Local server HTML validator

[Brian]

Re: Local server HTML validator

> Brian wrote ...[color=blue]
>[color=green]
>> W3C representatives aren't arguing in ciwah that the name "CSS
>> Validator" is appropriate. If they were, I've little doubt that
>> they would come under fire.[/color][/color]

[I actually wrote the quoted part above, though it does not appear that
way in your message. Please edit more carefully. Your postings make it
look like you said things that I said or the other way around.]

Albert Wiersch wrote:
[color=blue]
> By using the name, aren't they saying that it's appropriate?[/color]

They aren't saying it's appropriate *in ciwah*. If they were...
[color=blue]
> Anyway, I'm not arguing about it anymore.[/color]

We'll see.

--
Brian (remove "invalid" from my address to email me)

http://www.tsmchughs.com/

[Brian]

Re: Local server HTML validator

> Brian wrote ...[color=blue]
>[color=green]
>> W3C representatives aren't arguing in ciwah that the name "CSS
>> Validator" is appropriate. If they were, I've little doubt that
>> they would come under fire.[/color][/color]

[I actually wrote the quoted part above, though it does not appear that
way in your message. Please edit more carefully. Your postings make it
look like you said things that I said or the other way around.]

Albert Wiersch wrote:
[color=blue]
> By using the name, aren't they saying that it's appropriate?[/color]

They aren't saying it's appropriate *in ciwah*. If they were...
[color=blue]
> Anyway, I'm not arguing about it anymore.[/color]

We'll see.

--
Brian (remove "invalid" from my address to email me)

http://www.tsmchughs.com/

[Alan J. Flavell]

Re: Local server HTML validator

On Fri, 16 Apr 2004, Albert Wiersch wrote:
[color=blue]
> "Brian" <usenet3@juliet remblay.com.inv alid> wrote in message
> news:1080f2an7j qljf5@corp.supe rnews.com...[color=green]
> > Albert Wiersch wrote:
> >
> > W3C representatives aren't arguing in ciwah that the name "CSS
> > Validator" is appropriate. If they were, I've little doubt that they
> > would come under fire.[/color]
>
> By using the name, aren't they saying that it's appropriate?[/color]

CSS is not an SGML application, and makes no claim to being one. In
relation to CSS, the term "validation " is not pre-defined. What the
W3C validator does is to investigate the CSS syntax, and report errors
(which is what a validator does, in general terms), and optionally
offer warnings (which is no business of a validator as such, but is
nevertheless useful). It seems to me that the CSS thingy is primarily
a validator, and that its claim to be one does not clash with any
pre-defined specification of what a CSS validator should be.

So: it's a matter of personal taste (rather than of objective
definition) whether the term is appropriate.

In relation to HTML (starting with HTML2.0/RFC1866 - remember that?
- if you don't, at least some of us do) the term "validation " already
-had- a defined meaning, and was not open to marketing puffery.
[color=blue]
> Anyway, I'm not arguing about it anymore.[/color]

Didn't we already get to hear that before? The hole is getting
deeper.
[color=blue]
> There's no point to it.[/color]

Perhaps someone will bother to refer it under the Trades Description
Act.

Even Nick was willing to concede that it could be a useful tool, if
offered honestly for what it is: I say the same myself. As long as
you misrepresent it - despite your public admission that it's a lie,
with the implication that calling it a "validator" is a purely
marketing strategy to impress the gullible - we have no alternative
than to warn folks off it.

[Alan J. Flavell]

Re: Local server HTML validator

On Fri, 16 Apr 2004, Albert Wiersch wrote:
[color=blue]
> "Brian" <usenet3@juliet remblay.com.inv alid> wrote in message
> news:1080f2an7j qljf5@corp.supe rnews.com...[color=green]
> > Albert Wiersch wrote:
> >
> > W3C representatives aren't arguing in ciwah that the name "CSS
> > Validator" is appropriate. If they were, I've little doubt that they
> > would come under fire.[/color]
>
> By using the name, aren't they saying that it's appropriate?[/color]

CSS is not an SGML application, and makes no claim to being one. In
relation to CSS, the term "validation " is not pre-defined. What the
W3C validator does is to investigate the CSS syntax, and report errors
(which is what a validator does, in general terms), and optionally
offer warnings (which is no business of a validator as such, but is
nevertheless useful). It seems to me that the CSS thingy is primarily
a validator, and that its claim to be one does not clash with any
pre-defined specification of what a CSS validator should be.

So: it's a matter of personal taste (rather than of objective
definition) whether the term is appropriate.

In relation to HTML (starting with HTML2.0/RFC1866 - remember that?
- if you don't, at least some of us do) the term "validation " already
-had- a defined meaning, and was not open to marketing puffery.
[color=blue]
> Anyway, I'm not arguing about it anymore.[/color]

Didn't we already get to hear that before? The hole is getting
deeper.
[color=blue]
> There's no point to it.[/color]

Perhaps someone will bother to refer it under the Trades Description
Act.

