itsmistermoon Nov 23, 2022
Ian.Burnette

If you're like me, then you were really happy to learn about Mastodon's enthusiastic support for image descriptions, and you were eager to join in.

Then you went to actually write something and realized you have no idea how to present visual info in a way that is helpful/enjoyable to those who are #VisuallyImpaired or #Blind.

I found this guide really informative: https://uxdesign.cc/how-to-write-an-image-description-2f30d3bf5546

Post-viral Edit: Don't forget to give the author some love on medium. They did the work!

#Accessibility

How to write an image description

I wrote this how-to guide with the immensely helpful counsel and insights from Bex Leon and Robin Fanning, as well as through an online…

Medium
Nov 23, 2022 at 9:51amWeb
Show threadTikiusNov 23, 2022

@ianburnette This will help too.

https://mastodon.social/@ppatel/109371145817187752
#BDSM #Nudes

Show threadEvidently_ZNov 23, 2022
@ianburnette Thank you, that's a really helpful and informative article.
Show threadLen BeltaNov 23, 2022

@ianburnette

A simple rule is to imagine you're talking to a friend on a phone, see an awesome picture in a newspaper and describe it to them. You would tell them everything that's important, based on context. The things that make this picture matter to you, and what you want others to see as well (the object & context part in the linked article).That's what you do with #AltText.
If you say "Damn, I would struggle in this situation too!": Yeah, that's a thing, you're not alone. We're all different.
It helps a lot to read other people's alt..You even might discover things in pictures you didn't spot yourself.

Show threadIan.BurnetteNov 23, 2022
@Len Yeah, the hardest thing for me is definitely figuring out what context to assume. Different people might follow a hashtag for different reasons. I liked the articles suggestion to start with a short, "literal" description and then add details so that listeners can make a judgement call on whether they want to hear more.
Show threadZinzy Waleson GeeneNov 23, 2022
@Len @ianburnette love this tip! I always remind myself that alt text is like having a secret side conversation with visually-impaired people, and I find that so encouraging and sweet that I try to do better with my descriptions than I ever have.
Show threadAila LeuNov 23, 2022
@Len @ianburnette thank you so much, this tip helps me a lot! I want to add Alt Text to the images I upload but I'm struggling... describing a painting is already difficult... and my idea is to do it in english and catalan (my language). I'm learning but I can see that it will take some time to do it well...
Show threadLen BeltaNov 24, 2022
@ailaleu @ianburnette You're very welcome. 💙
Show threadBronwyn CleeNov 23, 2022
@Len @ianburnette brilliant tips thank you both! Like many have mentioned, I struggle to add appropriate alt text, however I feel more informed now. And I love the notion of it being shaped around a conversation, in parts. #connectionsmatter
Show threadJoan ConcilioNov 23, 2022
@ianburnette An alt text tip I find often overlooked is to try not to rely too heavily on color, which is very subjective even among those without what we traditionally think of as visual processing differences like low vision or color-blindness. There are colors that have meaning regardless of visual perception - so to say a “red light” generally imparts more info than a “traffic light” - but describing someone’s clothing color is usually extra words without a lot of extra value.
Show threadIan.BurnetteNov 23, 2022
@joan This was literally the first problem I ran into that got me thinking about this. I read what I'd written, realized that it was like 20% colors, and thought "hmm, that's probably not very interesting if you've never experienced Blue"
Show threadJoan ConcilioNov 23, 2022
@ianburnette Right?? And then you can really go down a rabbit hole in your head about what words’ meanings we derive from visual info - like I got in my own head about size comparisons and stuff (which I do feel like are useful but it really made me think in a way I never had)!
Show threadIan.BurnetteNov 23, 2022
@joan For sure! I've been wondering about how to caption "hidden twist" images too. Like, "A log floats in a peaceful pond-- just kidding it's actually the tail of a very large alligator!"
Show threadOrla BreslinNov 23, 2022
@joan @ianburnette I post a lot of landscape photos on various social media, both on work & personal accounts. Only realised recently how much I use colour to describe the image 🤦. I'm finding it challenging but fascinating learning new ways of describing.
Show threadmask-wearing, socially distant entityNov 23, 2022
@ianburnette @ryanadams Many thanks! I have such a hard time with this. What is pertinent? What is not necessary? Going to read it now, so I don’t forget.
Show threadjimbodonNov 23, 2022
@ianburnette super helpful, thanks!
Show threadJohn LuskNov 23, 2022

@ianburnette
Mind if I add a tag? Object if so.

