Matthew Weier O'Phinney wrote:
even with comlpiant browsers, valid html and valid css you can still get display difference between browsers!-- Paul Johnson <[EMAIL PROTECTED]> wrote (on Wednesday, 19 February 2003, 10:15 PM -0800):On Tue, Feb 18, 2003 at 03:07:59PM -0600, DvB wrote:I've never done this, but I've seen it done (with me own eyes! :-) I don't think it worked as well as the native Linux browsers and probably would crash as soon as it started doing its Direct-X crap but, for your purposes, it would probably work (one would assume you do standards compliant development).Well, if that's the assumption, why bother getting IE to work at all? If you go to the standard, and it works in one browser, than it'll work anywhere. Save yourself the trouble. 8:o)Because IE has around 90% share of the browser market -- if it doesn't work on IE, you lose your audience. And, contrary to popular belief (hint: sarcasm!) coding standards-compliant HTML and CSS does not mean that if "it works in one browser, than[sic] it'll work anywhere." Not all browsers implement standards the same or correctly -- and, with the number of older browsers out there, you have to be worried also about graceful degradation of the code so that bugs in older browsers don't make a site unreadable.
take the following (abrevieted) example
<html>
<head>
<style>
#big {width : 100px; height :100px; background-color : red;}
#small {width : 50px; height: 50px; background-color : blue;}
</style>
</head>
<body>
<div id="small">
a small div containing
<div id="big">
a big div
</div>
</div>
</body>
</html>
now there is nothing invalid about either the CSS or the HTML above
But it is clearly bad coding - the big div can't fit inside the small one - and the browser has to either increase the size of the small div, decrease the big one, break the containment ...
unfortunately different browsers cope with this situation in different ways...
and while in this example it is easy to see what is wrong: in more complex pages similar things happen but are much harder to debug.
and I have not seen any tools that validate html and css together (though the mozilla DOM inspector comes close)
--
Sean Burlington
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
