On Mon, Jun 22, 2009 at 10:42 PM, Remember the dotrememberthedot@gmail.com wrote:
In Håkon Wium Lie's recent analysis of Wikipedia image markup ( http://www.princexml.com/howcome/2009/wikipedia/image/), he makes a good point: we include image captions both below images and again in the images' tooltips. Also, for inline images without explicitly defined tooltips, the image name is used as the tooltip even though it is also shown in the URL when mousing over the image. Neither of these automatic tooltips are really useful, and they slow down page load time on image-heavy pages.
What do you think? Should we keep the redundant tooltips, or start leaving them out?
We should definitely leave them out. They're clearly redundant. I tried to do this before, but the code for this is horrible and I gave up. The code to pass image attributes from the parser to the appropriate image-generating class (mostly in Linker) needs to be scrapped and rewritten from scratch.
On Tue, Jun 23, 2009 at 12:21 AM, Tim Larsonlarson@towncommons.com wrote:
One suggestion that hasn't been mentioned is leaving these as titles in interactive settings (tooltip in web page) and using CSS to generate inline text in others (caption when printed).
This would prohibit captions from containing markup, and not work in IE6 or IE7. Also, tooltips are not at all discoverable -- captions are a much better place to put info you expect people to actually read.
On Tue, Jun 23, 2009 at 12:52 AM, Brianna Laugherbrianna.laugher@gmail.com wrote:
This is actually only the case if you use the keyword 'thumb' or 'frame'. If you just do [[file:foo.jpg|this is my caption]], then you only get the "tooltip" (usually called "alt" text for images). http://en.wikipedia.org/wiki/Alt_attribute
We're discussing the title attribute, not the alt attribute. They're entirely different. Alt text is incorrectly rendered as a tooltip by Internet Explorer if no title attribute is present, for historical reasons, but no other browser does this AFAIK. I don't think any screen reader uses the title attribute in place of the alt attribute, either.
Alt text can be set by itself now with [[Image:Foo.png|alt=xxx]], and this is currently the *only* way to set it for images with captions -- it's no longer the same as the caption by default.
They might not be useful for you, but they are useful for others.
Could you explain how a title attribute that duplicates caption text is useful to *anyone*?