W3C home > Mailing lists > Public > public-html@w3.org > March 2010

Re: Bug 7034

From: Henri Sivonen <hsivonen@iki.fi>
Date: Fri, 26 Mar 2010 03:56:31 -0700 (PDT)
To: HTMLwg WG <public-html@w3.org>
Cc: James Graham <jgraham@opera.com>, Sam Ruby <rubys@intertwingly.net>
Message-ID: <1307082186.37907.1269600991349.JavaMail.root@cm-mail03.mozilla.org>
"Henri Sivonen" <hsivonen@iki.fi> wrote:

> Personally, I'd be more inclined to allow more
> interoperably-implemented presentational pre-existing markup features
> as valid to allow authors to use an HTML5 validator as a development
> aid when adding HTML5 features to pre-existing content that has
> interoperably supported legacy cruft in it.

Based on IRC discussion, I feel I should clarify:
I'm not suggesting that all interoperably-implemented presentational pre-existing features should be valid as a blanket policy. However, I think it would be worthwhile to allow things that are used so much that not allowing them makes validators too noisy for use with legacy markup--but only when the harm arising from the use of legacy markup features is directed mostly to the maintainer of the markup and not to the users.

Examples:
Tweaking table borders in CSS vs. tweaking them in presentational attributes is probably an issue where the harm is directed at the page author.
Using <font size> most likely means that <h1>, <h2>, etc. aren't being used, which is a case where the harm is directed at users of non-conventional browsing software.

-- 
Henri Sivonen
hsivonen@iki.fi
http://hsivonen.iki.fi/
Received on Friday, 26 March 2010 10:57:04 UTC

This archive was generated by hypermail 2.3.1 : Monday, 29 September 2014 09:39:15 UTC