Stefan BohacekQuestion for #blind #VisuallyImpaired folks, what kind of image descriptions do you prefer?
Short, to the point
As detailed as possible
Add a little whimsy!
Show results.
Poll ended at .
Jake in the desert
Christy S@jake4480 @stefan It depends on the image. The way I've always described to them people is, imagine you're talking to someone on the phone with just voice, no video. How would you explain the image so it made sense to them and got the point across? If there is text in the image, putting that text in the description is probably important, either a summary or exact depending on what the image is trying to convey.
Dani 
@stefan wow 😂 70% la última opción. Me sorprende como va de momento
@dMjolnir96 I like to add a "view results" placeholder option to my polls, so that people who are not addressed in them don't end up tilting the results.
Hopefully Mastodon will add a way to show results without having to vote, that would be very useful, and I think some fediverse platforms might even have it already.
⁂67guanacos89llamas181alpacas@stefan I'm no visually impaired, but sometimes I need a little help to fully understand social nuances in images. I've found ALTs useful once in a while.
@337guanacos Yes, it's definitely useful for adding context!
Quinn Comendant@stefan How can I view the answers without voting? (I’m not visually impaired.). Perhaps you can post a screenshot of the results?
@com The "I don't rely on image descriptions" option was intended for that purpose, does that work?
Lee Fife
Yvonne Frindle@stefan @com Oh… see I interpreted that as being an option for _visually impaired_ people to say that they themselves didn't rely on image descriptions, so I would not have chosen it in order to see results. If intended for interested people to simply see results, that could have been more explicit.
@frindley @com Yep, definitely agree, completely my fault here.
But the good news is that this won't be a problem after the next version of Mastodon comes out. https://stefanbohacek.online/@stefan/110849870609110598
Stefan Bohacek (@[email protected])
Attached: 1 image Nice. The next version of Mastodon will have an option to view results of polls without having to vote. The screenshot is from v4.1.5+nightly. #mastodon #nightly #polls
Ned Yeung@com @stefan Interesting. I just use my web browser to access my Mastodon instance (no downloaded app), and I get a "See Results" option without a need to vote. This is what the results are at the moment:
Michael Porter@ned @com @stefan …Hoping that 67% isn’t made up of people who ignored the #blind #VisuallyImpaired tags…
@MichaelPorter @ned @com My bad, that option was intended as a "show results" option, I should have made that clear. Sorry about that!
So did I. I saw, "Question for the impared" and immediately opted myself out of voting. ;)
@MichaelPorter @ned @com A few people pointed that out, definitely an oversight on my end.
So I was pleasantly surprised to learn about this upcoming update to Mastodon just now. https://stefanbohacek.online/@stefan/110849870609110598
Stefan Bohacek (@[email protected])
Attached: 1 image Nice. The next version of Mastodon will have an option to view results of polls without having to vote. The screenshot is from v4.1.5+nightly. #mastodon #nightly #polls
@stefan Funny you posted that. I'm curious not so much about the results, but how many blind people already came around and started to use mastodon. 😉
Here are some related links:
https://saltkrokan.test.rrze.uni-erlangen.de/blog/2022/08/16/der-perfekte-alt-text-interview-mit-blinden-und-sehbehinderten-menschen/
@sbe Thank you for sharing these!
lifewithryan@stefan I would love to know this answer. I’m in awe of how well people describe things on here in the alt text. I have much work to do…
Baron Woke of Carlsberg@lifewithryan @stefan I've just found the more you do it, the better you get. My alt text descriptions generally have far more information than is in the actual post. A lot of the time, as I'm typing, it actually helps me understand the image better.
@SteveJonesnono1 @lifewithryan This is so true.
Back on Twitter, I used to follow a bot that would remind you when you posted an image without a description. Now it's second nature, no reminders necessary!
Kenneth@stefan Camp Whimsy here.
#accessibility #a11y #alttext #imagedescription #blind #VisuallyImpaired
MostlyBlindGamer@stefan I think the distinction between alt text and descriptions is very useful here: short alt text can get the point across and a detailed description can provide access to further relevant information.
