- From: Charles 'chaals' (McCathie) Nevile <chaals@yandex.ru>
- Date: Thu, 26 Aug 2021 14:06:10 +1000
- To: w3c-wai-ig@w3.org
Hi Wayne, thanks for the further explanation, and the note on how knowing about things helps identify problems elsewhere which is a brilliant piece of anecdote I hope to use. I also have personal experience of the effect - as a kid there were places I would specifically go for that, in what amounted to a completely naturally generated and in me very mild but interesting psychotropic effect. But then, as a kid I used to love things that would spin me around until I was somewhere between dizzy and sick. To grapple with the technical detail: I think there are a couple of issues here. One is that we should be noting in guidance that there is a potential problem with word highlighting - and that in most cases the best solution is unlikely to involve slowing the reading rate. This can occur in multiple ways - a user's assitive technology does the highlighting, or there is content-generated highlighting or subtitling that have the effect. Magnification of important focus areas obviously increases the likelihood of the problem surfacing. I think we should address the potential for author-created interaction between dynamically changing content that is potentially a flash, especially under magnification. Cross-linking requirements on magnification and on flash suppression might be an obvious place to start trying this. Slightly less short-term I wonder if it is time for WAI to have another go at describing requirements for user agents more comprehensively. Perhaps doing that as an informative document would reduce the problem of providers working to water diwn the guidelines so their products are not in breach, instead of recognising and documenting things that cause problems for users, and working to improve the products until they can solve the problems. cheers Chaals On Thu, 26 Aug 2021 02:54:41 +1000, Wayne Dick <wayneedick@gmail.com> wrote: > Hi All, > I brought this up because it bothers me. > Here is the problem. When you read and listen at the same >time, you > keep your visual focus on the word being read. The >visual information > reinforces what you cannot hear, and the >audio reinforces what you > cannot see. So, unlike non-visual >access to reading, other people with > print disabilities use the >see/hear combination to boost comprehension. > This requires >looking at the word being read. > > The word highlighted appears at the same region in the >visual field. It > looks like a flash. I have tried not looking at >the highlighted word, > but that disrupts comprehension. > > As far as slowing down reading speed to below 180 wpm, >that works but > reading drags. The typical person reads >200-250 words per minute on > average. > Good ebook or article readers give the user a choice to >highlight: > * None, > * Word, > * Phrase or > * Line. > > > I brought this up because I observed the problem, and could >calculate > that it does not meet Web guidelines. That means >it may be an > accessibility issue. Log it. You may not be able >to address it, but it > is a potential problem. > I did not start with the numbers and go backwards. I started >with > persistent nausea and worked backwards. > At CSU Long Beach we had a lattice lath structure over a >sidewalk that > caused flash issues for people who walked >under it. We discovered it > when the vice president of >publications got extremely dizzy whenever > she walked under >it at lunch time. She was a member of our Accessible > >Technology Initiative, and recognized the issue. The >University > changed the orientation of the lath and the >problem went away. That is > not a Web problem, but the >Web guidelines enabled us to identify the > problem. When a >person walked under the lath at a normal speed it > created a >flash effect that was greater than 3/second. > This highlight flash may not be a web content issue, but it >may very > well be an IT accessibility issue. > Best, Wayne > > > > > > > > > On Wed, Aug 25, 2021 at 8:51 AM Janina Sajka <janina@rednote.net> wrote: >> Good points, John. >> >> >> >> In terms of mitigation -- APA is looking at a technological solution. >> >> We're expecting a presentation of tech developed at MIT at an upcoming >> >> RQTF call. The concept has been proven to work. Now we need to get the >> >> licensing and user agent adoption. >> >> >> >> John Foliot writes: >> >>> Hi Janina, >> >>> >> >>> Not sure if you are aware of the PEAT testing tool from the Trace >>> Center ( >> >>> https://trace.umd.edu/peat/), however they state: >> >>> >> >>> "In general, web or computer content will not provoke seizures if >>> either of >> >>> the following is true: >> >>> >> >>> - There are no more than three general flashes and no more than >>> three red >> >>> flashes within any one-second period, or >> >>> - The combined area of