been started up for the first time 10 minutes ago, and which the WWW knows exactly zilch about, and I feed that picture to an LLM, I do not think the LLM will correctly pinpoint the place where the image was taken. It will not be able to correctly say that the picture was taken at
<Place> on
<Sim> in
<Grid>, and then explain that
<Grid> is a 3-D virtual world, a so-called grid, based on the virtual world server software OpenSimulator, and carry on explaining what OpenSim is, why a grid is called a grid, what a region is and what a sim is. But I can do that.
If there's a sign with three lines of text on it somewhere within the borders of the image, but it's so tiny at the resolution of the image that it's only a few dozen pixels altogether, then no LLM will be able to correctly transcribe the three lines of text verbatim. It probably won't even be able to identify the sign as a sign. But I can do that by reading the sign not in the image, but directly in-world.
By the way: All my original images are from within OpenSim grids. I've probably put more thought into describing images from virtual worlds than anyone. And I've pitted my own hand-written image description against an AI-generated image description of the self-same image
twice. So I guess I know what I'm writing about.
CC:
# # # # # # # # # # # # # # # #That's probably because threads work differently on Mastodon from Hubzilla. Mastodon doesn't know threaded conversations. And Mastodon users only receive messages
- posted by actors that they follow
- boosted (repeated) by actors that they follow
- which mention them
But they do not receive messages that are comments on posts which they've already received.
Let's assume Alice posts something on Mastodon, Bob comments on Mastodon, Carol replies to Bob on Mastodon, Dave replies to Carol on Mastodon and you reply to Dave on Hubzilla.
If this was an all-Hubzilla thread, you'd only mention Dave to show that you're replying to Dave. Your comment goes straight to Alice, and Bob, Carol and Dave pick it up from Alice because they've already got Alice's post on their stream.
But since Alice, Bob, Carol and Dave are on Mastodon, if you only mention Dave, then only Dave will receive and be notified about your reply because you've mentioned Dave. Alice, Bob and Carol will never see your reply because you haven't mentioned them.
So in this constellation, if you want all four to see your reply, you have to mention them all.
Granted, other reasons may be delivery delays on Hubzilla's side, or that enough Mastodon users have muted or blocked you because, from their point of view, you as a Hubzilla user act too disturbingly un-Mastodon-like, and you break Mastodon's unwritten rules left and right.
# # # # # # # # # # # #
Bajando temporalidad
A 70/1 each-way play tops these 4 US Open bets
Cubs 8, Phillies 4: Ian Happs two homers lead a long-ball barrage
In the cases of some Mastodon users, I actually wonder if it's worth telling them a) that the Fediverse is not only Mastodon, b) that I'm on something that's very very much not Mastodon and c) the implications of all this. Especially if they give the impression of wanting the Fediverse to be only Mastodon oh so very much.
Or whether I should simply Superblock them so that they'll never appear on my stream again.
# # # # # # # # # # # # # #Are you referring to my mentions being and rather than what you're used to, namely and Using the long name rather than the short name and keeping the outside the link rather than making it part of the link Likewise, the # being outside the hashtag link rather than being part of it
This is because
I'm not on Mastodon. The Fediverse is not only Mastodon. It has never been. So this is not a toot.
No, really. This is what I post from: , . I ask you: Does this look like Mastodon Have you ever seen Mastodon look like this
Where I am, this style of mentions and hashtags is hard-coded. And it has been since long before Mastodon was even an idea.
I'm on something named
Hubzilla. Hubzilla is not a Mastodon instance. Hubzilla is not a Mastodon fork either. Hubzilla has got absolutely nothing to do with Mastodon at all.
It is its very own project, fully independent from Mastodon (, , ).
Hubzilla has not intruded into "the Mastodon Fediverse" either. The Fediverse is older than Mastodon. And
Hubzilla was there before Mastodon.Hubzilla was launched by in March, 2015, eight months before Mastodon, by renaming and redesigning his own Red Matrix from 2012,
almost four years before Mastodon. And the Red Matrix was a fork of a fork of his own
Friendica, which was launched on July 2nd, 2010, 15 years ago, five and a half years before Mastodon. (, , , )
Friendica was there before Mastodon, too.Here's the official Friendica/Hubzilla timeline on Hubzilla's official website to show you that I'm not making anything up: . Scroll all the way down and notice all the features that you may right now know for a fact that the Fediverse doesn't have, but that Friendica has introduced to the Fediverse 15 years ago, five and a half years before Mastodon was launched.
Again, Mastodon has never been its own network. The Fediverse has never been only Mastodon. When Mastodon was launched in January, 2016, it immediately federated with
- GNU social, the successor to 's StatusNet from 2008 (the actual start of the Fediverse, eight years older than Mastodon)
- 's Friendica, now maintained by and
- 's Hubzilla
Friendica has been formatting mentions and hashtags the way I just did for 15 years now. When Mastodon was launched, Friendica has been formatting them that way for five and a half years already, and Hubzilla has done so for ten months. It is hard-coded there. It is not a user option.
