Unity: Not Quite So...Unified?

True, I have noticed it initially. But I am not tempting to read through extensive documents however, to search for information, if there are any tools, designed for use, by impairment people. Hence maybe there are as well.

In the end, adobe allows install plugins also. So something somewhere may be out there.

 

Okay, I just caught up on this thread, and there are a few things I will respond to.
First, to everyone criticizing the formatting of this post, I fail to see why the aesthetics are so important to you as long as the content can be understood.
Secondly, thanks to Murgilod and others for understanding what I am saying in spite of the anger. The reason why I am angry is that Unity has known for at least 4 years that blind people have tried to use this product, and have done nothing. I feel that some anger at an entire group of people being ignored for no reason is justified; this isn't just a Unity problem, but one that is pervasive industry-wide.
Lastly, people have asked what specific improvements I believe could be made to make the program more accessible to screen readers. A huge step toward API accessibility would be to add labels, alt tags, and/or descriptive text to graphical controls and menus so that screen readers can navigate them. The reason that the screen reader cannot effectively navigate the UI is that it's a picture, a graphical image of text or other things. So make sure that if you have graphical text for a button, combo box, or any other control, you add a label or alt tag and descriptive text that gets passed to the screen reader, so that we know what the control is.
The second thing would be to make sure that keyboard support can accomplish the same things that a mouse can. Allow me to use keyboard input to make choices or modifications, since using a mouse is not an option for me. Just those two steps would be a massive improvement.

 

It might help to put Unity3D in context. It was always aimed at the 3D market (2D support wasn't added until later). Particle systems (two versions), terrain, shaders, 2 animation systems, fog and LOD... . There are still lots of game engines. Unity is aimed towards the highly visual end. Others are specialized for things like choose-your-adventure story games. It seems like those would work better for the sightless, and they're still popular.

It supports pure programming, which is perfectly doable by the completely blind. But the goal has always been moving to visual systems like Mechanim.

You could argue that Unity is now the market leader, so has extra responsibilities, but that only applies if laws were relaxed to let you be a natural monopoly, or you gets lots of generous government work, like IBM or Microsoft.

 

This is deliciously ironic. Just like you find it difficult to use the software in question thanks to a lack of accessibility options, some people find it difficult to read a post without formatting. Good formatting makes a massive difference to readability and asking people to just live with it is not too dissimilar to asking visually impaired people to live with no accessibility.

 

Just responding to this part:

It isn't about aesthetics. Large walls of text are difficult to read visually. I think it has to do with moving your eyes from the end of a line to the beginning of the next line. It is easy to go to the correct next line when there are 4 lines in a paragraph. Not so much when there are 50 lines.

Think of it like if you were using screen reader and every few sentences it started saying sentences out of order, or chopping and mixing them up together. It becomes quickly very hard to follow the point the author is making.

 

Presentation is really important to basically everything. It impacts things ranging from initial impressions (which make a huge difference to the opinions people form and the actions they take) to how easy something is to understand to the amount of attention people will give to something.

You're probably right that it shouldn't make a difference but, rational or not, to most humans it makes a huge difference.

Nothing has made it into the Editor yet, and it also frustrates me when there are known issues which appear simple which they don't do anything about for years. So I understand your anger there, particularly considering the issue at hand.

However, as has already been pointed out, this doesn't mean that they're doing nothing, and it doesn't mean that you're being ignored. They are working on a revamp for the Editor interface, which is a prerequisite issue to handling things like screen reader accessibility and flexible editor styling and things like that. This is also fundamental to basically all of the Editor's workflows, so they're taking the time to get it right, and it's simply not done yet.

If it finally comes out and doesn't support or provide hooks for improved accessibility then I'd agree that you've been ignored.

As inconvenient as that post was for me to read, there's a world of difference between "inconvenient" and "inaccessible".

 

Agreed, but I have known people that would literally be unable to read the text they've written due to formatting, and I have had major problems myself in the past with walls of text.

 

I would fully would understood, if you stop at this part of the sentence.
But since you decided to continue

Sorry, but this is very hypocritical.

Formatting has high importance in a way, how information can be conveyed. It helps grab important information and jump into right part of the content.

If you use reader and read text, which has no headers and punctuation, spaces, paragraphs etc., then enjoy getting context correctly. If you know what I mean.

 

True but, unlike Unity's Editor widgets, a screen reader could help with that.

Sure, but does that mean we should ignore anyone with less than 20/20 vision just because Unity happens to have a Scene view?

