- From: Charles F. Munat <chas@munat.com>
- Date: Wed, 20 Dec 2000 11:57:33 -0800
- To: "'Bailey, Bruce'" <Bruce_Bailey@ed.gov>, "'Chris Croome'" <chris@webarchitects.co.uk>, <w3c-wai-ig@w3.org>
- Cc: <webmaster@altavista.com>
Yes, the CSS validator is really useful. I generally work with the style sheet until I get no warnings at all. There is one instance where I ignore the warning, however, because I believe it is only a reminder and my action was intentional: margin: 0em; margin-bottom: 1em; Here I get warned that I've redefined the margin-bottom. I know. That's what I wanted to do. It's easier than writing out all the margins. Frankly, I think that the validator should provide a warning if you use a fixed size for fonts. Which brings up another issue: one area where I believe the guidelines are unclear is 3.4: Use relative rather than absolute units in markup language attribute values and style sheet property values. Well, unclear isn't really the word I want. It's pretty clear what this says. But I think that it should really be concerned with font size, not *all* property values. For example, suppose that I have a right-aligned header (h1) followed by a right-aligned subhead. They are white text on a red background. I want just a little room between the edge of the red background and the end of the text, and I want the text to align vertically. Now, if I use, say, 0.1em padding, the larger heading will move farther from the edge than the subhead. In this instance I see no advantage to the relative units. All I want is a couple of pixels offset, so I should be able to use padding-right: 2px. Checkpoint 3.4 clearly prohibits this, and I think that's too limiting. Has anyone else run into this? Charles F. Munat Bruce Bailey wrote: Thanks for taking a look at this Charles. I have not been able to get to it and you provided a MUCH better analysis than I could have. I will also point out that while the page in question does, strictly speaking, use "valid css", the w3c jigsaw tool returns over two dozen warnings! All of the warnings have to do with color and background-color, so they speak directly to the usability problems noted. >
Received on Wednesday, 20 December 2000 14:51:29 UTC