That's because
not everything in the Fediverse is a Twitter clone or Twitter alternative. bFriendica was designed as a Facebook alternative with full-blown long-form blogging capability. And Hubzilla adds even more stuff to this.
This is why Friendica and Hubzilla don't mimic Twitter.Another shocking fact: As you can clearly see here,
Friendica and Hubzilla don't have Mastodon's 500-character limit. Friendica's character limit is 200,000. Hubzilla's character limit is 16,777,215, the maximum length of the database field. And it's deeply engrained in their culture, which is many years older than Mastodon's culture, to not worry about the length of a post exceeding 500 characters.
One more shocking fact:
Friendica has had quote-posts since its very beginning. So has Hubzilla. Both have always been able to quote-post any public Mastodon toot, and they will forever remain able to quote-post any public Mastodon toot. And Mastodon will never be able to do anything against it. (By the way: In 15 years of Friendica, nobody has ever used quote-posts for dogpiling or harassment purposes. Neither Friendica nor Hubzilla is Twitter.)You find this disturbing You think none of this should exist in the Fediverse, even though all this has been in the Fediverse for longer than Mastodon
Then go ahead and block all instances of Friendica and Hubzilla as well as all instances of Mike's later creations, (streams) () from 2021 and Forte () from 2024.
Or you could go ask and as well as of to add every last instance on any of these lists to their blocklists for being "rampantly and unabashedly ableist and xenophobic by design" due to not being and acting and working like Mastodon and just as rampantly and unabashedly refusing to fully adopt and adapt to the Mastodon-centric "Fediverse culture" as defined by fresh Twitter refugees on Mastodon in mid-2022 as well as refusing to abandon their own culture which is disturbingly incompatible with Mastodon's. Essentially try and have four entire Fediverse server applications Fediblocked once and for all because they're so disturbing from a "Fediverse equals Mastodon" point of view.
Or you could go to Mastodon's GitHub repository (), submit a feature request for defederating Mastodon from everything that isn't Mastodon
by design and then go lobbying for support for your feature request.
As for why I have so many hashtags below my comments, here is what they mean. Many of them are meant to trigger filters, including such that automatically hide posts behind content warning buttons, a feature that Mastodon has had since October, 2022, that Friendica has had since July, 2010, and that Hubzilla has had since March, 2015.
- #, # = This post is over 500 characters long. Create a filter for either or both of these hashtags if you don't want to see my or anyone else's long posts.
- #, # = CW: long post (over 500 characters long). Create a filter for either or both of these hashtags if you don't want to see my or anyone else's long posts.
- #, # = This post talks about the Fediverse. Create a filter for either or both of these hashtags if you don't want to see me or anyone talk about the Fediverse.
- #, # = CW: Fediverse meta. Or: CW: Fediverse meta, Fediverse-beyond-Mastodon meta. Or: CW: Fediverse meta, non-Mastodon Fediverse meta. Create a filter for either or both of these hashtags if you don't want to see me or anyone talk about the Fediverse.
- #, #, #: This post talks about the Fediverse not only being Mastodon. Create a filter for either or multiple or all of these hashtags if you don't want to see me or anyone else talk about the Fediverse being more than Mastodon. Otherwise, click or tap any of these hashtags to read more about it in your Fediverse app.
- #: This post talks about the Facebook alternative in the Fediverse named Friendica. Create a filter for it if you don't want to see me or anyone else talk about Friendica. Otherwise, click or tap it to read more about it in your Fediverse app. It is also meant for post discovery.
- #: This post talks about the Swiss army knif of the Fediverse named Hubzilla. Create a filter for it if you don't want to see me or anyone else talk about Hubzilla. Otherwise, click or tap it to read more about it in your Fediverse app. It is also meant for post discovery.
- #, #: This post talks about the Facebook alternative in the Fediverse commonly referred to as (streams). Create a filter for either or both of them if you don't want to see me or anyone else talk about Friendica. Otherwise, click or tap either of them to read more about it in your Fediverse app. It is also meant for post discovery.
- #: This post talks about the Facebook alternative in the Fediverse named Forte. Create a filter for it if you don't want to see me or anyone else talk about Forte. Otherwise, click or tap it to read more about it in your Fediverse app. It is also meant for post discovery.
- # = This post talks about alt-text and/or contains an image with alt-text. It is primarily meant for post discovery.
- # = This post talks about alt-text. Create a filter for this hashtag if you don't want to see me or anyone else talk about alt-text.
- # = CW: alt-text meta. Create a filter for this hashtag if you don't want to see me or anyone else talk about alt-text.
- # = This post talks about image descriptions and/or contains an image with an image description. It is primarily meant for post discovery.
- #, # = This post talks about image descriptions. Create a filter for either of these hashtags if you don't want to see me or anyone else talk about image descriptions.
- # = CW: image description meta. Create a filter for this hashtag if you don't want to see me or anyone else talk about image descriptions.
- #, #, # = This post talks about hashtags. Create a filter for either of these hashtags if you don't want to see me or anyone else talk about hashtags.
