- From: Jukka K. Korpela <jkorpela@cs.tut.fi>
- Date: Mon, 22 Nov 2004 10:31:45 +0200 (EET)
- To: Jon Ribbens <jon+www-validator@unequivocal.co.uk>
- Cc: www-validator@w3.org
On Mon, 22 Nov 2004, Jon Ribbens wrote: > I must admit I don't entirely get this. Why does the word "valid" have > such a strange meaning when applied to HTML? Because HTML is officially defined as an application of SGML or XML, and the SGML standard and the XML specification define "valid" in a specific technical meaning - and validators work with that meaning. > In any normal situation, > data which did not conform to the text of a specification would be > considered "invalid". You can use words like "incorrect", "wrong", etc. > For some reason, with HTML data can be wrong but > "valid". Surely in such an example as above it should be described as > "valid SGML" but "invalid HTML"? If not, why not? The phrase "invalid HTML" (or "valid HTML") would best be forgotten, but people use it in an attempt to be more understandable. Officially, HTML is a markup system (or "language") defined in a particular way, and this includes the use of a DTD and validity (by SGML or XML rules), i.e. that the markup complies with the DTD. Hence, a non-valid document is not an HTML document at all. -- Jukka "Yucca" Korpela, http://www.cs.tut.fi/~jkorpela/
Received on Monday, 22 November 2004 08:32:18 UTC