I have mixed feelings about this. On one side, I'm very happy for blind people being enabled by software and hardware.
On the other side, I'm concerned that people have to use proprietary software to enable their sight. I doubt that there is open source software for blind people of similar quality. Am I wrong? I hope there is such software.
What would RMS do if he had a choice between using a mobile phone with proprietary software to enable sight and remaining blind?
If I could add my two cents here, I am quadriplegic and if I want to use a modern mobile phone and laptop or desktop I have to use Apple products. Their accessibility software is second to none and no other company comes close, I can take a new device out-of-the-box and have an able-bodied person set it up for me in about five minutes and from then on I need no other help.
Everything an able-bodied person can do with an iPhone I can do with my chin, usually with Microsoft and android and other open source offerings the accessibility software is always some subset of options and not the full experience. With Apple I get to use every aspect of every device despite the fact I can't move my arms and legs.
I would absolutely love to switch to Linux as my daily operating system, but I can't because the development has been done and it just isn't possible for me to use them as they are. So stuck with Apple I am. If Linux had the kind of support Apple did, I would switch in a heartbeat but they don't unfortunately.
They really are world leading in providing accessibility software for those of us with profound disabilities, and I've spent a decade looking.
Full disclosure: I help beta test the last couple of versions of the accessibility software for iOS.
Preface: I'm able-bodied, have reasonable sight, etc. In other words, I'm commenting on things I'm certainly not qualified to.
As someone who's used Linux for many, many, many years... I want to find fault with your conclusion that Linux and Open Source in general can't do this, but everything I've seen suggests I'm not very likely to be able to.
Even for able-bodied people, Linux isn't all that accessible - Some examples include poor font rendering (Yes, chrome renders nicely, but it's nowhere near consistent across all the applications I use day to day), another example is the fractured UI landscape, where apps tend to either be KDE, Gnome or Java, and each platforms idiosyncrasies leaks in leaving things like iconography, menu placement and organisation, and general UI styles different between each application.
I can't help but think, until we can sort this style of issue out, we have no chance of sorting out true accessibility. I'm reminded of XKCD's competing standards comic, where the most likely outcome here is probably yet another competing "standard".
I think it's just that Apple in this case has complete control over the ecosystem so they can mandate screen reader support(there's probably a Linux utility out there which only supports GTK), has a culture of perfection(there's no way that these features provide a good RoI), and is able to charge a high enough premium to subsidise these things.
I read Free Software: Free Society and I was amazed at how RMS's economic plan was basically 1) unis fund early development, 2) developers flesh out the ecosystem for free, 3) users start flooding in and, 4) these users magically start paying(or the hardware manufacturers do).
What this model misses is that it takes a lot more investment to make great software than just some developer time. It takes designers, artists, testers and depending on the software experts in specific fields. That's a huge investment in resources completely outside the software development field itself. Universities by themselves just don't have the research and investment capacity to carry advances in every single domain software can be applied to. They have other things to do too.
Back in the 60s it might have looked like the majority of effort to develop software just took some time by developers, but nowadays in many areas the developer time is a fraction of the overall effort required. It takes huge investment and if there is one thing Capitalism is good at, it's the productive and efficient deployment of capital. It's right there in the name.
Apple has control over the hardware, the operating system and their own apps but not over the entire ecosystem. This gets them far but only so far.
The big difference they've made is that they've created tools to make integration with Voice Over so easy, that it's a no brainer to do what little work needs to be done to make an app accessible and to increase demand for it.
They've managed to turn a burden into something that has an obvious economic benefit for app developers.
Also I'm fairly sure that there is actually a good RoI. It might not be massive but in the long term having the monopoly on phones for blind people surely pays off.
> I think it's just that Apple in this case has complete control over the ecosystem so they can mandate screen reader support
I don't believe they mandate it (how would that work for games?), but Apple provides excellent tooling and support throughout the system and it's built into all native controls so adding it to an application requires relatively little effort. See joshaidan's comment above: https://news.ycombinator.com/item?id=13846160
First, I don't think that "enabling sight" is the right term to use here. It's just making a certain device usable to the blind.
