- From: Robert Clary <bclary@netscape.com>
- Date: Fri, 06 Dec 2002 13:23:35 -0500
- To: David S <david@schrimpf.tv>
- CC: public-evangelist@w3.org
- Message-ID: <3DF0EB27.3010501@netscape.com>
David S wrote: > > On Friday, December 6, 2002, at 05:31 AM, Robert Clary wrote: > >> You gain no real benefit from using XHTML content if the page is to be >> served as text/html. > > > If nothing else, the syntax rules are more consistent and simply make > more sense than HTML4. Don't you think? I know it has been much easier > to teach strict XHTML than strict HTML4. > > The problem is that it is still HTML and will be treated as HTML by "other" browsers in wide distribution. This can lead to incorrect coding practices which the "other" browser will be forgiving of and which will result in another generation of web authors learning the wrong way to code XHTML/XML. Of course, since whatever broken markup they use works in the "other" browser, then any browser which disagrees is considered broken. Believe me, I've been there and seen that many times before. Of the last two years I have had to deal with numerous sites which thought that "XML was cool" and would incorrectly use XML syntax in HTML documents. These people did not really understand what XML was about nor how to use it. How do you deal with sites that use empty element syntax for non-empty elements such as <form /> or <option /> ? Believe me, it is not just minor sites which make these kinds of mistakes. Search the Tech Evangelism product in bugzilla.mozilla.org for examples of major sites which have made this mistake. If IE is so broken that it can not handle XHTML served as XHTML then send it a pure HTML page which it can handle. Any web author using it who looks at the W3C home page for inspiration will see a good example of using CSS to create a table-less layout in HTML which is a good thing. They will not be confused by the XML syntax and try to use it inappropriately in their own pages. The W3 home page is served by Apache/1.3.26 (Unix) PHP/3.0.18 which should be able to distinguish downlevel browsers such as Internet Explorer and serve them pure HTML while sending XHTML with the appropriate content type to browsers which can support it. I believe that is the most appropriate thing to do. Bob
Received on Friday, 6 December 2002 13:24:36 UTC