For example, my current game involves creating a lot of puzzles, level block-ins, that kind of thing. Yes, that's an "inherently visual task", but there's no reason someone with imperfect vision couldn't do them, because at that stage it's mostly snapping large objects to grids and hooking up properties to one another. In that context, the Scene view is less of an issue than the Inspector.

Then there's lots of things like objectives, tutorial messages, and audio cues to hook up. Making games generally involves a whole lot of "data integration" in general, much of which isn't inherently visual at all.

See my previous post. Telling people to go use other tools because they're different from us seems practical at a glance, but does not bear scrutiny as soon as you start thinking longer term.

For the vast majority of people, if they want to work in games, simulation, etc. it's common advice to grab industry standard tools and start learning to use them. If those tools don't have reasonable levels of accessibility then both individuals and the industry in general are missing out on opportunities.

 

Arrg!!!! Paragraphs are not formatting. They're as much a part of writing as sentences. Each paragraph expresses an idea. I'm sure if you think back you can remember the 4th grade "topic sentence and 3 supporting sentences" heuristic, and not the non-existent rule "indent every now and then for variety".

A wall of text is bad because you can't tell one idea from the next. It leads you to believe there aren't distinct ideas; maybe it's just bullet points smashed together. The author probably didn't take the time to organize thoughts, and hasn't tried to explain them, or else there would have been paragraphs.

IndomH's second post was perfect use of paragraphs. They aren't indented and don't have blank lines between, but that's not important. _Those_ are aesthetics and formatting.

 

Well I guess the formatting was the ultimate argument to fully convey why accessibility is important a bridge has been made so you can understand each others

 

I found Joe-Censored's response to be very interesting from a neurological standpoint (I wonder if it has something to do with the way the visual cortex interprets), and the most helpful. I wasn't being sarcastic when I asked that question. I truly wanted to know. For me, formatting doesn't matter. I don't care whether there are 4 lines or 400 in a paragraph, what color the text is, or whether something is indented, as long as my screen reader can voice the words and I can understand them. For me, it's about content of the message.

That being said, one of my pet peeves related to posts on fora and blogs is the utter disregard I see for grammar, spelling, and punctuation. You would not believe how often I see posts with ridiculous spelling errors, run-on sentences, subjects and verbs that don't agree, and punctuation entirely left out of the equation. That is worse to me than any type of formatting issue. However, I do still manage to interpret and understand the message conveyed.

I did note that all of the responses received dealt with my question about formatting, but none addressed the second part of that post, where I discussed implementation of some specific improvements, i.e., labels, alt tags, keyboard support, etc.

 

Well unity has been terrible on that, even with visually able people. For example if you have a 4k screens, the letter are so small you are doomed to decrease the resolution. At their core, they still have legacy garbage they have to deal with, resulting from early days decisions when they were just an internal engine, all while having to progress at the same time. And they have been struggling to both catch up to that, and be at the state of art at the same time. They are slowly getting rid of that legacy, while maintaining as much as possible a continuous, compatibility, to not break people project suddenly.

There has been a huge rolling out of deprecated features recently, aiming at transitioning out this. They are still held back by some. For example the text resolution was hard coded so deep in the core, they couldn't change it without breaking important features. They have know that for a while, and have been working on it. Idea like ECS DOTS are basically a way to soft reboot the core and rebuild the engine on top. I know text is coming, they said so, but we don't actually know how hard it will be to do so.

 

Thank you for formatting your post.

 

One issue is this forum is an English language forum, yet a large number of its users have learned English as a second language. Yes they can be difficult to understand, but I try to remember they are generally making their best effort. The worst are the posts that obviously were just fed into Google translate. They are often easy to pick out because they will use synonyms for words that are proper names (referring to Unity as Unify or such nonsense).

Labels and Alt tags seem like the low hanging fruit Unity could jump on, though I don't know if there are any technical blockers as to why they aren't in there already.

The keyboard support is something I think would be most exciting though, because while it helps the visually impaired, it also helps virtually everyone. Not that doing so for solely the visually impaired isn't something worth considering by Unity. Just that a feature with wide appeal is going to be more likely to get implemented, maintained, and done right than a feature for a niche audience.

As an old school DOS user I've always favored interfaces with good keyboard support. I think this is largely a lost skill on the Millennials who have always had point and click interactions with computers. (those that didn't become Linux fans at least)

 

This is something that should have been addressed from day one. If you are going to have graphical text in lieu of real text, or even worse, you choose to only have pictures of actions to convey the use of a control, then labeling and alt tags should be a no-brainer.

