Transcription A Bluesky post from "Slippy", @damnslippy.slippy.me, with a profile picture of a woman with short, purple hair holding a knife: Sincerely delighted to discover, 45 minutes into this nearly-wordless three-hour documentary about French monks who take vows of silence, that among the reasons they \\\can\\\ talk is "to make sure the monastery cats know when it's mealtime by making little kitty-calling noises at them." :::
I blame the tools. On Lemmy, unless you are using Tesseract, you typically cannot see the alt text unless the image fails to load or you have a specific tool to do so like an actual reader for the blind. For at least the past year I have always put alternative text onto every image I have posted, but using Firefox on Android and I have no easy way to even know if that text is there, e.g. to check spelling. In PieFed this is an active area of development, and e.g. alt text was added 9 months ago to the post creation page (roadmap) but I don’t know if Lemmy has any plans to ever show alt text to someone who is not blind (or has some other means to view that text).
And even if it did, not all apps may make use of it. Therefore if these prolific posters can’t even see it themselves after they post something, I can well understand why they would feel like it is not worth the bother. The incentivization structure is just all wrong: it’s all cost to them and virtually no benefit, at least none that they can see for themselves, immediately. Someone would need to put in the requisite time and effort to make better tools, if we truly wanted this to change. However, I personally have given up on Lemmy every getting better at a reasonable rather than snail’s pace.
And similarly Firefox as well. If the developers choose not to care about such issues, it sends a strong message to the users of those tools that the concept is unimportant.
Something that people could do regardless that is an even better workaround than writing up alt text into image posts would be to copy and paste text rather than use screenshots. This offers so many advantages, including the ability to change font sizes and match whatever dark/light mode the user prefers at that moment, and be represented in whatever font face the user chooses, etc.
Personally I often use the body to add a transcription (not to different from what I put in my comment above), rather than use true “alt text”. The main reason is that I often want to write more than the amount that feels appropriate for an alt text (including having paragraphs), but the visibility of it is an added advantage.
It honestly shouldn’t be that hard. Just use both alt and title text properties, the latter of which shows up on hover, and on mobile at the top of the OS’s default long press menu.
But I do agree that the tools should be made to assist. Mastodon basically yells at users to add alt text. Pixelfed already does my suggestion of mirroring the alt text to the title text.