flashes occurring concurrently occupies no >>> more >> >>> than a total of one quarter of any 341 x 256 pixel rectangle >>> anywhere on >> >>> the displayed screen area when the content is viewed at 1024 by 768 >>> pixels. >> >>> " >> >>> >> >>> So my reading between the lines suggests that it's less the location, >>> and >> >>> more the size. I wonder aloud how this is applicable to Wayne's >> >>> concern/observation? >> >>> >> >>> As a TTS tool that also provides text highlighting processes the >>> individual >> >>> words it is reading, are those 'highlighted words' less than or greater >> >>> than "...a total of one quarter of any 341 X 256 pixel rectangle"? Or >>> is it >> >>> more looking at a combination of "flashing highlighting" that is also >> >>> introducing horizontal movement (L to R in English, but R to L in, say >> >>> Hebrew) that it then becomes a concern? (In other words, while each >>> word is >> >>> highlighted individually, and thus likely below the individual >>> measurement >> >>> thresh-hold, is the *proper* way to evaluate this to instead think of >>> this >> >>> as the block of text that is flashing or strobing in the aggregate?) >> >>> Perhaps an issue that requires clarification and review in WCAG 3? >> >>> >> >>> Nonetheless, while I recognize the issue and concern, I'm also not sure >> >>> there is anything an individual content author can do to mitigate or >> >>> remediate this edge-case scenario. >> >>> >> >>> JF >> >>> >> >>> On Wed, Aug 25, 2021 at 8:51 AM Janina Sajka <janina@rednote.net> >>> wrote: >> >>> >> >>> > There's an additional nuance I've not previously considered ... >> >>> > >> >>> > Is the flash sensitivity specific to location on screen? i.e. more >>> than >> >>> > 3 per second at the same x,y location? >> >>> > >> >>> > Or does successive highlighting of words on screen also trigger the >> >>> > hazzard? >> >>> > >> >>> > To me this seems like it would be worth clarifying. >> >>> > >> >>> > Best, >> >>> > >> >>> > Janina >> >>> > >> >>> > John Foliot writes: >> >>> > > Hi Wayne, >> >>> > > >> >>> > > Ah, math <smile>. This presumes that the end-user has configured >>> their >> >>> > TTS >> >>> > > engine to read at this speed - but since all screen readers I've >>> seen >> >>> > also >> >>> > > allow the end-user the ability to adjust the reading rate, this by >> >>> > > extension means they can also adjust "flashing" in the use-case >>> context >> >>> > you >> >>> > > provided. (And if the user-agent stack doesn't, this is a failure >>> of >> >>> > UAAG, >> >>> > > which is non-normative, sadly.) >> >>> > > >> >>> > > Additionally, this will manifest on *any* content rendered in the >> >>> > > user-agent - this cannot be mitigated by the individual content >> >>> > > author/owner - it is a concern rooted at the user-agent level. >> >>> > > >> >>> > > One has to presume that a user who both requires >> >>> > > text-to-speech+highlighting AND is also sensitive to flashing >>> content >> >>> > will >> >>> > > have previously adjusted their user-agent stack to address this >>> issue. >> >>> > > >> >>> > > JF >> >>> > > >> >>> > > On Tue, Aug 24, 2021 at 6:57 PM Wayne Dick <wayneedick@gmail.com> >>> wrote: >> >>> > > >> >>> > > > If you are reading with a text-to-speech reader and it highlights >> >>> > words at >> >>> > > > more than 180 words per minute, then you have more than 3 >>> flashes per >> >>> > > > second. >> >>> > > > >> >>> > > > >> >>> > > >> >>> > > -- >> >>> > > *John Foliot* | >> >>> > > Senior Industry Specialist, Digital Accessibility | >> >>> > > W3C Accessibility Standards Contributor | >> >>> > > >> >>> > > "I made this so long because I did not have time to make it >>> shorter." - >> >>> > > Pascal "links go places, buttons do things" >> >>> > >> >>> > -- >> >>> > >> >>> > Janina Sajka >> >>> > https://linkedin.com/in/jsajka >> >>> > >> >>> > Linux Foundation Fellow >> >>> > Executive Chair, Accessibility Workgroup: http://a11y.org >> >>> > >> >>> > The World Wide Web Consortium (W3C), Web Accessibility Initiative >>> (WAI) >> >>> > Co-Chair, Accessible Platform Architectures >>> http://www.w3.org/wai/apa >> >>> > >> >>> > >> >>> >> >>> -- >> >>> *John Foliot* | >> >>> Senior Industry Specialist, Digital Accessibility | >> >>> W3C Accessibility Standards Contributor | >> >>> >> >>> "I made this so long because I did not have time to make it shorter." - >> >>> Pascal "links go places, buttons do things" >> >> >> >> -- >> >> >> Janina Sajka >> >> https://linkedin.com/in/jsajka >> >> >> >> Linux Foundation Fellow >> >> Executive Chair, Accessibility Workgroup: http://a11y.org >> >> >> >> The World Wide Web Consortium (W3C), Web Accessibility Initiative (WAI) >> >> Co-Chair, Accessible Platform Architectures >> http://www.w3.org/wai/apa >> >> >> > -- Using Opera's mail client: http://www.opera.com/mail/
Received on Thursday, 26 August 2021 04:07:04 UTC