- # = CW: hashtag meta. Create a filter for this hashtag if you don't want to see me or anyone else talk about hashtags.
- #, # = This post is talking about character limits. It is primarily meant for post discovery. But if you don't want to see me or anyone else talk about character limits, create a filter for any of these hashtags.
- #, #, #, # = This post talks about quote-posts and/or contains a quote-post. If this disturbs you, create a filter for any of these hashtags.
- #, #, #, #, # = This post talks about quote-posts. Create a filter for either of these hashtags if you don't want to see me or anyone else talk about quote-posts.
- #, # = This post talks about quote-posts. Create a filter for either of these hashtags if you don't want to see me or anyone else talk about quote-posts.
- # = This post is talking about fediblocks. It is primarily meant for post discovery.
Lastly:
Having all hashtags in one line at the very end of a post that only contains hashtags is the preferred way in the Fediverse. For one, hashtags in their own line at the end of the post irritate screen reader users much less than hashtags in the middle of the text.
It's actually hashtags in the middle of the text that are ableist. Besides,
Mastodon is explicitly designed to have a separate hashtag line at the end of the post. A 66/1 shot leads our US Open Specials
What if I transcribe text within my image (for any definition of "text within my image") in a long image description in the post itself which I write in addition to the actual alt-text And the alt-text explicitly mentions the long description at its end E.g. "A more detailed description including explanations and text transcripts can be found in the post."
I often have so many bits of text to transcribe (in addition to describing where in the image they are) that I can't fit them all into the 1,500-character limit for alt-texts that Mastodon, Misskey and their respective forks impose on the whole Fediverse.
I'm not talking about screenshots from social media or something. I'm talking about renderings from 3-D virtual worlds where there may be 20, 30, 40 or more bits of text strewn across the scenery within the borders of the image. The rule says that all text within an image must be transcribed 100% verbatim, and it doesn't explicitly mention any exception, so I do have to transcribe them all. In addition, if they aren't in English, I must additionally translate them as literally as possible. There's no way I can fit all this plus a sufficiently detailed and accurate visual description into 1,500 characters.
But if you (or others) insist that
all text within an image must be transcribed verbatim in the alt-text, and if you sanction image posts that transcribe the texts in the image elsewhere than in the alt-text, then I simply won't be able to post certain images in an appropriate way.
# # # # # # # # # # # # # # # # #
NBA Free Agency: Will the Bucks look to extend Portis long term
Find the latitdue and longitude of any place At , we are committed to providing to help your pets live , , and lives. From and to services, , and , our experienced team offers personalized plans tailored to your pets unique needs.
To know more:
Don't mix up quotes and quote-posts. They're something
very different, not only in technology, but also in use-case and especially in cultural implications.
Whereas on platforms with threaded conversations, quotes are usually used in comments within a thread, and on many platforms, such comments could be removed by the thread owner.
This is a quote. Like in a bulletin-board forum. This is actually only used in comments. For Mastodon users, it's basically unimaginable that this ever happens in social media. Twitter has never had this, so most Mastodon users don't even know the very concept of this. Thus, it is not what they're upset about.
What Mastodon users are so upset about are quote-posts. What they refer to as "quote-posts" or "quote-toots" is what we call "shares" or "shared posts", what Twitter/X calls "quote-tweets", and what is used on Twitter/X for harassment and dogpiling purposes, namely this:
It should be noted that Twitter-style platforms use quote posts in an entirely different way than platforms with threaded conversations.
With Mastodon, someone can quote you, criticize you, and then people dogpile on. Since it is not part of a thread, and is its own top level post, nothing can be done about it.
Whereas on platforms with threaded conversations, quotes are usually used in comments within a thread, and on many platforms, such comments could be removed by the thread owner. Yes, they can create new top level post quoting someone, but that seems to be used less on threaded conversation platforms than on Twitter style platforms.
This creates a different culture surrounding quoting people, since one has potential consequences and one does not.
Plus, I think there is also a cultural difference between people who want to broadcast their thoughts versus people who want to join conversations. People who want to participate in conversations are typically less hostile since they get banned or blocked pretty quickly. People who broadcast their posts just want as many followers to see it as possible and tend to block anyone that disagrees with them. It is a different mindset.
That is why Mastodon has to implement quote controls, but thread conversation platforms do not.
You should see that it's something completely different. This
never happens within the same thread. It wouldn't make sense to quote-post/quote-tweet/share a post in a comment on that same post.
This is what Mastodon users what to have control over. This is what they want to prevent. Entirely. They would want a switch that makes Hubzilla hide the Share button under a post of theirs if they knew that a) Hubzilla exists, and b) Hubzilla can share posts/quote-post.
This is what keeps preaching over and over and over that literally the only possible way to keep this from happening is by not posting in public. For this is what neither Hubzilla nor Mike's own (streams) and Forte have a permission setting for because even their highly advanced and fine-grained permissions systems have no way of implementing actually water-tight quote-post control. So what chance does Mastodon have with its total lack of a permission system and no understanding of Hubzilla's, (streams)', Forte's or even only Friendica's permission system
Mastodon doesn't have either implemented, neither quotes nor shares/quote-posts, at least not beyond displaying quotes properly formatted (it is working on displaying quote-posts properly formatted now). Hubzilla has inherited both from Friendica which has had both for 15 years.
