camlorn38

It can, but not with a suitable latency for actual use. What you have to do is send the speech and ersynthesize on the far end. There's an NVDA issue for supporting RDP channels [0], but NVDA Remote made this very low priority and it's potentially blocking on another one to refactor how NVDA handles speech.

0: https://github.com/nvaccess/nvda/issues/3564

I'm pretty sure the NVDA Remote that runs in the browser still exists. You have to have the add-on on the target machine, however.

Windows 10 just works and now we have WSL (i.e. bash i.e. run all your favorite CLI tooling). In general, Microsoft has gotten a lot better about Windows just working. Perhaps Linux has too, but you still had to make sure you had fully Linux-compatible hardware and then do extra steps to get Wi-fi, last time I played with it.

Also, the web browsing experience on Windows is so much better, and the audio stack doesn't fall down at the drop of a hat because you edited a config file wrong (hope you have someone sighted who knows how to unedit it for you). I'm not sure I'd call Linux a stronger foundation; this was not at all my experience with it. OS X is, but then desktop Voiceover sucks to the point where you can't really program with it (basic things like terminal do odd things, nevermind the 10 or so keystrokes needed to navigate from code to the project explorer in Xcode. And we have to mention the speech latency). Then they just killed the function keys, which is an additional problem knocking OS X off the list.

But I think the biggest thing about Windows for me is that it's got synths which are capable of being intelligible upwards of 800 words a minute. Linux didn't even let you get at these settings via Orca last I tried it, and you can't set the inflection either, so it never emphasized punctuation. When your interface is linear and top-to-bottom, the biggest bottleneck in the general case is how fast you can go with the synth, and any platform which significantly cuts this down is therefore not a winner in my book.

But whether or not you agree with my points, I consider it pretty clear-cut that only a blind programmer even has the option of trying Linux in the first place, and certainly not a new one at that. You need too much knowledge to have even a halfway decent experience. In terms of making things accessible and having them matter, you've got to hit Windows first.

Orca is a bad idea. Very few blind people use Linux because Linux desktop sucks, and Orca is always at least somewhat behind all the other screen readers. I have met a few blind people over the years who have used it for an ethical reason or because they can't afford Windows, but no one really likes it. Linux is indeed the cheapest option, but it's also almost completely ineffective in terms of making it actually work well for your users.

For starters, even getting low latency out of the Linux audio stack is a major headache, and the synth situation is abysmal. You can't even touch the config files for these yourself because if you break either--even temporarily--you now can't use the computer. Then you get into how all the graphical desktops have accessibility issues to one degree or another and how you have to use a separate screen reader for anything outside them.

What you want if you want a testing VM that actually has value is Windows and NVDA. NVDA is free in both senses and kind of the industry standard for sighted testing now. Jaws is still more popular, but this is slowly shifting in NVDA's favor. This would be because Jaws costs roughly $1000 per user. The advantage of NVDA is that you can be sure all users can have it, and if it works with NVDA then it's very likely to work with jaws without too much more work.

But sadly you can't just test with one; in the end you have to test with all of them. Things like aria are nice if used correctly, but the aria spec doesn't say much about what screen readers have to do, and no one implements 100% of it. It's very close to the situation with needing to test on multiple browsers.

I'm a blind programmer. I'm currently working on the Rust compiler [0] and a large library for 3D audio that is essentially desktop WebAudio [1]. I'm the kind of person who people often ask for help with their college classes because I went through everything without trouble and came out of college with a 3.9 GPA, and the only reason I'm not making significant amounts of money at the moment is that I have other health problems that I won't go into here (but I would trade with someone who is only blind in a heartbeat). I think I am qualified to say that this is a bad idea.

Firstly, just offhand, the following stacks should be fully accessible with current tools: Node.js, Rust, Python, truly cross-platform C++, Java, Scala, Ruby, PHP, Haskell, and Erlang. If you use any of these, you can work completely from a terminal, access servers via SSH through Cygwin or Bash for Windows, and do editing via an SFTP client (WinSCP works reasonably, at least with NVDA). Notepad++ also makes a perfectly adequate editor, again with NVDA; I'm not sure about jaws if you're using that.

GitHub has a command line tool called hub that can be used to do some things, and is otherwise pretty accessible. Not perfect, but certainly usable enough that NVDA (one of the most popular screen readers) uses it now. Many other project management systems have command line tools as well. If you write alternatives to project management tools, you will have to convince your employer to use them. Replacing these makes you less employable. You need to work to make them accessible, perhaps by getting a job on an accessibility team.

The stacks you are locked out of--primarily anything Microsoft and anything iOS--can only be fixed with collaboration from the companies backing them. Writing a wrapper or alternative to msbuild that can let you do a UWP app without using the GUI is not feasible. I have looked into this. Doing this for Xcode is even worse, because Xcode is a complicated monster that doesn't bother to document anything--Microsoft doesn't document much, but at least gives you some.

I imagine this is not what you want to hear, but separating all the blind people into the corner and requiring custom tools for everything will just put us all out of work. if you're successful, none of the mainstream stuff that cares even a little right now will continue to do so, and you'll end up working on blind-person-only teams at blind-person-only companies.

0: My most notable Rust PR is this monster: https://github.com/rust-lang/rust/pull/36151 1: https://github.com/camlorn/libaudioverse