- From: Bailey, Bruce <Bruce_Bailey@ed.gov>
- Date: Tue, 6 Feb 2001 14:15:47 -0500
- To: "'Kynn Bartlett'" <kynn-edapta@idyllmtn.com>, w3c-wai-ig@w3.org
Dear Kynn, My sincere apologies for upsetting you. I really didn't mean to insult you, and my "depending on the weather" cut was uncalled for. As you wrote, perhaps you are the paragon of philosophical consistency -- and I am just too dull to appreciate it. I don't agree with promoting techniques (e.g., using TITLE attributes superfluously) that interferes with the accessibility of current products (e.g., JFW), even if future products (e.g., Edapta) will remediate such coding. The fact that such a practice is (technically) syntactically valid, hardly mitigates against the current usability issue. Yep, titles are metadata. Yep, GUI browsers tend to render them as ToolTips (what I was calling mouse-over pop-ups). Yep, tool tips can be useful (which is why I cited their use in links, abbr and acronym). Those points we agree on. The disagreement is that you are willing to promote (currently useless -- the search engines don't even process them) metadata when we know that IN ACTUAL PRACTICE this sacrifices accessibility?!? Forgive me, but I don't believe the pre-Edapta Kynn would have suggested such a thing! You didn't answer my hard question: Given the three (!) reasonable choices, what is an algorithm for deciding what the ToolTips text content should be for the following example? <a href="foo.html" title="Big World of Foo."><img src="foo.png" title="Foo Globe Icon" alt="Come to Foo!"></a> It doesn't get much easier when one has to decide the reasonable behavior of a screen reader! JAWS, when faced with this vague situation, favors TITLE content over ALT. This choice is arbitrary, but not entirely illogical. The simple solution to this quandary is for authors to skip the TITLE attribute -- unless one has some clear expectations for browser behavior. Don't expect the browser (screen reader or not) to make the right choice. As a content provider, don't expect the machine browser to read your mind! Avoid this ambiguity. -- Bruce
Received on Tuesday, 6 February 2001 14:16:08 UTC