CC:
# # # # # # # # # # # # # # # # # # # # # # # # # # #
Our 5 best bets for the US Open at Oakmont
An iconic landmark in Nairobi
.
.
.
.
.
.
.
.
.
.
MLB scores: Mike Yastrzemskis 3 RBI day was long overdue
How long until Aaron Bummer gets an opportunity to close
" "
" and the ", or "what all works should have been."
It's okay, you don't have to overthink it! Write how you'd describe the image to a friend over the phone.
This only works with simple real-life photos.
If your image shows more obscure stuff (like mine), . (especially Mastodon users: The link goes to a Fediverse post that you may import into your timeline by copying the URL and searching for it.)
# # # # # # # # # # # # You can follow PeerTube accounts from Mastodon.
You seem pretty new to the Fediverse, so this may sound completely unimaginable because it's akin to following a YouTube user from Twitter. But the Fediverse is not a bunch of enclosed decentralised networks that only connect one kind of software. It's one big network in which everything is connected to everything, regardless of server software.
PeerTube is part of the Fediverse. This means that PeerTube is federated with Mastodon.
Consider this: I'm not replying to you from Mastodon either. I'm on something that's basically Facebook meets WordPress meets Google Cloud Services meets Google News with a whole lot more stuff on top. Notice how Twitter isn't in this. What I'm on is
very much not like Twitter or Mastodon. And yet, I can reply to you, and you can read it on Mastodon.
# # # # # # # # # # # # # Bonfire hat einen gewaltigen Vorteil gegenber Hubzilla: Es ist bekannter. Denn es macht effizientere Werbung. Soweit ich wei, hat es nmlich ein eigenes Mastodon-Konto.
Kein Witz: Wenn du ein Fediverse-Projekt richtig gut im Fediverse bewerben willst, dann brauchst du dafr ein Mastodon-Konto. Egal, ob dein Projekt selbst sich problemlos mit Mastodon verbinden kann. Mit einem Mastodon-Konto hast du auf Mastodon sehr viel mehr Reichweite als mit einem Nicht-Mastodon-Fediverse-Konto.
Auerdem macht Bonfire sehr viel mehr Werbung sogar noch als Mike fr (streams) und Forte. Und Hubzilla macht im Grunde immer noch fast gar keine Werbung, weil es zum einen noch zu sehr mit sich selbst beschftigt ist und damit, tageslichttauglich zu werden, und wir zum anderen kaum Leute haben, die das Talent htten, Hubzilla innerhalb des Fediverse anzupreisen, noch dazu auf eine mastodonkompatible Art und Weise.
Bonfire ist Hubzilla auf Wish bestellt. Aber VHS ist ja auch Betamax auf Wish bestellt. Betamax kennt keine Sau, und Hubzilla kennt auch keine Sau.
In sptestens zwei Jahren wird es Leute auf Mastodon geben, die glauben, Hubzilla sei der relativ neue schlechte Versuch, Bonfire abzukupfern.
# # # # # # # # # # # # #
<>
If you talk about non-Mastodon Fediverse server software, always mention that it's part of the Fediverse.
There are people in the Fediverse who read your message, but who don't know the software that you're talking about, and who don't know that it's in the Fediverse. In addition, there are people on Mastodon who are still fully convinced that "Fediverse" refers to an alleged Mastodon network, that the Fediverse is only Mastodon.
Also, always mention that the software you're talking about is connected to Mastodon. If you say it's in the Fediverse, that doesn't automatically imply for everyone that it's connected to Mastodon.
Case in point: Early this year, there were lots of people who were on both Mastodon and Facebook, and who wanted to migrate away from Facebook. So they created Friendica accounts because they had read somewhere that Friendica is "the Facebook alternative in the Fediverse", and they wanted their friends to move to Friendica, too.
Why didn't they invite their friends to Friendica and follow them from Mastodon instead Did they want to move away from Mastodon as well and use something better instead No. Did they want to try Friendica so that they could help their friends with it No.
No, they moved to Friendica because they thought they'd have to in order to be in touch with other Friendica users. After they had learned that the Fediverse is, in fact, not only Mastodon, that there's also Pixelfed and PeerTube and, well, Friendica, they thought that "Fediverse" refers to a bunch of decentralised networks which, however, are completely isolated from one another. They're all totally different things that run totally different software after all. Mastodon users can only follow Mastodon users, Friendica users can only follow Friendica users. Why should it be any different
Well, and then someone told them that, in fact, Mastodon and Friendica are connected. Like, you can follow a Friendica user from Mastodon. That was completely unimaginable to them until they tried following their own Friendica account with their own Mastodon account. And they succeeded. So much to their surprise that they stared at the successfully established connection between a Twitter clone and a Facebook alternative like a deer staring at the high beams of your car at night.
To you, it's only natural that "Fediverse" means everything is connected. But to many Mastodon users, this concept is completely alien until you explicitly mention it to them.
