- From: Sean Palmer <sean_b_palmer@yahoo.com>
- Date: Thu, 22 Jun 2000 05:23:29 -0700 (PDT)
- To: www-html@w3.org
Hello everyone, Has anyone noticed that valid XHTML can sometimes not be valid XML? Try it: simply change the extension of a valid XHTML document to xml, or change the MIME type at what ever server you have, and run it through a validator. Doesn't work, does it? I'm surprised that W3C haven't seen this bug before, but there you go. This also ties in with the recent problems outlined with the MIME type of XHTML - personally I think a new one "text/xhtml" may be needed, because it seems that XHTML is neither HTML or XML at all.You may say "well it gets read by the browser as HTML (text/html), but what if you had a <br/> tag instead of <br />. In some browsers this would give problems because it isn't HTML. XHTML needs more thought put into it before XHTML 1.1 is released. Did anyone see the comments I made about XSLT and CSS? Someone pointed out to me that it may be better to break away completely from CSS and the old HTML semantics of it. XHTML 1.1 should be a pure XML language compatible with XSLT for the new upcoming branch of XML browsers (e.g. IE5. Does anyone know if Netscape 6 supports XML?). The Internet is evolving to fast for some people to catch up, so backwards compatability is always going to be an issue, but I can see the days where we will be able to have raw XHTML, our own XML, and XSLT all combined into one document. Ad, what with the coming XForms and XLink specifications, the future of XHTML could be very bright indeed. Imagine this:- You could have a single document that is firstly a structural document for all of your screen output: this would be done with the XHTML modules. Then, you could style this with a mixture of CSS and XSLT (or only XSLT if I had my way!). You could then have another part of the document that contained raw XML to be interpreted by (perhaps) some script on the viewable page, or maybe it itself could be transformed into XHTML by a further XSLT document. Then, to add insult to injury, you could further enhance the document by adding interactivity, say including other documents from a different server using XLinks, and maybe a form here and there? My suggestion is simple, can anyone work out a way of making sure that a future XHTML document can do all that I described above, as well as maintaining backwards compatability? I'm sure W3C will manage it, but they need to act quickly. Overall, everyone will be saying that this is too much of a gap to bridge in just one go, but does anyone remember how quickly HTML caught on in the first place? And what of ASP and all of the other Server Side Scripting languages? Actually, that brings me to another quick point, what is W3C doing to address the unscrupulous people who detect what browser the client is using, and then deliver shoddy code as appropriate? This means you could have a document with a marquee tag (perish the thought) but that still validates as an XHTML page through W3Cs validator. A bit of a nightmare problem that! Come on W3C, flex your corporate muscle, any carry through all of your XML dreams - make us proud! Sean B. Palmer WapDesign ORG U.K. - http://www.wapdesign.org.uk/ __________________________________________________ Do You Yahoo!? Send instant messages with Yahoo! Messenger. http://im.yahoo.com/
Received on Thursday, 22 June 2000 08:24:10 UTC