@MostlyBlindGamer I'm using these interchangeably, because Mastodon specifically adds image description as an alt text, but this is a good point, thank you!
@stefan yup, it does. It also uses the term “caption,” which might be a good indication that the idea is to be short. There’s nothing wrong with adding “description in reply” though.
Gregory J. Rosmaita@MostlyBlindGamer @stefan The idea that ALT provides a thumbnail by which a visually impaired individual can decide whether to either follow a graphically defined link or query the image for an extended description does NOT apply to social media where there is only one descriptor available that programmatically binds the image to its description. Yes, ALT was originally intended to be a textual thumbnail, but on Mastodon there is no way to link programmatically the description with the actual image.
Note that I keep harping on the term "programmatically". There MUST be a direct relationship between the image & its description, so that an assistive tech can provide the alternative IN PLACE of the inaccessible content in the natural document flow, as the document is being as the document is being rendered accessibly.
(Edited to remove repeated phrases caused by my iPhone's inability to properly transcribe my voice, not all of which I caught before I posted. I will always edit after-the-fact when clarity is in question)
@MostlyBlindGamer @stefan The distinction between ALT and descriptions is a relic of HTML 4.01 and does NOT apply to social media.
HTML 4.01 contains both a terse descriptor—the ALT attribute—to give a verbal thumbnail of the image, & a verbose descriptor—the LONGDESC element—to provide a rich, formatted descriptions which can contain MathML, SVG, & other markup necessary to actually provide an equivalent experience for the blind/VI & those for whom supplemental speech enables them to make sense of a complex image, equation or table.
Mastodon only provides ALT, so it must serve as a hybrid descriptor. A mere caption is NOT sufficient.
You may wonder why I cited HTML 4.01 & not HTML5. Over my formal objection, as well as the formal objection of many others, LONGDESC was deprecated in HTML5.
Of course, deprecating an element does not mean that that element is no longer necessary. Both a terse & a verbose descriptor MUST be part of any ML's native semantics in order for the dyad to function correctly and provide us with the information we need.
ALT in HTML 4.01/XHTML 1.0 was envisioned as a thumbnail, to provide the user with an informed opinion about whether it was worth it or not to follow a link, or to request more detailed information about the image. On Mastodon ALT is the only means to provide an image description that is programmatically bound to the image it describes.
ahimsa@stefan Tagging @oedipusnj - thought you might want to vote and/or add a comment
@ahimsa_pdx @stefan Thanks! Much appreciated, especially as I have not been on social media much recently. 🐱😎
AidanBSo I'm reading this as "To the point, with added whimsy." Perfect.
Your favourite booster@stefan
would love to see the results and improve my ALTs, is it possible to add a "just show the results" option?
would love to see the results and improve my ALTs, is it possible to add a "just show the results" option?
@mustamakkara That was my intention behind the "I don't rely on image descriptions" option, I could've made that more clear, sorry!
@stefan
ah of course, makes sense :)
TBH I (sighted) actually often do read image descriptions as they can be anything from entertaining to actually clarifying or providing context.
ah of course, makes sense :)
TBH I (sighted) actually often do read image descriptions as they can be anything from entertaining to actually clarifying or providing context.
@stefan My challenge is that most of my images are visual art (sketches in different media), and I worry that my descriptions are too long, because I'm trying to convey not just the subject matter/composition but also the medium/artistic technique/treatment.
@frindley Yes, that's definitely a challenge. But I think people appreciate putting in the effort.
Jupiter Rowland@StormyNight Challenge accepted: I have a description for an extremely detailed image which has grown into over 13,000 characters of description, explanation and more than 20 transcriptions. It's my longest one by far, but I still find it lacking. Here is the link to an article with only the image and the description.
Jupiter Rowland
Description of this image as announced in this thread: My avatar in the Metropolis Metaversum, waving good-bye at the on-looker from the Metropolis welcome building before Metropolis shuts down for good. The Metropolis Metaversum, Metropolis or Metro in short, was a virtual 3-D world, also refer
@Yvonne Frindle @Stefan Bohacek Yes, some kinds of images actually require enormous #ImageDescriptions if done properly.