# # # # # # # # # # # Nur da ich nirgendwo ein Zitat sehe (und ich sehe den kompletten Thread auf einem Haufen) und auch keinen Quote-Post (hier auf Hubzilla sind das zwei total verschiedene Sachen, und Hubzilla hat schon immer beides untersttzt das gilt fr die ganze Software-Familie).
Wahrscheinlich wenden Phanpy und Tusky irgendeinen Trick an, der Mastodons Einschrnkungen umgeht, der aber mit etablierten Standards vllig inkompatibel ist. Wahrscheinlich wissen die Entwickler weder von Phanpy noch von Tusky, da es Quote-Posts im Fediverse schon gibt, geschweige denn, wie die gemacht werden. Also hat sich da jemand etwas vllig Neues ausgedacht, das dann nur von ein, zwei Mastodon-Apps untersttzt wird, aber von keiner einzigen Serversoftware, die Quote-Posts eingebaut hat.
# # # # # # # # # # # # # # # # # # # # # # # #I'm very certain that there are people on Mastodon who
- have joined in autumn of 2022 or even earlier
- have only ever used the official Mastodon mobile app, but never a third-party app or even the Web interface
- don't even know that Mastodon has a Web interface in the first place because it's just another phone app to them
- follow a four-digit number of Fediverse actors, all of which are on Mastodon
- are fully convinced that the Fediverse is only Mastodon, and Mastodon is an enclosed network that doesn't connect to anything else
I'm just as convinced that there are people on Mastodon to whom largely the same applies with one difference: They do not only follow Mastodon users. But they think they do. That's because those on Pixelfed, Sharkey, Friendica etc. whom they follow happen to never get caught behaving in ways that Mastodon users may find odd, they never even get caught posting more than 500 characters at once, and they never point out that the Fediverse is, in fact, not only Mastodon. And seriously, if you've never heard of Pixelfed, you may believe that even pixelfed.social is another Mastodon server.
# # # # # # # # # # # # #
Anyone can copy text from anywhere, start a new thread in the Fediverse, copy it in there, mark it as a quote or not, and make fun of the author.
And if that fails, they'll resort to what they've always been doing: screenshots. Not even (streams) and Forte with their advanced permissions systems can keep people from taking screenshots. (Then again, they don't have a quote-post permission setting either because such a thing wouldn't work across the Fediverse anyway.)
In the meantime, Friendica has had quote-posts for a decade and a half, and they've always been used sincerely, believe it or not. Same on Friendica's various descendants.
If the author is not fairly mentioned in the thread, he may never find out about it and be able to defend himselve.
If you're quote-posted from Friendica, Hubzilla, (streams) or Forte, you're automatically notified as if you've been mentioned. I'm not sure about those server apps that have implemented Misskey's way of quote-posting, though.
# # # # # # # # # # # # # # # # # # # # # # # # # # #
Con Chut Mt Lng Mun Thnh Cng Nhng T May Mn Li Khng Ln No P2
Con Chut Mt Lng Mun Thnh Cng Nhng T May Mn Li Khng Ln No P2 Your browser does not support HTML video. Con Chut Mt Lng Mun Thnh Cng Nhng T May Mn Li Khng Ln No P2
New entry of AI-generated and added to our :
Why Did the So
BYU Signs Kevin Young to Long-Term Contract Extension
The Atlanta Braves are mediocre and have been for too long now
I tried to be a chef, but I burnt everything. Now I just sell existential toast.
Hot Blonde with Long Socks Footjob
Das Traurige ist doch: Die allerallermeisten Mastodon-Nutzer "wissen" aktuell, da es im Fediverse keine Quote-Posts gibt, weil Mastodon keine hat, zumal geschtzt immer noch jeder zweite "wei", da "Fediverse" und "Mastodon" dasselbe meinen und das Fediverse nur aus Mastodon besteht.
Die wenigsten werden diese Ankndigung gelesen haben. Von denen, die sie gelesen haben, werden viele "quote posts from other servers and software" nicht verstehen. Was fr andere Server Mastodon kann doch nicht quote-posten. Und was fr andere Software Was heit andere Software Eine andere Handy-App fr Mastodon Oder meinen die Bluesky Kann Bluesky quote-posten Wenn ja, dann gehrt Bridgy Fed sofort gefediblockt!1!!
Das heit brigens auch, da diese Pseudo-Berechtigung, wer die eigenen Trts quote-posten darf, 99% der Mastodon-Nutzer in Sicherheit wiegen drfte. Die sind absolut berzeugt, da sie damit 100% wasserdicht sicher verhindern knnen, da irgendjemand im Fediverse sie quote-postet. Das wird noch zu vollgemachten Hosen und panikbedingten berreaktionen bis hin zu versuchten (oder gar tatschlichen) Fediblocks gegen ganze Nicht-Mastodon-Instanzen fhren.
berhaupt: Mastodon kann Quote-Posts schon jetzt anzeigen. Das konnte es schon 2016. Nur hat es die bisher noch nicht anstndig formatiert. Ich hoffe, es kommt bald nicht nur mit Quote-Posts wie von Misskey klar, sondern auch mit solchen wie von Friendica.
