Follow

RT @kentcdodds
You ๐˜ต๐˜ฉ๐˜ช๐˜ฏ๐˜ฌ it's ๐’ธ๐“Š๐“‰โ„ฏ to ๐˜„๐—ฟ๐—ถ๐˜๐—ฒ your tweets and usernames ๐–™๐–๐–Ž๐–˜ ๐–œ๐–†๐–ž. But have you ๐™ก๐™ž๐™จ๐™ฉ๐™š๐™ฃ๐™š๐™™ to what it ๐˜ด๐˜ฐ๐˜ถ๐˜ฏ๐˜ฅ๐˜ด ๐˜ญ๐˜ช๐˜ฌ๐˜ฆ with assistive technologies like ๐“ฅ๐“ธ๐“ฒ๐“ฌ๐“ฎ๐“ž๐“ฟ๐“ฎ๐“ป?

ยท ยท Moa ยท 6 ยท 87 ยท 70

Here's the thing, the fancy-looking unicode characters almost never create words on their own, especially not those from the web, gemini or IMs, nor the ones for STEM fields.

A rudimentary look ahead "algorithm" can toggle transliteration on and off. We have transliteration mappings in every web CMS that deals with user uploaded files, because they need clean paths for SEO, just take the mappings from Drupal.

#a11y

cc: @lena @aral @devinprater

@walter @lena @devinprater Yeah, sadly, part of the problem is that accessibility tools do little more than the bare minimum. Yet another lovely gift of capitalism where things that donโ€™t generate huge amounts of profit find little investment and love. Itโ€™s so sad that the most effective accessibility pitches are those that aim to convince corporations that improvements will lead to higher profits.

@aral @lena @devinprater @walter to that point, current tools donโ€™t even handle providing any emphasis for bold or italicized text, they pretend that quote marks and parentheses donโ€™t exist.

I would suggest that claiming they do โ€œthe bare minimumโ€ is giving them too much credit given and how many years text to speech has been a usable technology

@masukomi @aral @lena @walter Yes, I hate that too. Although, most blind people don't seem to care. And of course that should be configurable too. But it *should* be an option.

@devinprater @aral @lena @walter

that sounds kinda like saying they donโ€™t care if the text is in blue ink or black. They donโ€™t have the option of telling the difference. Consider โ€œi hate catsโ€ vs โ€œi HATE catsโ€ . Itโ€™s like the difference between โ€œcats are unpleasantโ€ and โ€œcats are the devil!โ€ Thatโ€™s the power of italics.

Not hearing parentheses can make a sentence nonsensical or incredibly confusing

@masukomi @devinprater @aral @lena
I also have opinions about this, lots of them actually, some wrong, but here me out:

We keep using the "screen reader voice" terminology" like it's the one true voice, the one and only. This is B.S., there should be multiple tracks in parallel, something like backing vocalists or backup singers, which can add emphasis to various parts of the text, and sometimes have them take over 1/2

@masukomi @devinprater @aral @lena
In some cases, the extra voices would add emphasis, but in other cases, they would add extra words while the main one is paused, like "opening double quotes", and you would hear this with a 3rd sound in the background, one specific to that phrase, similar to that "Windows XP error" sound we were all used to. In time, you can shorten the phrase from the 2nd voice or completely remove it. Butโ€ฆ

2/3

@masukomi @devinprater @aral @lena
But in the end, from my experience of using screen readers, this is not going to be enough for a great experience.

While trying to find an example here, I remembered a, clearly staged, "impromptu acoustic" performance by a band. If you hear all the voices, you'll get a feel for what's happening, but if you don't see what's in the background, you'll never get the full "picture", pun intended. 3/4

with a YouTube link to Nu-Di-Ty by Kylie Minogue 

@masukomi @devinprater @aral @lena
While the band was playing in the foreground, the people in the background had the most interesting choreography, everybody, even the janitors, were in for the action. And as the C.W. says, it's a bit lewd, but you might wanna get somebody to explain the background action, screen readers aren't gonna help here.

youtube.com/watch?v=zzALnGeJ0t

@masukomi @devinprater @aral @lena, what I'm trying to say is that is, as much as we like to think of computers as a some code heartless machines, our interfaces, the human interfaces can not be.

Not while reading a blog post or while writing code or debugging errors, we need to improve on that. Maybe we can get some marketing budget if we rephrase "accessibility" as "human friendly interfaces".

En Fin or backslash-backslash-zero (\\0) for the all cold ones out there.

6/6

@walter @masukomi @aral @lena Mmmmm, I'd have to hear this to see if it is distracting or not.

@walter @masukomi @aral @lena Having one voice speaking text, and another saying "italics on" and such, yeah that would be too distracting.

@devinprater @masukomi @aral @lena @walter

I'm not sure why you would say that "most blind people don't seem to care." The blind/low vision people I know have explicitly stated that they think it's horrible, but they don't have a choice.

The example video does an excellent job at demonstrating how inaccessible it is. While it'd be great if the tools were better, it's really not difficult to produce more accessible content.

even worse: a lot of TTS engines just... get this right now. all of it. mappings, punctuation, everything. by default. they'd just have to be used by screenreader software.

still, using Unicode fonts is likely a bad idea, precisely because a lot of accessibility tools may suck.

@masukomi @aral @lena @devinprater @walter

@kescher @aral @lena @devinprater @walter

Any idea why screen-readers don't leverage that? I used to work at ๐ŸŽ and spent a lot of time talking with their accessibility folks. They REALLY care, so i feel like there must be some good technical reason why the screen reader's at co's like ๐ŸŽ don't.

@masukomi @kescher @aral @lena @walter Lol the people at Apple can't even use a proper speech queue on MacOS, let alone create something like this.

@LunaDragofelis @masukomi @kescher @aral @lena @walter Basically, when a screen reader is speaking one thing, like a book, and something else comes in, like a notification, you don't really want the notification interupting the book mid-sentence. So a queue is usually made to hold the notification text to send off to the synthesizer after a time or the current text has been spoken. Apple doesn't really have that. And it's one reason why VoiceOver in the Terminal sucks.

@lena Even if these symbols were used for their intended purpose, this would be a terrible way to read them.

@lena all i care about is *emphasis* and having sourcecode rendered as monotype (or anything to make it look different from regular text)

@lena its so ironic that this has no content warning that its bad for screenreaders

Sign in to participate in the conversation
La Quadrature du Net - Mastodon - Media Fรฉdรฉrรฉ

Mamot.fr est un serveur Mastodon francophone, gรฉrรฉ par La Quadrature du Net.