Even Nick was willing to concede that it could be a useful tool, if
offered honestly for what it is: I say the same myself. As long as
you misrepresent it - despite your public admission that it's a lie,
with the implication that calling it a "validator" is a purely
marketing strategy to impress the gullible - we have no alternative
than to warn folks off it.

[Rijk van Geijtenbeek]

Re: CSE HTML Validator not legal in Norway (Was: Local server HTML validator)

On Fri, 16 Apr 2004 16:45:03 -0400, Brian
<usenet3@juliet remblay.com.inv alid> wrote:
[color=blue]
> Albert Wiersch wrote:
>[color=green]
>> I should have said DTD validator to avoid the confusion.[/color]
>
> Except that, in terms of SGML, there is no other kind.[/color]

What passes as HTML on the web has, on average, very little to do with
SGML... Albert has a point here :)

--
Rijk van Geijtenbeek

The Web is a procrastination apparatus:
It can absorb as much time as is required to ensure that you
won't get any real work done. - J.Nielsen

[Rijk van Geijtenbeek]

Re: CSE HTML Validator not legal in Norway (Was: Local server HTML validator)

On Fri, 16 Apr 2004 16:45:03 -0400, Brian
<usenet3@juliet remblay.com.inv alid> wrote:
[color=blue]
> Albert Wiersch wrote:
>[color=green]
>> I should have said DTD validator to avoid the confusion.[/color]
>
> Except that, in terms of SGML, there is no other kind.[/color]

What passes as HTML on the web has, on average, very little to do with
SGML... Albert has a point here :)

--
Rijk van Geijtenbeek

The Web is a procrastination apparatus:
It can absorb as much time as is required to ensure that you
won't get any real work done. - J.Nielsen

[C A Upsdell]

Re: Local server HTML validator

"Alan J. Flavell" <flavell@ph.gla .ac.uk> wrote in message
news:Pine.LNX.4 .53.04041623562 20.22333@ppepc5 6.ph.gla.ac.uk. ..[color=blue]
> CSS is not an SGML application, and makes no claim to being one. In
> relation to CSS, the term "validation " is not pre-defined ...[/color]

Surely it would be possible to devise a strict grammar for CSS, in which
case software that identified syntax errors with respect to this grammar
could reasonably be called a validator.

[C A Upsdell]

Re: Local server HTML validator

"Alan J. Flavell" <flavell@ph.gla .ac.uk> wrote in message
news:Pine.LNX.4 .53.04041623562 20.22333@ppepc5 6.ph.gla.ac.uk. ..[color=blue]
> CSS is not an SGML application, and makes no claim to being one. In
> relation to CSS, the term "validation " is not pre-defined ...[/color]

Surely it would be possible to devise a strict grammar for CSS, in which
case software that identified syntax errors with respect to this grammar
could reasonably be called a validator.

[Nick Kew]

Re: Local server HTML validator

In article <FT_fc.12497$QG 5.4212@news04.b loor.is.net.cab le.rogers.com>,
"C A Upsdell" <cupsdell0311XX X@-@-@XXXrogers.com> writes:
[color=blue][color=green]
>> CSS is not an SGML application, and makes no claim to being one. In
>> relation to CSS, the term "validation " is not pre-defined ...[/color]
>
> Surely it would be possible to devise a strict grammar for CSS, in which
> case software that identified syntax errors with respect to this grammar
> could reasonably be called a validator.[/color]

Indeed, that's exactly what tools like the W3C CSS validator do (bugs aside).

But your words "could reasonably be called", when contrasted with something
more definite like "is by definition", make Alan's point. And - crucially -
once we have "is by definition", we need a formal and well-understood
definition not merely of the syntax to be validated, but of a validation
process. No such thing exists for CSS, which is why it's considered
to be legitimately open to armwaving like "could reasonably be called".

--
Nick Kew

Nick's manifesto: http://www.htmlhelp.com/~nick/

[Nick Kew]

Re: CSE HTML Validator not legal in Norway (Was: Local server HTML validator)

In article <opr6kp76iwicz8 n2@news.individ ual.net>,
"Rijk van Geijtenbeek" <rijk@opera.com > writes:
[color=blue]
> What passes as HTML on the web has, on average, very little to do with
> SGML... Albert has a point here :)[/color]

That point would be that validation is only one element in the QA process,
and probably not the most important.

It justifies having checking tools other than validators.
It doesn't justify selling a lie to innocent end-users.