# # # # # # # # # # # # # # # # # # # # # # # #
10 to 12 actors bowed out of the performance where will be in attendance, but they didn't bow out of the Kennedy Center run, as this NPR article incorrectly claims (I read the CNN article they link). Good context for why the tone deaf administration will applaud a revolution against themselves LOL
it's
How long will the Braves carry Rafael Montero
The Mastodon devs are talking as if either the Fediverse is only Mastodon, or the Fediverse as a whole doesn't have quote-posts.
Neither of this is true. The Fediverse has had quote-posts since July 2nd, 2010 when Mistpark (now known as ) was launched. Mastodon toots have been quote-post-able since Mastodon itself was launched, for when Mastodon was launched, it immediately federated with at least two Fediverse server applications that have quote-posts, namely Friendica and , a fork of a fork of Friendica by Friendica's own creator.
Nowadays, at least Pleroma, Akkoma, all other Pleroma forks, Misskey, Calckey, Firefish, Iceshrimp-JS, Iceshrimp.NET, CherryPick, Sharkey, all other Misskey forks, Mitra, Friendica, Hubzilla, (streams) and Forte
can quote-post Mastodon toots with no problem.
And
Mastodon won't be able to stop them. No, seriously, it won't. Not with a non-standard, proprietary, home-brew opt-in or opt-out switch that doesn't tie into anything that the other Fediverse server apps have. And whatever switch Mastodon is working on
will not tie into anything that already exists.
Let me put it this way: Hubzilla has the second-most advanced and fine-grained permissions system in the Fediverse. It goes
well beyond most people's imagination. It works on three levels: (that's similar to a Mastodon account), (that's "followers" in Mastodon lingo, but Hubzilla doesn't distinguish between followers and followed), . (streams) and Forte are the only ones with an even more advanced and fine-grained permissions system.
But even they don't have a quote-post permission setting. And they have permission settings for
just about everything. You want reply control in the Fediverse Hubzilla has reply control, and (streams) and Forte have reply control
on steroids. But what they don't have is a quote-posting permission because that's next to impossible to control across the Fediverse even with the most advanced permissions system.
As (professional software developer for almost half a century, designer of two Fediverse protocols, creator of Friendica and Hubzilla, inventor of nomadic identity, creator and maintainer of (streams) and Forte) says:
The only way to make your posts un-quote-post-able is by not posting in public and not allowing everyone in the Fediverse full access to your posts. Set your "Who can quote" however you want, I'll always be able to quote-post all your public posts with no problem and with no resistance.
So what chance does Mastodon have then Mastodon which doesn't even know what permissions are Developed by Eugen Rochko who actually has a history of head-butting with Mike Macgirvin, and who would
never take
any step towards
anything that Mike has ever developed
I'm commenting from Hubzilla right now, and I'm also on (streams). And I can tell you: If you make any of your posts "un-quote-post-able", this still won't make my Share buttons on Hubzilla and (streams) disappear.
CC:
# # # # # # # # # # # # # # # # # # # # # # # # # # # # # # # # # # # # #
I've seen no mention (perhaps I missed it) of who exactly gets to --> create <-- these "remote quote posts"
"remote" implies no one on my local instance
A common misconception on Mastodon is that the Fediverse doesn't have quote-posts. Anywhere.
As a matter of fact, however:
- Pleroma users can quote-post Mastodon toots.
- Akkoma users can quote-post Mastodon toots.
- Misskey users can quote-post Mastodon toots.
- Calckey users can quote-post Mastodon toots.
- Firefish users can quote-post Mastodon toots.
- Iceshrimp-JS users can quote-post Mastodon toots.
- Iceshrimp.NET users can quote-post Mastodon toots.
- CherryPick users can quote-post Mastodon toots.
- Sharkey users can quote-post Mastodon toots.
- Users on any other Forkey can quote-post Mastodon toots.
- Mitra users can quote-post Mastodon toots.
- Friendica users can quote-post Mastodon toots.
- Hubzilla users can quote-post Mastodon toots.
- (streams) users can quote-post Mastodon toots.
- Forte users can quote-post Mastodon toots.
- Many other non-Mastodon Fediverse users can quote-post Mastodon toots.
None of them have included quote-posts with the explicit intent to harass Mastodon users.
To give you an example: Friendica was launched 15 years ago, more than five years before Mastodon. It already had quote-posts back then. When Mastodon was launched, it was
Mastodon that connected itself to
Friendica and not the other way around. And ever since that very moment, Friendica was able to quote-post Mastodon toots.
If Mastodon adds its own, proprietary, home-brew, non-standard quote-post opt-in or opt-out, ball of the above will still be able to quote-post any public Mastodon toots with zero resistanceb. In fact, they won't even know whether a Mastodon user has opted out of or not opted into being quote-post-able.
To make matters worse, the Fediverse has at least two different quote-post technologies.
