Sébastien Hinderer <sebastien.hinde...@ens-lyon.org> writes: > Samuel Thibault (2021/08/30 18:56 +0200): >> Sébastien Hinderer, le lun. 30 août 2021 17:54:29 +0200, a ecrit: >> > Regarding the counting feature, I find it very clever! But then I'm >> > wondering, why not ocunting tabs directly? >> >> The screen reader cannot actually "see" the tab characters, it only sees >> spaces. > > Ah yes, this of course makes sense. Sorry.
The editor would need to do that. However, note that the fight on tabs vs spaces isn't really being won by any side either. In fact, if I were to aim for primarily speech based coding, I'd look into Emacspeak or similar solutions. You really want structured information when reading code with speech I think. Its not just about identation. You might want to change voices based on syntax highlighting, just to name one example. T.V. Raman really has a point in saying that the artificial boundary between application and screenreader, which is the screen, is a hinderance to good screen reading. If the screen reader had more direct access to the application it is presenting, it could do much more with less heuristics. Which is what Emacspeak demonstrates. There is a learning curve yes, but I guess every sufficiently customizable editor could be used to write a screen reader plugin for. It just appears Emacspeak is about the most complete attempt of that idea. -- CYa, ⡍⠁⠗⠊⠕ _______________________________________________ This message was sent via the BRLTTY mailing list. To post a message, send an e-mail to: BRLTTY@brltty.app For general information, go to: http://brltty.app/mailman/listinfo/brltty