--
Nick Kew

Nick's manifesto: http://www.htmlhelp.com/~nick/

[Nick Kew]

Re: Local server HTML validator

In article <FT_fc.12497$QG 5.4212@news04.b loor.is.net.cab le.rogers.com>,
"C A Upsdell" <cupsdell0311XX X@-@-@XXXrogers.com> writes:
[color=blue][color=green]
>> CSS is not an SGML application, and makes no claim to being one. In
>> relation to CSS, the term "validation " is not pre-defined ...[/color]
>
> Surely it would be possible to devise a strict grammar for CSS, in which
> case software that identified syntax errors with respect to this grammar
> could reasonably be called a validator.[/color]

Indeed, that's exactly what tools like the W3C CSS validator do (bugs aside).

But your words "could reasonably be called", when contrasted with something
more definite like "is by definition", make Alan's point. And - crucially -
once we have "is by definition", we need a formal and well-understood
definition not merely of the syntax to be validated, but of a validation
process. No such thing exists for CSS, which is why it's considered
to be legitimately open to armwaving like "could reasonably be called".

--
Nick Kew

Nick's manifesto: http://www.htmlhelp.com/~nick/

[Nick Kew]

Re: CSE HTML Validator not legal in Norway (Was: Local server HTML validator)

In article <opr6kp76iwicz8 n2@news.individ ual.net>,
"Rijk van Geijtenbeek" <rijk@opera.com > writes:
[color=blue]
> What passes as HTML on the web has, on average, very little to do with
> SGML... Albert has a point here :)[/color]

That point would be that validation is only one element in the QA process,
and probably not the most important.

It justifies having checking tools other than validators.
It doesn't justify selling a lie to innocent end-users.

--
Nick Kew

Nick's manifesto: http://www.htmlhelp.com/~nick/

[Brian]

Re: CSE HTML Validator not legal in Norway (Was: Local server HTMLvalidator)

Rijk van Geijtenbeek wrote:
[color=blue]
> On Fri, 16 Apr 2004, Brian wrote:
>[color=green]
>> Albert Wiersch wrote:
>>[color=darkred]
>>> I should have said DTD validator to avoid the confusion.[/color]
>>
>> Except that, in terms of SGML, there is no other kind.[/color]
>
> What passes as HTML on the web has, on average, very little to do
> with SGML...[/color]

You appear to know more about SGML than I do (that wouldn't be hard;
there are likely scores of folks here know more that I do). But I'll be
brave and ask questions.

If the HTML one finds on the www has little to do with SGML -- no
argument from me on that one -- does that change the meaning of HTML
validation? Does that divorce HTML validation entirely from SGML?
[color=blue]
> Albert has a point here :)[/color]

I'm not so sure. Whatever one may say about HTML on the www, I don't
think that's an excuse to mischaracterize a linter as a validator and
confuse the matter for the ignorant.

--
Brian (remove "invalid" from my address to email me)

http://www.tsmchughs.com/

[Rijk van Geijtenbeek]

Re: CSE HTML Validator not legal in Norway (Was: Local server HTML validator)

On Sat, 17 Apr 2004 12:48:48 -0400, Brian
<usenet3@juliet remblay.com.inv alid> wrote:[color=blue]
> Rijk van Geijtenbeek wrote:[/color]

...
[color=blue][color=green]
>> Albert has a point here :)[/color]
>
> I'm not so sure. Whatever one may say about HTML on the www, I don't
> think that's an excuse to mischaracterize a linter as a validator and
> confuse the matter for the ignorant.[/color]

By creating a CSS checker and calling it a 'CSS validator', the W3C itself
indicates that there is more to the word 'validator' than parsing an SGML
document according to a DTD...

I think it shouldn't be too hard to build in 'real' DTD validation in
Albert's product, it would make it a better product. This criticism of CSE
is years old, and obviously either Albert is very stubborn, or he doesn't
understand why it can be useful to know a document is valid in the sense
of condforming to a specified DTD. Either way, it doesn't inspire much
confidence in the quality of the product.

--
Rijk van Geijtenbeek

The Web is a procrastination apparatus:
It can absorb as much time as is required to ensure that you
won't get any real work done. - J.Nielsen

[Jan Roland Eriksson]

CSS as an application of SGML? (Was: Re: Local server HTML validator)

On Sat, 17 Apr 2004 00:12:42 +0100, "Alan J. Flavell"
<flavell@ph.gla .ac.uk> wrote:

[crossed, and fup's set, to ciwas]

[...]
[color=blue]
>CSS is not an SGML application, and makes no claim to being one.[/color]

[...]

A thought that strikes me rather late in the day is if it would be
possible to create an SGML declaration that is fitting for the available
CSS syntax, i.e. could a retrofit work be done that brings the CSS
syntax into an SGML conforming state?

At time of writing this post I don't know if the idea of an SGML
declaration for available CSS syntax, and a DTDefinition that makes use
of that SGML declaration, is possible to turn into reality.

But it sure has a nice ring to it to be able to have a unified method of
true validation (based on nsgmls most probably) for both (X)HTML markup
and CSS stylesheets; No?

--
Rex .. <rex@css.nu> .. <http://css.nu/>