Misskey and the various Forkeys put
RE: <URL of the quote-posted message>
into a message, and it's automatically rendered as a quote-post. It references the original, and when the original is edited, so is the quote-post. I don't know whether or not the quote-posted user is notified.
Friendica, Hubzilla, (streams) and Forte generate a dumb copy of the original message from an eight-digit reference number. It comes with a "headline" containing the name of the original poster, profile picture included, and a link to the original message. Since it's a dumb copy, the quote-post will not change when the original is edited. The quote-posted user is notified similarly to being mentioned when being quote-posted.
Immediate, effective counter-measures are completely impossible against both.
CC:
# # # # # # # # # # # # # # # # # # # # # # # # # # # # # # # # # # # # # I have found a few guides for alt-text and image descriptions, but they may contradict each other.
- (this one still mentions SEO, but it doesn't make HTML a requirement, and it's very informative when it comes to how to actually describe an image)
- (this one is geared towards websites and blogs, too, but it goes into detail when it comes to the actual description, too)
- (this has loads of individual pages about alt-text and image descriptions, and the author is visually impaired herself she actually advocates two descriptions for each image, one in the alt-text, one in the text itself)
- (again, website-focused, but again, no HTML, and it includes details like telling you not to mention the race of people in your descriptions and not the gender either if it isn't 100% proven and what to mention instead it also recommends two descriptions for complex images)
- (this one advocates alt-text that's no longer than 100 characters, and it contradicts the above in that alt-text must never repeat what's already in the text itself, but it underlines the rules about race and gender)
- (this is a more technical one it points out that and why alt-text must never include line breaks)
The existing guides on how to write alt-text in social media aren't worth the effort. They don't tell you anything the guides above, at least not beyond walking you through the process of adding alt-text to images in certain social media, step by step. Most of them only cover American corporate social networks and social media (Facebook, Instagram, , LinkedIn). A few add TikTok.
Very few also add Mastodon, but even they only walk you through adding alt-text on Mastodon's standard Web interface. They do not deal with Mastodon's special alt-text culture. They assume that all social networks and social media have either the exact same alt-text culture as websites and blogs or none at all. And literally not a single guide covers anything in the Fediverse that is not Mastodon.
Hence my wiki plans. For one, I want to explain alt-text and image descriptions in the Fediverse as a whole. I won't include step-by-step walkthroughs because I can't possibly know every Web UI and every phone app out there, but I will point out that alt-text doesn't work exactly the same everywhere in the Fediverse as on Mastodon. Besides, I want to take Mastodon's alt-text culture into consideration which is being forced upon the whole rest of the Fediverse. Finally, I want to write guides on certain aspects of describing images and writing alt-text and not only compile the information that's strewn about the Web in lots of individual guides, but also link to these guides as references and point out when they contradict each other.
# # # # # # # # # # # # # # # Just a pity that this one particular guide doesn't really work in the Fediverse.
First of all, the concept of "too long alt-text" or "too detailed alt-text" doesn't exist in the Fediverse, at least not on Mastodon where accessibility standards were defined by overly eager laypeople.
Next, there are no decorative images in Fediverse posts.
Also, only few Fediverse server applications support adding HTML tags to posts. The vast majority of Fediverse users, especially everyone on Mastodon, have a dedicated text entry field for adding alt-texts to image file attachments.
Finally, SEO does not matter in the Fediverse at all.
The whole guide is about alt-text on static websites designed by paid professional Web developers. As opposed to social media users, two out of three of whom can only post plain text.
All this is why I've started putting together a wiki specifically for alt-text and image descriptions in the Fediverse.
CC:
# # # # # # # # # # # # # # # Was, wenn das Bild selbst der Kontext des Posts ist
Was, wenn das Bild etwas zeigt, was Sehende so neugierig macht, da sie auf den Kontext pfeifen und sich das Bild in allen Details ansehen, ob die jetzt fr den Kontext relevant sind oder nicht Es heit doch, da Barrierefreiheit und Inklusion bedeuten, da Nichtsehende immer exakt dieselben Chancen haben mssen, dasselbe zu tun, was auch Sehende tun. Was, wenn jetzt jemand, der blind oder sehbehindert ist, so neugierig auf das Thema und das Bild ist Und wenn das Bild dann auch noch etwas so Unbekanntes, Obskures zeigt, da niemand, der das Bild nicht klar vor Augen hat, auch nur eine grobe Vorstellung hat, wie
irgendetwas auf dem Bild aussieht
Die Gedanken habe ich mir schon vor ziemlich langer Zeit gemacht und meine Beschreibungen daran ausgerichtet. Ich meine, ich poste nicht gerade Katzenfotos.
CC:
# # # # # # # # # # # # Fr mich hinkt der Vergleich mit dem Telefon.
Wenn ich eins meiner Bilder aus einer virtuellen Welt jemandem ber das Telefon beschreiben wollte, knnte ich diejenige Person vorher fragen, was sie ber virtuelle Welten allgemein und ber diese virtuellen Welten ganz speziell schon wei und was ich vorher erklren mu. Und ich knnte fragen, wie detailliert ich das Bild beschreiben soll. Und dann knnte ich mich ganz individuell auf diese eine Person einstellen.
