Post Archive

› November 18, 2004

HTML mess

  • Reported by liorean

Due to the harddrive that held my Windows partition (among other things) dying a loud, screeching, horrible death I wont post the nice XHTML and accessibility post I had prepared for some time. Instead, I'll give you two questions, school-book style:

What does, according to you, the 'text/html' media type signify and imply? How do user agents and specs differ in what they interpret it to mean?

What is the reason for these SGML heaven - Tag Soup - XHTML documents, which all are valid and hold the same structure, to have such radically different treatment?

Comments

1. November 18, 2004 10:34 AM

Quote this comment

Charl van Niekerk Posted…

Sorry, I forgot to do my homework. Please don't make me stand in front of the class...

Just kidding. :-)

2. November 18, 2004 10:45 AM

Quote this comment

Hayo Posted…

xhtml with text/html signifies it's not xml. bit of a paradox. Battles were already fought.

#2: Well, because they aren't done according to spec?

3. November 18, 2004 01:20 PM

Quote this comment

Mathieu 'P01' HENRI Posted…

#1 means that the document in its whole does not need to be a valid XML document. That is that short attributes are usable ( e.g.: selected="selected" can be replaced by selected ), and some closing tags are not necessary.

#2 A little bird told that's because the browsers do NOT support SGML. ;)

4. November 18, 2004 04:33 PM

Quote this comment

XHTML Posted…

XHTML stands for Extensible Hypertext Markup Language and is a reformulation of HTML 4.0 in XML 1.0.

For that matter, do we really need a separate media type for it? Why would text/html not suit the purpose? IMO it should be perfectly valid.

5. November 19, 2004 03:49 AM

Quote this comment

Curcan Ovidiu Posted…

XHTML cannot be read by human beings. That's why they decided on an application/* media type ;)

6. November 19, 2004 06:06 AM

Quote this comment

Ben de Groot Posted…

The text/html media-type implies that you are sending the browser an HTML document, and it will consquently treat it as such. Slapping an XHTML doctype on top of this is basically a lie, it will switch the browser into tag-soup mode anyway.

7. December 2, 2004 01:12 PM

Quote this comment

Nikole Jolie Posted…

Yeah, XHTML cannot be read by human beings. True. :)