And yes, there are open source and free options available. iOS' accessibility APIs are closed to third parties and VoiceOver is the only option on that platform. However, there is free and open source screen reading software for Windows, Linux and Android.
I can't say what the state of Orca, the free Linux screen reader, is these days, but I know development is still going on. Orca wasn't so much of a problem in the past when I tried it, it was more that it could be a real pain to get everything working together. Think reasonable low-latency sound output for speech, driving a braille display through the BRLTTY software, getting the screenreader runnign at the login screen etc etc. I hope that has improved by now, but I only interact with Linux through SSH sessions or local text console these days.
Then there is NVDA for Windows. A free and open source screenreader mainly developed by two blind guys. On many fronts it has feature parity with the very expensive commercial offerings and even surpasses the commercial offerings on certain points. I use it as my daily driver.
In the past I also used a Mac near fulltime, but the VoiceOver of Mac OS became to buggy for my professional work. Also, usually updates only came when the OS was updated, so fixes and new features could take a while. So, long story short, open screenreader on a closed operating system that provides stable APIs seems to be the best of both worlds for now.
RMS is more of a realist than people think. He doesn't believe in owning phones, but he'll gladly use someone else's who has decided to make that tradeoff. He's well-aware of the fact that the proprietary regime leads to more software in more niches. He chooses to stand as a beacon of light to those who want a better world.
He considers his efforts to have been wildly successful.
I don't know why people always assume other's positions boil down to illogic.
People don't eat meat for many reasons. Those reasons may or may not include a moral component. If it does include a moral component, there are varying amounts of pragmatism you can add to it. I eat meat, but I'd happily switch over to vegetarianism if, say, I lived in India and the vegetarian food there is amazing, and everyone around me was also eating vegetarian.
see, that assumes you have infinite time & money at your disposal. But most of us have jobs and other commitments which mean we can't just drop everything to spend years of our lives researching and building something
There is Talkback on Android, which sometimes gets a source release with a long delay at https://github.com/google/talkback
It, and Android accessibility in general, is still far behind Apple. Unfortunately Linux desktop accessibility is also far less usable than Windows and MacOS. There is however an excelent open source screen reader for Windows called NVDA. I am using it right now. http://nvda-project.org/
Apple are investing resources far beyond the direct economic benefit into accessibility, out of moral concern (with a small PR benefit).
This is hard to replicate in the open source world where the model is "everyone contributes what they have an economic incentive to create".
On the other hand, regular command line tools are fairly accessible to blind users (I imagine, though I wouldn't like to read a man page in Braille, or by TTS), though of course the usual issues of inconsistencies between tools are magnified.
You might be interested in the results of the 2016 GOV.UK assistive technology survey. [0]
iOS VoiceOver ranked as one of the 3 most popular screen readers, second to JAWS for Windows [1] (closed source, paid) and above NVDA for Windows [2] (closed source, free, donations encouraged).
ChromeVox [3] is AFAICT the only free and open source screen reader that came up in the survey, at 1%.
You are assuming that Apple is making these accessibility features for capitalist reasons rather than to comply with disability and accessibility legislation.
No assumption is needed, just look at the record. Apple has been putting accessibility technology into its phones and desktop machines for over 20 years, long before it was ever mandated.
Apple has guided developers into using this technology in apps they write. As an example, Apple's VoiceOver screen reader technology uses widgets that are tagged with information about what kind of control they are. Developers can use this same tagging information when writing automated tests for their apps, which is a great way to enable developers to add assistive technology to their apps without making it difficult.
If that's the case, why isn't anyone else reaching this level of functionality? It is clear they are going above and beyond any level required by regulation.
On the other side, I'm concerned that people have to use proprietary software to enable their sight. I doubt that there is open source software for blind people of similar quality. Am I wrong? I hope there is such software.
What would RMS do if he had a choice between using a mobile phone with proprietary software to enable sight and remaining blind?