Und wenn ich dann angefangen habe, und der Person wird es dann doch zu detailliert, dann kann sie mich jederzeit unterbrechen.
Wenn ich es aber im Fediverse beschreiben soll, dann mu ich es beschreiben fr mglicherweise Millionen an Fediverse-Nutzern und Milliarden an Internet-Nutzern. Die kann ich nicht vorher fragen, auf die kann ich mich also nicht vorher einstellen. Da mu das, was ich sowieso liefere, sofort passen. Und unterbrechen knnen die mich auch nicht. Selbst wenn sie es knnten, gibt's vielleicht jemanden, der gerade eben nicht will, da ich aufhre, sondern, da ich bis zum Ende weitererklre und weiterbeschreibe.
Das heit: Statt fr nur eine Person mu ich fr eine unbekannte Anzahl an Personen schreiben. Und vor und whrend des Beschreibens gibt es keine Dialogmglichkeit.
CC:
# # # # # # # # # # # # # # Nur was, wenn z. B. ich ein Bild auf eine Art und Weise beschreibe, die ausgerechnet dir nicht pat
Stell dir vor, du bekommst von mir einen Bildpost mit einem Alt-Text, der exakt 1500 Zeichen lang ist. Etwa 1400 davon sind Bildbeschreibung, aber trotzdem ohne Text-Transkripte. Der Rest verweist auf eine weitere, lange Bildbeschreibung im Post selbst. Der Post ist verborgen hinter einer Zusammenfassung mit einer Inhaltswarnung vor einem extrem langen Post. ber 62.000 Zeichen, davon sind ber 60.000 Bildbeschreibung. Du klappst die Zusammenfassung auf, und tatschlich: ber 60.000 Zeichen an Bildbeschreibung.
Wre das fr dich immer noch "besser als nichts" Wre das immer noch in deinem Sinne
So einen Post habe ich im letzten Herbst brigens tatschlich gemacht. Es hat mich zwei volle Tage gedauert, das eine Bild zu beschreiben. Ich kann gern drauf verlinken, allerdings ist der Post fast komplett auf Englisch.
Du sagst: "Beschreibe, was du siehst!" Das habe ich getan.
Aber: Ich sehe mehr als all die Sehenden, die meinen Post sehen. Meine eigenen Bilder sind Renderings aus kaum bekannten virtuellen 3-D-Welten. Ich beschreibe meine Bilder nicht anhand dessen, was ich auf den Bildern sehe, schon gar nicht anhand dessen, was die frs Fediverse eingeschrumpften Bilder noch zeigen. Ich beschreibe sie anhand dessen, was ich in-world sehe.
Auf besagtem Bild steht an zwei Stellen eine Teleportkonsole mit zehn whlbaren Zielen. Die nimmt wahrscheinlich jeweils keine 200 Pixel ein. Kein Sehender knnte sie als Teleportkonsole identifizieren. Ich aber kann sie in allen Details sehen und beschreiben und sogar die 14 Textschnipsel darauf wortwrtlich transkribieren, und ich tue das auch. Wohlgemerkt, auf dem Bild ist gar nicht zu sehen, da auf den Konsolen berhaupt Text drauf ist, weil die so winzig sind. Das einzige, was ich mir gespart habe, ist eine superdetaillierte Beschreibung der Ansicht von oben, die als Karte verwendet wird. Dieses Mal wollte ich aus guten Grnden keine Bilder auf dem Bild in allen Details beschreiben.
Das ist brigens mein normaler Detailgrad bei meinen eigenen Bildern. Die Lnge der Bildbeschreibung variiert allerdings mit dem Detailgrad des Bildes. Portraits arrangiere ich inzwischen so, da es mglichst wenig oder besser noch gar keinen Hintergrund gibt. Aber dann habe ich einen Avatar im Detail zu beschreiben, von den ntigen Erklrungen ganz zu schweigen. Bei dem Bild, das ich oben erwhnt habe, entfallen 40.000 Zeichen alleine auf das eine Gebude im Bild, das kein bichen wie ein reales Gebude aussieht.
Diesen Aufwand wrde ich brigens nicht betreiben, wenn ich keine guten und sogar meines Erachtens ziemlich fundierten Grnde dafr htte.
Ich habe dieses Jahr noch kein einziges Bild gepostet. Ich arbeite seit Ende letzten Jahres immer mal wieder an den Beschreibungen fr eine Reihe von Portraits in verschiedenen Outfits. Jeder Post wird mehrere Bilder mit jeweils mehreren Portraits enthalten, alle vom selben Avatar in derselben Pose, aber in leicht unterschiedlichen Outfits. Die gemeinsame Prambel fr die langen Bildbeschreibungen steht bereits. Aber damit die Leute die Bilder und die Bildbeschreibungen auch verstehen, beginnt sie mit ber 12.000 Zeichen an Erklrungen, bevor irgendetwas Visuelles beschrieben wird.
CC:
# # # # # # # # # # # # #