Allowing keyboard input should be a no-brainer as well.

 

The keyboard support has improved. A few updates ago they finally let Enter and Arrows work on enum dropdowns in the Inspector (I remember since I have a Q about it from before that, and it was definitely impossible then).

Traditionally, specialized technical products like Unity don't have lots of tooltips. You're expected to read the docs and become an expert, and the settings tend to be complex. Pick a setting in the particle system. Very few can be explained in a few words.

 

I'm pretty sure that they would if they were starting Unity today. But they started it in 2002 and they did it mainly for themselves (although they thought they will eventually license out the editor), so it wasn't a big concern of theirs.
And since then they were struggling with the IMGUI, it's very hard to replace it without breaking everything.
So they're doing it step by step as they can and I think they do a great work so far (see the new UI system in the 2019 versions).
It's not there yet, but a good first step.
Also you can find the keyboard shortcut editor, you can assign any keystroke to any command, so do whatever you want with it, you can set up a fully keyboard-enabled editor. Also in the 2019 lineup.

Well, I'm sorry, this is an international forum, a lot of us speak English as a second language. My grammar is not as good as I want it to be and it's not as good as on my mother tongue, but I'm trying, I'm so sorry that it does not reach your first class expectations. In exchange I speak more than one languages.

 

English is quite hard language to learn. In fact its grammar is quite a mess on its own. But as pointed already, yes we have multinational community, with different English writing and speaking skills. What is common however between many languages, is formatting and punctuation.

But I can imagine, grammar is much more critical for auto reader, than for someone, who can read directly.

 

But these things have been broadly addressed prior to that post, with talk about early oversights making them hard to address without major changes which are already in progress.

Building on what @Lurking-Ninja has already said, this is really easy to complain about in hindsight. It's a no-brainer now precisely because tools like Unity (and competitors) becoming generally more accessible* has highlighted the cases where they're still not accessible.

Also, while screen readers are provided by OSes as standard now, was that the case back when Unity was first made? We can hardly be upset at them for not supporting a thing if it didn't exist and/or wasn't standardised at the time. Were screen readers a standard thing back then?

That said, as I already mentioned, I do strongly agree it's something they should work on as soon as they can.

I wouldn't be too upset if I were you. You write better English than many native English speakers.

* Before Unity's time, the main issues preventing access to industry-standard game engines for most people were 1. the price (often being hundreds of thousands of dollars) and 2. being much harder to use in general.

 

Yes and no. Windows XP came with Microsoft Narrator but it was only capable of reading text in a few applications that came standard with the OS (Windows Setup, Control Panel programs, Windows desktop, Internet Explorer, Notepad, and Wordpad). Anything more than that needed a third party application.

https://www.geeksinphoenix.com/blog/post/2009/06/07/Read-text-aloud-in-Windows-XP-with-Narrator.aspx

Only just barely before the public release of Unity. Apple's VoiceOver was released with OS X 10.4 which was released in 2005 and Linux's Orca was released in 2006 replacing the previous solution Gnopernicus which was released in 2001. I believe Windows Vista is when Narrator gained support for other applications which would have been 2006.

https://en.wikipedia.org/wiki/VoiceOver
https://en.wikipedia.org/wiki/Orca_(assistive_technology)
https://en.wikipedia.org/wiki/Gnopernicus

 

Where is a TL;DR when you need it?

 

12-ish years ago we had JAWS to read many web-pages, and it was known to use alt-tags on IMG elements - mostly for low-bandwidth that couldn't show pictures, but also for the blind.

Don't most game developers use Scrums? Because we know everything is something we need to work on as soon as we can, but they can't all be. We learn to write it on a card, put everything else we want added to Unity on more cards, and sort them. And many things we need to work on as soon as possible go on the end and get cut.

 

I wasn't referring to any one person's posts here, and I certainly wasn't referring to persons who speak and write English as a secondary language. In fact, I was referring to those people who are American, live in the United States, and speak and write English as their primary language. For those people, there is no excuse for the grammatical garbage that I've come across.

Yes, they were. The first version of Jaws for DOS was released in 1989. The first version for Windows was released January 1995, and a new version for Windows has been released about once a year thereafter. Screen readers arrived long before the beginning of Unity. For information on Jaws for Windows, please see:
https://www.freedomscientific.com

Incorrect. Please see my above reference to both Jaws for Dos and Jaws for Windows, which by that time had been going strong for a decade, if Unity was released in 2005.