#imageDescriptionHowTo

Show threadIan.BurnetteNov 23, 2022
@tarheel Go right ahead (also, go Blue Devils! 😜)
Show threadJohn LuskNov 23, 2022

@ianburnette

😜 lol

Show threadJohn LuskDec 2, 2022
@ianburnette
#imageDescription
Adding another tag somebody else used.
Show threadDrLisaKNov 23, 2022
@ianburnette Thanks for posting this. So helpful.
Show threadwesruvNov 23, 2022
@ianburnette That article looks great! As a pro that has to do these professionally, I really liked this article too:
https://webaim.org/techniques/alttext/
WebAIM: Alternative Text

Show threadFuck Elon MuskNov 25, 2022

@wesruv @ianburnette I'm surprised by this: "we recommend alt="Astronaut Ellen Ochoa"."

That tells nothing other than sighted readers get to see what she looks like. What value is there to have alt text that's essentially a section header?

Show threadIan.BurnetteNov 25, 2022

@MayWeAllRise @wesruv

I think this just supposed to be an example to demonstrate a guiding principal, which is that if the information conveyed by the image is already described in the article, then you shouldn't repeat it in the alt-text because it's boring and just slows the listener down.

I think this guide is a bit more relevant for people writing an article or designing a website rather than captioning a post, but there are still some good tips.

Show threadFuck Elon MuskNov 25, 2022

@ianburnette @wesruv I hear you, but it's not a good example. To provide only her title and name is less helpful than no alt text because it basically conveys, "there's an image but you don't get to know what it looks like."

Better to provide even a very basic description like, "A professional portrait of Astronaut Ellen Ochoa in her orange NASA spacesuit, helmet off." Better still to actually describe her features.

Show thread1505helen❤️Nov 23, 2022
@ianburnette that is really helpful my attempts have been clumsy
Show threadMarine Beaufils 🪡Nov 23, 2022
@ianburnette very informative guide indeed.
Show threadDreadnought HolidayNov 23, 2022

@ianburnette My biggest tip is to simply imagine someone sat next to you asks you what the picture is of? What would your immediate reply be?

That instinctive level of detail is generally correct for your account and audience, and can always be added to if you think it's necessary.

Show threadDreadnought HolidayNov 23, 2022
@ianburnette (also, that guide is one of the better ones I've seen - the emphasis on context and audience is often missing entirely in favour of THIS IS THE ONE TRUE ALT TEXT)
Show threadNiles JohnsonNov 23, 2022

@ianburnette Thanks! I've been looking for this kind of guide for a long time. My other resource is this thread (yes, from twitter, but it's worth the trouble).

https://twitter.com/thingskatedid/status/1512286241928081414

Helpful advice from (I think) somewhere in those replies is that if you can't think of how to describe an image, try to describe how it makes you feel. That seems consistent with the advice in the guide you posted, although that guide is more thorough.

Kate on Twitter

“now you can all see alt text, i'm going to share a thread of some of my favourite image descriptions 💜 this first one is a sequence of photos, it tells a small story from one image to the next https://t.co/sEBxdvTaoi”

Twitter
Show threadKaSaPoNov 23, 2022
@ianburnette Thank you for the link, interesting read. I'm still learning my way to make good descriptions. Ever since I joined Mastodon (so not that long) I add the Alt-Text, not only is it a way to communicate my images to visually impaired people, but it also made me realise I think a bit more about what's actually IN my pictures while adding the description as in: what is the essence.
Show threadTanjaNov 23, 2022
@ianburnette Bookmarked! Thank you for sharing! #Accessibility #AltText #ImageDescription
Show threadAlexander Refsum JenseniusNov 23, 2022
@ianburnette Thanks for sharing. I think the object-action-environment concept makes a lot of sense. Will try it on my #soundactions recordings.
Show threadRyan SchultzNov 23, 2022
@ianburnette Boosting and bookmarking this resource, thank you!
Show threadCatEyeJayneNov 23, 2022
@ianburnette THIS was super helpful! I want to participate in #AltText #BeKind to #LowVisionUsers to be more #Inclusive Thank you for posting this. Please #Boost this post #DogsOfMastodon #CatsOfMastodon #ZSHQ
Show threadRosieNov 23, 2022
@ianburnette thank you, I have learned so much from that article.
Show threadGaysisNov 23, 2022
@ianburnette I don't really post images here, but this will be very useful for the websites I'm creating, so thanks!
Show threadanne cunninghamNov 23, 2022
@ianburnette Thank you for the guide.
Show threadSkykristal | commissions open!Nov 23, 2022
@ianburnette this is something I never really did or thought about. Very informative, thank you!
Show threadKelly LepoNov 23, 2022

