W3C home > Mailing lists > Public > www-validator@w3.org > August 2003

Re: Validation result uri link encoding

From: Ville Skyttä <ville.skytta@iki.fi>
Date: Sun, 31 Aug 2003 17:40:33 +0300
To: www-validator@w3.org
Message-Id: <1062340832.24993.75.camel@bobcat.mine.nu>

On Sun, 2003-08-31 at 15:06, martin@lucas-smith.co.uk wrote:

> Aren't / and : legal characters in <a href= > links?

They're reserved characters in RFC 2396 (and 2732), along with '&' and
friends which are certainly more offensive if not properly treated in
query strings.  All those reserved chars are escaped in URIs emitted by
the validator, dunno if it's strictly necessary for all of them in this
case.  The only downside of encoding all of them in query strings is

> 2) There appears to be a space between check? and uri in the visible
> (rather than underlying link) version of this text as it appears. This
> makes cutting and pasting into HTML source potentially error-creating if
> people don't realise. I don't see any reason for the existence of this
> space.

It was added for cosmetic reasons, to make browsers wrap loooooong
URIs.  It's a hack, CVS log says:

   Add quick and nasty hack to insert a space in result URLs to enable 
   line wrapping in browsers. There *must* be a better way to do this! 

Less hacky suggestions welcome...

> 3) The beta is a great improvement. My only comment would be to replace
> the mailto: links on the right in fussy mode with a link to a feedback
> form, which automatically carries the error number. Presumably you would
> also get less spam that way, and not everyone has immediate access to a
> MUA.

Hm, how about Bugzilla, something like:
Received on Sunday, 31 August 2003 10:40:35 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 1 March 2016 14:17:38 UTC