 

JAWS is proprietary though and while the non-commercial home license is affordable (only $90/yr) the commercial license is far less so (one-time $1,200). Incidentally these are the current costs of the software. Back in 2007 the cheapest version (linked below) was $900. Searching further back shows that this was the price for a very long time.

https://web.archive.org/web/2007061...fic.com/ProductInfo.aspx?productid=340014-001

His point wasn't that there weren't screen readers available at the time. His point was that they weren't important enough to be included with the OS. Setting a standard requires more than just making your program available. You need to make them widespread enough that people are going to start supporting it with their apps.

Up until screen readers became a main accessibility feature of an OS support for them was basically an afterthought and Unity being the age that it is was one of those programs where it wasn't important enough to have support from day one.

 

TL;DR: Unity should be screen reader accessible, although a lot of the posters in this thread find ways to find this unreasonable.

 

Indeed, I asked if they were standard, and whether they were provided by the OS.

That said, I wouldn't say that accessibility features "weren't important enough". They were important, but back then it wasn't seen as an OS's responsibility to do that kind of thing (Remember, there was once a time when games had to provide their own support for different bits of hardware!) They needed to grow into it, and eventually they did.

Hopefully the same happens with Unity.

 

What you don't realize is that even now, now that standardization, as it were, has occurred, screen reader support is still an afterthought in 99% of cases, and is still not considered to be important enough to implement from day one, and most developers absolutely refuse to fix the inaccessibility in their products, or if they don't outright refuse, they just ignore the situation and hope we'll go away and not push the issue. Standardization hasn't helped on that score very much.

 

I'm very interested in this topic and it's really helpful to hear from @IndomitableHeart and the perspective of someone trying to use a game engine like Unity. Clearly, a game can be made that is not predominately visual, so it stands to reason that an engine should allow for as many people as possible to make that game.

I also see a potential conflict, as Unity is clearly starting to move towards visual graphs that seem to perform as good or better than written code, i.e. ShaderGraph. This makes writing shaders accessible to non-programmers, which means many more people with diverse skillsets are able to jump into advanced areas of the program. At the same time, if there is not an existing template for how to convert a node-based graph for a non-visual user, that's going to be a major challenge.

This reminds me of a talk I attended by a Danish architecture firm that was working in the US on implementing bike lanes in a very dense US city. They know how to design good bike systems in Copenhagen, but some of the laws in the US were in conflict, i.e. accessibility setbacks and design requirements. The dilemma they were hitting is accessibility for one group was making a less safe path for another group. They ended up going with what the law required, but they also felt the design was less safe and less usable for bicyclists, meaning a lower adaption rate and more accidents.

I don't really have an answer vis a vis Shadergraph, but I'm just pointing out the challenges involved when developing complex systems like a game engine.

 

Is this still in reference to video games?

Perhaps this would be best for a new thread, I'd love to hear your thoughts on how today's popular games could be more accessible. Or if there are examples of popular games which are particularly accessible, I'd love to hear what it is about them that helps to achieve that.

 

No, the problems with developers not wanting to address accessibility issues isn't solely limited to videogames. It's in literally any area of software. Ask a dev to fix it so that we can use their stuff, and they either outright say no, or worse, they ignore us and don't say anything. Even these huge corporations with billions of dollars won't work well with us, and they've got the manpower, the cream of the crop in coders, the resources, and anything else they need to do it, and yet they refuse. That's part of the reason that it appears to us at times to be very much a mentality issue, a belief that we're inferior and less important than people with sight. But we're not.

As for making videogames accessible, there are a number of ways to improve accessibility. Add spoken menus and in-game text, either through TTS or through voiceovers by humans. TTS can be used on most operating systems now, as well as both XBox and PS4 consoles. Add audio cues to certain things like doors, objects, and interactable NPCs. Add lots of haptic feedback.

 

I think it's a direct response to the thought "we had screen-readers back when Unity was being invented, but they weren't in common use. Now they are and Unity should get with the times". IdomH is correcting that last part - saying screereader-enabled Apps are still not in common use.

 

I am disabled and Unity has engaged me about my disability and we (me and Unity) care very much about any and all accessibility problems so I thank the community for their answers and the original poster.

But let me make it clearer than I did before:

If someone is struggling with something and has an opinion of that due to the struggle it is NOT the responsibility for everyone to jump on that and defend it but simply listen and take it on board. Nothing bad will ever happen from that. Remember, it's a perspective from someone who is very much struggling hard.

So listening to that and seeing what can be done (there are colour blind accessibility options and more to follow!) is a fantastic thing.

Thanks for understanding!