@ianburnette
This is a great guide. Some other ones that I have found useful:

Cooper Hewitt, Smithsonian Design Museum guide for image descriptions:
https://www.cooperhewitt.org/cooper-hewitt-guidelines-for-image-description/

Jake Archibald, Writing great alt text: Emotion matters
https://jakearchibald.com/2021/great-alt-text/

Léonie Watson, Thoughts on skin tone and text descriptions: https://tink.uk/thoughts-on-skin-tone-and-text-descriptions.md-notes-on-synthetic-speech/

Neiman Labs, “Space is for everyone”: Meet the scientists trying to put otherworldly images into words: https://www.niemanlab.org/2022/08/space-is-for-everyone-meet-the-scientists-trying-to-put-otherworldly-images-into-words/

Cooper Hewitt Guidelines for Image Description | Cooper Hewitt, Smithsonian Design Museum

On Striving for Digital Inclusion Museums provide robust content for people to interact with across digital platforms. As cultural organizations continue to develop more advanced experiences, it is essential that they consider all audiences during the creation of digital resources and tools. Digital accessibility ensures that people with disabilities have access to our online collections,

Cooper Hewitt Smithsonian Design Museum
Show threadJohn M FloresNov 23, 2022

@kellylepo @ianburnette Thanks for this great conversation and great links. A bunch of us follow the #ALT4me hashtag which people use when they encounter an image without a description and want assistance. We also use #ALT4you when we encounter an image without a description and add one.

We started using the hashtags on the bird site and had some success but the attitude and enthusiasm here is 100 times better!

Show threadDragon-sided DNov 24, 2022

@kellylepo @ianburnette Seems to me Mastodon should just implement an im2text plugin and automate the process.

https://github.com/OpenNMT/Im2Text

GitHub - OpenNMT/Im2Text: Im2Text extension to OpenNMT

Im2Text extension to OpenNMT. Contribute to OpenNMT/Im2Text development by creating an account on GitHub.

GitHub
Show threadKelly LepoNov 24, 2022

@dragonsidedd @ianburnette

For anything more complex than OCR, automatically generated image descriptions tend to be pretty terrible, and frequently wrong. They are better than nothing, but humans are much better than computers at describing images.

https://www.nytimes.com/interactive/2022/02/18/arts/alt-text-images-descriptions.html?unlocked_article_code=lprX1aUSOZHYylVssPc62Q2kcIemLm6FxrET8fEVJno9lKndvGn6mn4R6BnUCdzXf8ScQw3azfUzrgoE27R2x36Zp5Gdt4hiDuHExPm7djxCsN6pCeVhWuGOUQrvMgGKVPw1Yj4RMPIhvK_w6i3p5OBPZjSchYnkdc-FPspnbSwOvUP9nN8hSUbhLyLpckpcZ2t-SnrTaYVQYTVtfnVzRSc_XG_P-t0gzscT1t_f2WWCyT6nVoTvJOOCm2j0Bp9MeP9KQwqjXyGxBNG7-qbfXM_H7-ZMHYbS2V00Am22n81BA3oMndUmtSo_i8DIMzE2d-XwnpI4gUmh4T0yM20hnrwYZ9aqOHZtfu6SwFkt-JZ07N0&smid=share-url

The Hidden Image Descriptions Making the Internet Accessible

The challenge of describing every image on the internet, and the people who are trying.

The New York Times
Show threadDragon-sided DNov 24, 2022

@kellylepo @ianburnette The article seems to have two concerns:

1) AI back in 2016 did a terrible job. This is unsurprising. I'd be interested in specifically reviews of im2text-based models.

2) Training sets have biases. This is also unsurprising; all content produced by humans has biases. We know ways to reduce bias in training sets.

The solution is definitely not to make people work harder on each post; the solution is to automate that work away.