My images are from #VirtualWorlds. They show places which practically nobody is familiar with, which therefore have to be described, and they show lots and lots of things in these places which practically nobody is familiar with, which therefore need even more description.
On top of that, I have to explain everything that the average #Fediverse user is unfamiliar with, and that can be a lot again.
In fact, I don't describe my images by what can be seen in the image itself at the resolution at which it's posted. I describe them by what's actually there within the frame of the image, and I look around in-world rather than in the image.
Also, they say that if there's text in the image, it has to be transcribed. I've extended that: If there's text anywhere in the area shown in the image, it has to be transcribed, regardless of whether it's legible in the image or not. So there could also be quite a number of text transcriptions in my image descriptions.
If an image is extremely detailed, I may put more characters into one #ImageDescription than other users put into all image descriptions of two weeks or even a whole month. But nobody has complained about that yet.
My images are from #VirtualWorlds. They show places which practically nobody is familiar with, which therefore have to be described, and they show lots and lots of things in these places which practically nobody is familiar with, which therefore need even more description.
On top of that, I have to explain everything that the average #Fediverse user is unfamiliar with, and that can be a lot again.
In fact, I don't describe my images by what can be seen in the image itself at the resolution at which it's posted. I describe them by what's actually there within the frame of the image, and I look around in-world rather than in the image.
Also, they say that if there's text in the image, it has to be transcribed. I've extended that: If there's text anywhere in the area shown in the image, it has to be transcribed, regardless of whether it's legible in the image or not. So there could also be quite a number of text transcriptions in my image descriptions.
If an image is extremely detailed, I may put more characters into one #ImageDescription than other users put into all image descriptions of two weeks or even a whole month. But nobody has complained about that yet.
Chris_Swan@stefan I will be interested in this poll!
Ciro Durán 🎏@stefan huh, Mastodon _does_ have a See results link, so you can view the results without skewing votes
@chiguire Not for everyone! https://stefanbohacek.online/@stefan/110849870609110598
Stefan Bohacek (@[email protected])
Attached: 1 image Nice. The next version of Mastodon will have an option to view results of polls without having to vote. The screenshot is from v4.1.5+nightly. #mastodon #nightly #polls
@stefan ah right, me being on mastodon.social which is updated regularly would definitely have it. Good to know!
@stefan I can only speak for myself as a blind user, but my favorite image descriptions that I’ve seen so far are the ones that manage to explain the content of the image in detail while offering context specific to the poster’s tonal intentions. (humor or whit for a meme for example) regardless, I’m just glad to have the chance to actually interact with content.
Niknodder@stefan good question! I always wonder if I’m providing enough or too much detail.
🌕 Chelsea 🌻📚@stefan I was very torn between "as detailed as possible" and "add a little whimsy," but ultimately whimsy won out for me. I've never been a short and to the point person. I want to know what's in the image, and also why you felt it was worth sharing, and also to get a sense of your personality in your descriptions. It's a tall order so I never expect it, but it's nice when I happen upon it.
@greengaybles That makes a lot of sense. Keeping it short not to take up too much of people's time, but add some personality, to make it interesting.
SKUNK 🏳️⚧️@stefan I've been grappling with this as an artist. How much detail is necessary? My art is supposed to be an experience. Can I do a good job conveying that experience to someone who can't see it? Do they just want to know the literal contents of the image or should I add flair and flavor to the description? I'm not visually impaired so I honestly don't know the answer. I try my best to strike a happy medium, and hope it does the job.
@RATFOOD From the responses to this poll, I believe people really appreciate when you put in the effort.
👩🦯The Blind Fraggle@stefan A lot of people voted but hardly anyone answered, lol.
I said as much detail as possible mostly because I figure if you can see it, I should be able to see it too! As in, don't leave information out because you think it's unimportant. I should be the judge of that! But of course in practice it's a bit different, lol. Usually just saying everything in the picture should be enough, no need to say where everything is in relation to each other, for example. Whimsy is also good!
@Fragglemuppet Yeah, the results of this poll were bound to be lopsided, I wish I had the means to only ask blind people directly. But I think this is still very insightful, especially reading the comments.
So thank you for sharing your thoughts!
John Blair@stefan guttural (harde) g, like gif