W3C home > Mailing lists > Public > www-validator@w3.org > November 2002

Re: Beta: Fatal Error: No DOCTYPE specified!

From: Nick Kew <nick@webthing.com>
Date: Fri, 1 Nov 2002 18:49:51 +0000 (GMT)
To: Kim Brooks Wei <kimi@kimbwei.com>
cc: W3C Validator <www-validator@w3.org>
Message-ID: <Pine.LNX.4.21.0211011824200.1501-100000@jarl.webthing.com>

On Fri, 1 Nov 2002, Kim Brooks Wei wrote:

> I've been following this issue with interest. As a non-programmer, I 
> have a user's perspective that may bring into this discussion some 
> useful insights.

Thank you for contributing:-)

> The validator service is very useful to me. I am told by 
> coder-friends that sometimes html IN-validation is a good thing and I 
> take their word at face value; they have experience which I do not.

Side note: they might possibly be right, but most people who say that
are talking from ignorance, no matter how well-qualified they are.
See <URL:http://valet.webthing.com/page/why.html>

> On the home front, I need the validator service. [ ... chop ... ]

Yes, I should think that makes you a fairly typical user.

> As regards URI, the error messages it gives out could be a lot more 
> user-friendly.

The error messages have been reviewed at various times: the problem
is that they arise from an SGML parser, and can sometimes appear in
a context that wasn't envisaged.  Other validators may use different
messages (I think mine are better:-), but there's no solution that's
going to be right in every situation.

>	 I also often get confused by the little red arrows 
> pointing to a line of coding error.

Have you seen the validator at <URL:http://valet.webthing.com/page/>?
It's the same underlying parser, but a very different presentation
that you might prefer.

> Now, it happened that this week, I was unable to get my page 
> recognized as html. Someone in this list suggested the idea of 
> supplying a popup box in the case of incorrect doctype specification, 

Are you sure it was your doctype that was wrong?

I think we need to review the error message you get when your
server is misconfigured (whether or not that was the problem
in this specific instance) to save the possibility of ...

> This week, had I been absolutely certain that my doctype 
> specification and html declarations were correct, I could have gone 
> on to look elsewhere for the issue that was causing my documents to 
> be recognized as text ___ and saved myself about 5 hours of research 
> into making absolutely, unequivocably certain that I had in fact 
> stated the correct doctype.

> [chop]

> It makes no sense to search 2nd-or-lower tier tech knowledge bases 
> for the definitive answer to questions such as the one I faced 
> recently. I get too many opinions and won't be able to ascertain 
> without trial and error, between the good and bad amoung them and 
> unfortunately, if I am able to locate correct knowledge at the 
> 2nd-tier knowledge base level for one issue, this may be a fluke.

I'm not 100% sure what you're saying here.  But it's perhaps worth
noting that in any sufficiently "mainstream" public forum - such as
this list, or the major Usenet newsgroups - you may get good or
bad advice, but whatever you get will be subject to Peer Review.
So at least when someone posts bad advice, it's always likely to
get corrected (or at worst, trashed for what it is).

> I believe that w3c is overestimating the technical ability of its 
> community of potential users and is also unaware of how little 
> understood web standards are by individuals not directly associated 
> with their evolution.

Aaargh, I really want to use that as an anchor for a rant, but
I didn't realise quite how much good insight you posted when I started
my reply, and I haven't time now to do it justice.

Thanks again for the insights - they tend to lurk on the periphery
of vision for those of us who are hacking the technology all the time!

Nick Kew
Received on Friday, 1 November 2002 13:49:58 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 1 March 2016 14:17:35 UTC