jadelennox (
jadelennox) wrote in
dw_accessibility2009-10-30 11:33 pm
![[personal profile]](https://www.dreamwidth.org/img/silk/identity/user.png)
![[site community profile]](https://www.dreamwidth.org/img/comm_staff.png)
screenreader survey, and title attribute
- I'm going to add this to the "to document" wiki, but people should understand that the "title" attribute is not a tool to convey accessibility. Most users, screenreader or not, never see title attributes.
- Survey of results of assistive technology users testing title attribute access.
- A series of essays on "The title attribute - what is it good for?"
Except in form controls, title attributes are useless to 99% of users -- including sighted users, since most people don't hover long enough to see the tooltip. - WebAIM has published the results of their second screenreader survey. Summary of things that interested me:
- JAWS is still in the lead, but NVDA and VoiceOver are gaining.
- IE7 and IE8 are most common browsers, alas.
- most people have JavaScript on
- The answers about alt text are complex.
- The most annoying accessibility problems that would be a problem on dreamwidth are poorly named link text, keyboard access, and bad forms.
- On a lengthy page, many users navigate via the headings on the page.
- There is no typical screen reader user.
no subject
I assume this is talking about the discussion on bug 2037?
If that's so, I should point out that I, for one, do know that it's not something that would be very good for accessibility, and I knew that it isn't used by assistive technology.
However, having the alt text be just "user" and "community" may turn out to be better; I don't know and I don't want to judge on that, not being a screenreader user myself. But since we would otherwise have no title attribute anyway, for those who could use it, it might be better to have the "profile" appended if we were going to use that way.
However, I have little experience on this.
(By the way, I have an issue with that first experiment drawing conclusions from such a small dataset. In particular, two conclusions were drawn based on one user each. This is not how you do a good survey. :/ That said, I can see how some conclusions could be taken.)
no subject
One of the problems is that w3c talks about the title attribute from a potential basis, and doesn't really talk about the genuine limitations of user agents and screenreader configuration. The title attribute OUGHT to be useful, but given the way browsers implement it, and given the way most screenreaders are configured out of the box, it ends up not being potentially harmful, at least if people forget that the content in there is invisible to many users so it has to be used for supplementary information only.
Also, you are right that that one survey to which I linked is problematic. It's not just a very small pool, it's also dated. It stops at JAWS 7, I think. There aren't that many great pages online in which information about what's wrong with the title attribute is neatly compiled; this excellent summary, for example, had to get pulled off the wayback machine.
So in any case, as far as the information on that bug goes, I agree with you that I really don't have a clue about what's the right thing to put into that alt text. I just think that we need to make sure that the title attribute is entirely supplementary information that it doesn't matter if people never see/hear, kind of like the JavaScript pop-ups over the image.
no subject