W3C home > Mailing lists > Public > www-validator@w3.org > June 2008

Re: label for association

From: Jukka K. Korpela <jkorpela@cs.tut.fi>
Date: Sun, 1 Jun 2008 18:45:56 +0300
Message-ID: <01eb01c8c3fe$977efad0$0500000a@DOCENDO>
To: "alun" <a.jones@hccnet.nl>, <www-validator@w3.org>

alun wrote:

> This error/warning is not caught.

Which error divided by which warning.

> The /for/ associated /id/ is incorrectly placed in the *label* or
> *other element* and not in a control.
>
> <label for="foo" id="foo"><input type="text"></label>

It is an error, but it is not a reportable markup error, so any software 
that purports to be a markup validator and reports it as an error is in 
error.

> None of the validators catch this, although it will be corrected in
> the next release of Total Validator.

Sounds like snake oil. There is a lot of advertisement of "better than 
validator" validators that aren't no validators.

If you mean http://www.totalvalidator.com then I think you should have 
indicated your real name and your affiliation or non-affiliation with 
it.

I stopped holding my breath when that "Total Validator" refused to do 
validation:
Parsing Errors
E000 - 1 instance(s): Due to the online service being subjected to DoS 
attacks in the past we've had to limit pages to no more than 15000 tags. 
This restriction does not apply to the desktop tools

That's a blatant lie. It is no parsing error, and it is not a reportable 
markup error in the document. It is just an arbitrary restriction in the 
software that is advertized as a validator.

Besides, it is not a markup validator. It recognizes just a collection 
of HTML document types. It even babbles about

 The <!DOCTYPE> tag

which should be sufficient as a cluelessness indicator. So were you just 
dishonestly advertizing your own phoney "validator"?


Jukka K. Korpela ("Yucca")
http://www.cs.tut.fi/~jkorpela/
Received on Sunday, 1 June 2008 15:46:25 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Wednesday, 25 April 2012 12:14:29 GMT