Are there best practices when it comes to punctuation in alt text? For example, are semicolons a hassle for people who rely on screen readers? Is less punctuation better? Trying to use alt texts consistently, but I also want to optimize it for the readers who really rely on it.

@lrhodes Punctuation marks, you say? Let's see:
Use all the semicolons you want; I, for one, love them! I am often too lazy to go looking for the em dashes — there really should be an easy-to-reach key for them! — but those are lovely, too. And ellipses (it took me way too long to learn that ellipsis is singular and ellipses are plural)? Well ... yeah, those are just fine, too.
I'm sure there are a few that I missed, but I think you got the point. As long as it is grammatically correct, you can use all the punctuation you want. 🙂

@Mayana em-dashes are great, but let's not forget about en-dashes either; they're ones who define a range.
Which makes me wonder: does your screen reader pronounce a range produced with a hyphen (chapters 8⁠-12) differently than those that use an en-dash (chapters 8⁠–12)?

@FiXato @lrhodes 8-12 is pronounced as "eight dash twelve", where as 8–12 is pronounced as "eight en dash twelve".
But in your case, there is another symbol before both of your examples, one which isn't recognized by my screen reader. It, at least with the symbol reading level set to most or lower, isn't read when reading by line, but it does mess up the first example, making it read as "eight minus twelve", instead.

@Mayana @FiXato Oh, that's interesting. So the reader translates "dash" into "minus?" Specifically when it's surrounded by numbers, or consistently? So in that case, using the en-dash would actually help eliminate potential confusion, yeah?

@lrhodes Well no. As I said, 8-12 is correctly read as "eight dash twelve".
8 -12, for example, is "eight minus twelve".
And apparently, other symbols, like the one @FiXato used, can trigger that as well. Hmm.
Right. Apparently any character that is not a letter or a number will do that. Odd. 🤷

@Mayana @FiXato I assume the reader is geared to parse whole words, in which case, maybe it reads those characters as spaces when they follow directly after an alphanumeric character? So that it doesn't read "can't" as "can apostrophe t," for example…

@lrhodes @FiXato Well, no. Then it would read can't as can t. It does not. It reads it as cant.
The trouble here is that most of the rules are actually set by the TTS engine, with some additional ones being set by the screen reader. eSpeak and NVDA, in my case. So ... look, I'm not a programmer, I don't know how all this works. 🤷

@lrhodes What I can tell you though is that you really should not overthink this. We've generally been dealing with screen readers for a while. We're used to all sorts of weirdness when it comes to pronunciation, because most of the internet is far from accessible. Unless you start doing really weird things, you are unlikely to do something wrong in your image captions.
Look, we're just happy you're writing them. They're becoming more common on the Fedi, but on the internet in general, almost nobody does. We have low standards. 😀

@Mayana I guess that's what I get for copy pasting an example, rather than looking up just the en-dash glyph. 😳


Sign in to participate in the conversation

Merveilles is a community project aimed at the establishment of new ways of speaking, seeing and organizing information — A culture that seeks augmentation through the arts of engineering and design. A warm welcome to any like-minded people who feel these ideals resonate with them.