W3C home > Mailing lists > Public > w3c-wai-ig@w3.org > October to December 2014

"Supported technology"/"Until user agents…" was Re: text 'truncation' at normal scaling

From: <chaals@yandex-team.ru>
Date: Thu, 13 Nov 2014 18:57:32 +0300
To: Olaf Drьmmer <olaf@druemmer.com>, W3C WAI ig <w3c-wai-ig@w3.org>
Cc: Wayne Dick <waynedick@knowbility.org>, James Nurthen <james@nurthen.com>, Adam Cooper <cooperad@bigpond.com>
Message-Id: <44121415894252@webcorp02e.yandex-team.ru>
TL;DR: There is still an issue here, and while we'd like to avoid overloading authors, we should focus on telling them what users need.

13.11.2014, 11:35, "Olaf Drümmer" <olaf@druemmer.com>:
> This can be solved by tools,


> and does not have to be solved by authoring.


> Because of the flexibility of HTML you can adjust display in a zillion ways, including a decent combination of aspects like zooming, text only zooming, guided reading (retracing to beginning of next line), etc. Some features are even there in today's viewers.

Yes, sometimes this works already. But there is a concrete problem for actual humans when it doesn't.

> If we try to force authoring to take every tiny detail into account it's a battle that will be lost. We should choose our battles wisely, as everyone's resources are limited.

Yes, and yes. But…

WCAG should tell you how to make things accessible. That's what people believe it does, and that's what it "says on the tin". 

Whether you do everything it says or not is beyond the scope of the document. The battle we fight in making WCAG isn't the one of "everyone does the right thing", but the one of "everyone can find out what they should strive for, when they want to do the right thing".

In WCAG 1 we used the phrase "Until user agents…" to introduce a number of requirements for authoring, that were things which really *should* have been solved by tools, but weren't.

The idea is simple: If users can't access content effectively today, because the widely-deployed (i.e. *readily* available) tools are not good enough, but authoring practices can provide a workaround, we should say so.

The issue is a bit more complicated. 
If we convince authors to solve the problem, users will say "but there isn't an issue - the page does what I need" which leads tool developers to prioritise something else, based on "listening to what their users need". At a global scale, that argument is false - trying to get millions of authors to do something is generally vastly more expensive and less successful than trying to get hundreds of tools to do it. If authors didn't need to make the workarounds, they could spend more time on something else.

On the other hand, saying to people "it's OK, you can't pay your taxes or get a Doctor's appointment today because the tools don't really work, but in a year or two they will, and in the meantime we don't want authors to waste their time" has its own problems.

"Until user agents…" conditions led to endless discussions about "do user agents now…?". So it was dropped and in WCAG 2 we have "accesibility-supported technology". Which is the same thing, with a facelift. Is "the Web platform" (as people call it) *really* an accessibility-supported technology? In lots of ways and cases, Yes. Compared to say Silverlight, it's harder to judge since there are vastly more tools that people might reasonably have to use. PDF is somewhere in the middle - there is a very good set of products from Adobe, but it is not the case that everyone can use them to produce and read PDF.

Which means… what, exactly?
To a certain extent it seems that we have shut down the discussion, without actually dealing with the problem. That seems the wrong way round.

It is true that there are things which tools really *should* solve, but it isn't clear that they generally do. If we don't tell authors what those things are, they are far less likely even to spend a bit of spare time doing them *sometimes* - which would be far better than doing them *almost never*.

We need to be clear and honest about this, with users, authors and toolmakers. 

Providing the necessary information is how we enable the market to start sorting out the issues - for better or worse, WCAG is about the single most influential description of what is required to make the Web accessible. That also provides us with a clear reference point from which to talk about what is working well, and where there are problems that would be better solved by shifting burdens to those whose responsibility they "ought" to be.

Which includes dealing in annoying and complex issues of reality, as well as the clearer waters of "what is the most rational distribution of effort"?


> Olaf
> On 13 Nov 2014, at 07:06, Wayne Dick <waynedick@knowbility.org> wrote:
>>  It about time that we drop the myth that zoom constitutes reasonable accommodation.  There are so many things wrong with reading with zoom that it is hard to start enumerating them.  First there is retrace.  People who read with screen magnification spend about 17 to 50 percent of read time involved in retrace.  Next there is the challenge to short term memory.  Thus reading with zoom is slower and much more error prone.  Finally there is the noise to signal ration.  On any given page only the line being read has any meaning, the rest of the page is disconnected noise.  Since most people with vision loss have difficulty distinguishing signal from noise this especially problematic.
>>  Can some people read with zoom?  Sure, I can, and do when no other option is available. I have read amazingly difficult content with zoom, but the experience is profoundly distracting, and it is slow and ugly work.  You have to be very smart to succeed professionally with only zoom to help you, and you will never reach your intellectual potential.
>>  Wayne Dick

Charles McCathie Nevile - web standards - CTO Office, Yandex
chaals@yandex-team.ru - - - Find more at http://yandex.com
Received on Thursday, 13 November 2014 15:58:11 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 13 October 2015 16:21:53 UTC