Re: Massive HTML coding errors

Jack wrote:
Albert Wiersch wrote:
Besides, you could argue that the standard is Microsoft Internet Explorer, because that's what most people use.

You'd *never* catch me arguing that any 'standard' is defined by the
vendor of the most popular product. As far as I'm concerned, a standard
is a set of conventions established by a standards-setting body. But
this is beside the point; validity isn't about conforming to a
'standard', it's about whether a document is consistent with the DTD
that it says it's consistent with. Any well-formed XML or SGML document
can be valid, if it contains a DTD declaration, and is consistent with
that DTD. HTML documents that claim consistency with some DTD should
(FSVO 'should') be consistent with that DTD, and some people want to check.

In case the significance and importance of the distinction still isn't clear to Mr. Wiersch, let's try bringing Noam Chomsky's contribution into this: "Colorless green dreams sleep furiously". This is a perfectly valid English sentence. It's meaningless, but it's valid. "He ran very rapid" isn't valid, though most people would be able to figure out what it means. An application analogous to the CSE product might be able to tell that the former sentence is meaningless and that the latter sentence is meaningful, but doing so would be semantic analysis, not validation. Mr. Wiersch could call it validation all he wanted, but that would just be a demonstration of ignorance, obliviousness, or intentional disregard.
.

Relevant Pages

  • Re: contiguity of arrays
    ... > That isn't actually consistent with how the standard uses the term ... No, the standard doesn't seem consistently pedantic about consistency, does ... objects (i.e. regions of memory) accessed through an ... lvalue of a particular type must be aligned appropriately to the type. ...
    (comp.lang.c)
  • Re: clc-compliant [was Re: something else]
    ... results consistent with a reasonable ... consistent with my understanding of the standard, ... conforming, or both? ... allows a range of possibilities as unspec or impl-def, ...
    (comp.lang.c)
  • Re: OT: Need HTML Help
    ... Your initial DTD lines at the top are designed to define what standard ... If your code is compliant with standard X, ... standards-compliant browser ever produced although I don't know how MS ...
    (comp.os.vms)
  • Re: Model here, model there
    ... If T is consistent then G is ... and meta-theory or semantics. ... The point is that T itself can't even DISCUSS "T's standard model". ... model is true and the non-standard models are false? ...
    (sci.logic)
  • Re: Massive HTML coding errors
    ... in real-world browsers and they don't care about strictly conforming ... Your attitude seems to be that claiming conformance to a DTD is just ... This is consistent with your attitude that the term validator means what ...
    (comp.infosystems.www.authoring.html)