My Experience Playing BGA Totally Blind

[Caplin]

Hi All,

I wanted to write as I've recently started -playing around with BGA in ernest, though I have been a member for quite a while. I am totally blind, that is, have no vision at all, and use a screen reader program. The experience thus far is pretty awful, to be blunt.

Accessibility varies wildly from game to game. I am not confident that I could play fast enough to do real-time with strangers, though turn-based is a little easier if I can figure out what I'm doing. The only way to getgame information is to move the mouse around, which as a screen reader user is profoundly weird. There are sometimes tooltips, though they don't often display information in a format I can understand; some games use a lot of icons which have no text alternative so I have to guess at what is going on.
All in all, I'd say I can sort of play some of the games I've tried, having had most success with Race for the Galaxy. It's far from ideal however, and many games simply don't work very well at all.

Why am I willing to put up with this? I'm a bit of a masochist and I have no other way to play these games online, and in many cases couldn't play them in person either. I think there's a lot of potential in BGA but at the moment it's a struggle which is highly-dependent on the whims of the individual game developers.

I had intended this post to be a bit more positive when I started but it turned into a bit of a rant, and I apologize. I've been enjoying myself quite a bit, but the amount of effort required is a little -ridiculous. I work in accessibility professionally, and there is a lot which could be done to fix some of this stuff, but I haven't done a systematic study of the underlying platform.

Just thought this might be of interest to some people here and possibly spark some discussion. Happy weekend, all.

[RobertBr]

I suspect this is extremely useful to designers of games. I have this problem when I prepare talks online. I can put subtitles on my talks but as I have never used subtitles and do not need them myself I have no real idea what works well and what doesn't.
The same will be true for game designers here. All games have individual designers. Most will have no personal experience of what works and what does not. So if you provide feedback about what works and what does not I am sure that will help.

[katmk]

Thank you for sharing your experience! I hope that it contributes towards improving your, and other people's, access to all these games.

I also believe that when accessibility is paid attention to, it improves the experience for everyone.

[Sjarmtrollet]

I was just thinking about this the other day, thinking it would be hard because so much is mouse-driven. I really wish there were more keyboard shortcuts for actions, with each token having their own letter and such. Would that be helpful?

[Caplin]

Definitely, the more keyboard shortcuts the better. The other major area where things are difficult is identifying what's going on either on the board or with individual game pieces. A lot of games do have tooltips but they often use iconography wich the screen reader does not detect.

For instance, if one looks at Jumpdrive, there's no way to identify the different card types, military vs. exploration vs. income, etc. The game just reads the card info as a few seemingly random numbers. This is a frequent occurance throughout.

The more info which can be communicated in screen-reader-accessible text without having to move the mouse from place to place, the better. I'd love to play, say, Catan or Agricola but it's tricky right now because the board(s) are kind of black holes.

Thanks for the positive response. Apologies for the brain dump

[nik592]

I agree it's great of you to share your experience, and exactly what designers need to be able to build better sites for different levels of accessibility. If you have some general tips that people could employ (such as what you've said about Jump Drive, the iconography, keyboard shortcuts, etc.) and feel like preparing a list, maybe some devs would take that on board (depending what is possible within the framework, as you noted). I imagine it's tricky, much like real-life board gaming. I can't imagine how I would setup a game on a real table to help a blind player understand what was going on - I guess you'd have to just explain everything. But particularly with larger games, positioning is there to help with context, and you obviously lose all of that. Not to mention having to keep so much extra info in your head! I can look at (for example), objectives on offer to remind myself what I might want to work towards, but you've got to ask, or just remember it.

[thoun]

As a game developper, I must admit I'm not really thinking about it when adding a new game. I'm aware of accessibility topic, but I thought the main BGA site was not accessible at all, so I didn't think you would make it to the games! I'll try to describe more the game elements, but if BGA was accessible and would give us guidelines about it, it would probably help a lot.

[robinzig]

As a game developper, I must admit I'm not really thinking about it when adding a new game. I'm aware of accessibility topic, but I thought the main BGA site was not accessible at all, so I didn't think you would make it to the games! I'll try to describe more the game elements, but if BGA was accessible and would give us guidelines about it, it would probably help a lot.

I agree with thoun. Ironically in my day-job as a developer, accessibility (for screen-reader users, and others) is something I'm really passionate about making work, so I've tried to consider it when developing games for BGA (I only have one released so far - one more is in progress but an awful long way off being ready to see the light of day). But unfortunately I reached the conclusion that it's just not possible, and I thought - clearly wrongly! - that users who rely on a screenreader to access the internet simply wouldn't make it into BGA at all, so it wasn't worth the effort. I apologise deeply for this clearly incorrect assumption.

Unfortunately there is no thought whatsoever put into this topic by BGA (at least as far as I can tell), so even if I was determined to make games screenreader-accessible, it would be really hard to do. The biggest difficulty is figuring out what's going on on the gameboard. In almost all games, this is literally an HTML <div> element (a meaningless element that screenreaders will not treat as anything special at all), with a "background-image" that is in fact the game board and conveys all the information for users who can see the screen. Moves are often made by clicking on pieces that are on the board - these too are likely to be plain HTML <div> elements that screenreaders are not going to convey to their users are clickable. They could be made HTML <button>s instead and this would work better for screenreaders - but even then, a blind person has no way of knowing where that piece actually is on the board, and so on.

I don't doubt that it's possible to design an online board-game site where games are fully accessible to screenreader users, but unfortunately I think such a site would have to be built from scratch with this in mind. The BGA framework as it currently is was very much designed with sighted mouse users in mind, it seems to me - and while that's very unfortunate, it's asking an awful lot of individual game developers to try to (partially) fix this. That's not to make an excuse for not maing the effort - after reading this I'll definitely try to make a bit more effort! But it's very much swimming against the tide, and I'm not sure the end results would necessarily be all that accessible, just maybe a little but less in-accessible than the default.

Thanks for posting though, I really appreciate hearing this perspective!

[Caplin]

Yeah, I realize that from a development perspective it would be far easier to work on this if BGA was willing to put in some effort at the framework level. I don't blame developers for not doing the work because realistically, most screen reader users would not find this remotely usable.

Still, it's divs, which while meaningless on a semantic level are at least easier to work with, accessibility-wise, than canvass

I've noticed in some cases that where buttons do exist, they don't have detectable labels. It's probably another case of icons over text, which is understandable but the kind of thing that Aria, for instance, could solve.

I'll try to work on some sort of list but a lot of the main points have already been mentioned.

[thoun]

I've noticed in some cases that where buttons do exist, they don't have detectable labels.

Because frameworks use links shown as button, instead of real buttons. Not a big deal for most users, but clearly misleading for accessibility!