Show threadIan.BurnetteNov 24, 2022

@dragonsidedd @kellylepo

I'm not opposed to automation, and it should definitely be used to help fill in gaps, but I don't think it's going to fully substitute human interaction. I live in the Netherlands, and I still only speak very basic Dutch. AI translation makes a lot of things easier for me, but I really can't interact that much with non English-speakers.

Besides, this is a great opportunity to help build a better training dataset for alt-text generation.

Show threadThe Spoonless KitchenNov 26, 2022
@kellylepo These are great articles, thank you so much!
Show threadJames HavishamNov 23, 2022
@ianburnette Thanks from me for the link too. I was wondering how to describe photos and the obvious thing like colours and bright, dark seem challenging. Also what’s important to describe and what not. Now I’m going deeper into the guide. I hope it will help.
Show threadNullNoMoreNov 23, 2022
@ianburnette - I appreciate this thread and responses. Question: I often identify the format of the image (photo, screenshot, drawing) when using alt text. Is this redundant or useful?
Show threadIan.BurnetteNov 23, 2022

The T.701.11 guidelines that someone else posted has this to say:
"Identifying the type of an image provides a quick impression regarding the purpose and possible
contents of the image. People looking at an image use this information to give them a "first
impression" of what to look for when looking at the image."

So it seems like you've been doing good!

*Edit* Here's the aforementioned post: https://fosstodon.org/@asadotzler/109394093355591957

Asa Dotzler (@[email protected])

@[email protected] I'm a new fan of the ITU guidance at (PDF) https://www.itu.int/rec/dologin_pub.asp?lang=e&id=T-REC-T.701.11-202009-I!!PDF-E&type=items See sections 7-14 to avoid the parts you probably already know. The structure and rigor this provides and allows seems to be missing from many other approaches.

Fosstodon
Show threadBob | Ogdimora (Metal & Cats)Nov 23, 2022
@ianburnette thanks for sharing that. It's really helpful. I've actually really been enjoying adding descriptions to my photos. It makes me think more about the photo, and all that comprises it, and I get to enjoy that moment in time that I captured all over again. 😀
Show threadShield MaidenNov 23, 2022
@ianburnette Guilty of not doing it...useful info. Realise it's similar to assessments I had to do on first meeting a poorly patient strangely enough. Must try harder.
Show threadMiguel Guhlin 🦉 🇵🇦 🇺🇲Nov 23, 2022
@ianburnette @rickmans Thanks!
Show threadBokonon_LivesNov 23, 2022

@ianburnette This was very detailed and informative - thanks for sharing this insight.

It stood out to me that for social media, the author suggested adding details to the caption and not necessarily to the image alt text. Does that mean, for example, that they prefer adding detail to the main text of a toot, tweet, etc.? Or am I misunderstanding?

For visually-impaired members of Mastodon, is it still best to use the built-in image description feature that Mastodon provides?

Show threadIan.BurnetteNov 23, 2022

@bokonon_lives I've seen a lot of people with VI/Blindness requesting the built in alt-text/description. Keep in mind that you have a 1500 character limit for the built-in description, way more than the post text.

But again, I'm not an expert. If you want to help out just keep following #Accessibility. You can also follow # ALT4me (not using the hashtag because I don't want to pollute the feed) which people use to request an alt-text when the come across an undescribed image.

Show threadBokonon_LivesNov 24, 2022
@ianburnette Thank you!
Show threadNoelleNov 23, 2022
@ianburnette Thank you so much!🙏🏽 That really helps
Show threadZilla BlitzNov 23, 2022
@ianburnette Oh, much cool! Thanks for sharing. :)
Show threadAlison DomakinNov 23, 2022
@ianburnette @dyfrig may be useful for your meeting next week? And also for me!
Show threadDyfrig WilliamsNov 24, 2022
@AlisonDomakin @ianburnette Thanks Alison, this is really helpful!
Show threadNeeledgamNov 23, 2022

@ianburnette As someone who reads books for the blind/VI (volunteer job), and also books that come with pictures, graphs and maps, I've been told that you start with the big picture, then zoom in on the details.

So for instance: the painting called Mona Lisa shows a woman, set against a landscape. Only the top half of the woman is showing in the painting. She wears such & such, her hair is like this and that, and she has a very mysterious smile. Her eyes seem to follow you everywhere.