Learning how to do alternative text for images in coding (so image descriptions basically) and its crazy how wrong ppl do it on here imho. Its supposed to be as little words with as much context as possible. So instead of
[Image description: a white woman with blonde hair is saying: (text)]
Imagining someone couldn't see or see well, they might have watched community and reconize Britta and get the joke. But lady with blond hair means absolutely nothing and takes her out of context.
You also don't need [image description] depending on context. "Image:", "description:", or just jumping into the description is fine.
So a better caption might be;
[Image: Britta from Community: (text)]
I don't want to drag this person in particular or anyone doing image captioning, they're putting effort into making the world more accessible. I just see long captions with no context and SO MUCH unnecessary description all the time it annoys me to no end
So general rules:
As little words as you can
As much context as possible (not description)
Image intro seldom needed
Text within image is most important to translate
