I'm working on Reset Hard, a time-traveling puzzle game about learning how to do impossible things. One of my goals for the game is to make it extremely blind friendly.
At launch, I want all of the menus and the main interface to be navigable via screenreader. Post launch, I want the main graphics themselves to be navigable via screenreader. I want someone who's blind to be able to download the game and play through the entire thing without any input from a sighted person, at all, at any point.
To accomplish that, the game is essentially a webapp. All of the menus, interfaces, and dialog/options are rendered in semantic HTML. The main graphics are rendered as usual (a big blob of pixels), but my plan is to have a separate pure-HTML representation of those graphics that you can turn on, and some shortcuts for navigation like, "select a door, point my player at it, and then walk towards it."
I plan to distribute the game via
$1, but I have heard mixed reviews about Electron's accessibility controls and about its compatibility with common screenreaders. I'm coming into this without a ton of experience about what works and what doesn't.
So, **some questions:**
1. Is semantic HTML nice to navigate? If you were playing a game, would you want something else, even for menus?
2. Is a screenreader something you would want to use for a game in the first place? Is it better for me to defer to your existing setup, or is it better to build a custom set of controls specific to the game?
3. How fast are you at navigating HTML? If I build an HTML representation of a level and update that at, say, 30fps, should I try to narrow it down or build some kind of activity stream so that the information is easier to parse? What's your preferred way for a webapp to alert you to new information on the screen, especially if those updates might be coming somewhat quickly?
4. I'm doing pretty much everything myself and I don't have any budget to do stuff like voice-acting. This means that menus, interfaces, and dialog are probably going to need to be handled via text-to-speech. Is that a massive downside? Should I be looking into potentially even just voicing menus myself or something?
5. Are there specific communities/sites I should be reaching out to for playtesting and feedback as the game gets farther along in development?