Anne van Kesteren

Creating myths

You probably have noticed that Tantek (among others I assume) has restyled technorati. I think that this version is a whole lot better than the previous one. Everything is very clear and navigating the site has become much easier, but I actually wanted to talk about the "P.S." in the post from Tantek:

P.S. Since it is built with semantic XHTML, does not block any browsers — try it with Lynx or a Hiptop for example. However, the style sheets are withheld from v4 and under desktop browsers, and I'm considering withholding them from IE5.x/Windows browsers as well if those browsers end up mangling the styles too much.

Although the statement is correct, people will probably believe it works in older browsers because of XHTML, which is just plain wrong. "Semantic markup" instead of "semantic XHTML" would probably be better wording for this. HTML is more backwards compatible than XHTML is. Just check this example from Ian Hickson:

The "/>" empty tag syntax actually has totally different meaning in HTML4. (It's the SHORTTAG minimisation feature known as NET, if I recall the name correctly.) Specifically, the XHTML

<p> Hello <br /> World </p>, if interpreted as HTML4, exactly equivalent to:

<p> Hello <br>&gt; World </p>

...and should really be rendered as:

> World


  1. The more remarkable part of that comment is in my opinion:

    I'm considering withholding them from IE5.x/Windows browsers as well if those browsers end up mangling the styles too much.

    This is the first time I read about a major site deciding to ignore IE5.x browsers, which are still being used by a big part of the visitors.

    Posted by Jeroen at

  2. Euh, considering; they haven't decided yet I think.

    Posted by Anne at

  3. It's not particulary major, but Skyzyx recently decided to let go of IEw 5.x.

    Posted by David House at

  4. Jeroen, it's not impressive either. Browser stats of late show that IE 5.x has a share the same size of Mozilla and Firefox, yet unlike those two, IE 5.x's share is rapidly declining. It's currently averaged at 1,5% according to sites like Google. Sites I tend to rely on much less (like The Counter) show as "high" as 16%, yet also rapidly declining.

    Posted by Faruk Ates at

  5. I always thought that that method of closing open tags was hackish. I never assumed that it was the legit way to do so. As far as I know, you can go through your site doing <br></br>, but who the hell has the time for that?

    It hasn't caused me any troubles in NN4 or Lynx yet, so I assume that while it's theoretically correct that it should be rendered differently, it just isn't. Is there an actual browser that will render this monstrosity true to form? If so, what's it's usage numbers?

    I used to use 3.5's to hold text documents and small pictures. Now I use them as protection between my beer and my wood computer desk. Not technically correct, but it works and it's not hurting anybody - including the W3 or established practices.

    Posted by Ryan Christie at

  6. Ryan...I've never heard of any HTML rendering engine actually rendering "/>" as an HTML engine is supposed to render it. I've poked my head around for one. If there is one somewhere, it's probably way too old and underused to be a valuable thing to consider when deciding whether to use XHTML 1.0 or HTML 4.01.

    Posted by Devon at

  7. I think you should clarify that when Tantek says withholding, he is not saying they will get nothing. Just no styling - the site should still function.
    IMO this is a legitimate thing to do, and I do it on my sites. Its really not worth fighting the bugs to try and use CSS in the older browsers, although you should give them the basic functionality

    Posted by Sean at

  8. Sean, you are missing the point.

    Posted by Anne at

  9. <br /> rendering poorly was a problem with the planets (Debian, Gnome) - Blogger's feeds have it everywhere, and it was quite ugly. It was fixed in feedparser, so you won't see it now, but it has happened.

    Posted by James at

  10. Oh, and fwiw, I use Mozilla, but it wasn't the browser, but feedparser that transformed <br/> into <br>> or <br></br>>. Live examples (from planets that haven't updated their feedparser): Ramlings of a Retarded Monkey, Jessica.

    Today's lesson: not all consumers of (X)HTML are browsers.

    Posted by James at