W3C home > Mailing lists > Public > public-qa-dev@w3.org > May 2004

Re: 0.6.6 bug roundup

From: Ville Skyttä <ville.skytta@iki.fi>
Date: Sat, 15 May 2004 12:54:17 +0300
To: Bjoern Hoehrmann <derhoermi@gmx.net>
Cc: QA Dev <public-qa-dev@w3.org>
Message-Id: <1084614857.13842.319.camel@bobcat.mine.nu>

On Fri, 2004-05-14 at 18:24, Bjoern Hoehrmann wrote: 
> * Ville Skyttä wrote:
> >* "Show source" discards source indentation, there are many 
> >  www-validator@ reports about this.
> 
> Why do we do what we currently do?

I don't know, maybe it's an oversight.  It was introduced in 0.6.5b1:
http://dev.w3.org/cvsweb/validator/htdocs/results.css?only_with_tag=validator-0_6_0-branch#rev1.13.2.19

>  And how to fix it?

Candidate fix at:
http://www.w3.org/Bugs/Public/show_bug.cgi?id=717
Testing/comments welcome.

> >* Doctype/charset override/fallback issues (I have not looked into
> >  this in great detail, but have observer similar issues):
> >  http://lists.w3.org/Archives/Public/www-validator/2004May/0089.html
> 
> Yeah, this needs to be fixed... why do we default to UTF-8 and not
> ISO-8859-1 or Windows-1252? It seems rather unlikely that defaulting
> to UTF-8 yields in desireable results... maybe we should use
> Encoding::Guess with UTF-8, ISO-2022-JP, Shift_JIS, Big5, ISO-8859-1
> and Windows-1252 or something? Or just try UTF-8 first and then ISO-
> 8859-1 with Iconv?

Ahh... Encode::Guess!  I have been looking for something like this,
thanks for the pointer!  Dunno if it's the right thing to do here
though, we still have the "strong default" of HTTP, ISO-8859-1, for
cases where no encoding is explicitly set.

I think this is just a bug introduced somewhere near the final 0.6.5
release as I don't remember seeing it earlier.

Anyway, there should be an option to override the encoding, cases like
this are annoying:
http://validator.w3.org/check?uri=http%3A%2F%2Fwww.hut.fi

> >* Verbose on or off by default?  Many seem to prefer "on" but as Bjoern
> >  noted, we shouldn't change the defaults too often.
> 
> I am not sure whether those in favor actually had a close look at it...
> Many fell in love with this fuzzy fussy feature and we dropped it.. I'm
> all for verbose mode (and on by default), but I'd say it's currently not
> ready for prime time.

Ok.

>  If we keep it off, we should remove the hidden
> verbose config option, it generates overly long URIs; and turn it off
> in check if it defaults to on.

+1
Received on Saturday, 15 May 2004 05:54:39 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Thursday, 19 August 2010 18:12:44 GMT