W3C home > Mailing lists > Public > public-html-comments@w3.org > January 2008

Re: Silently deprecating XHTML (was: HTML5 vs content type sniffing)

From: Anne van Kesteren <annevk@opera.com>
Date: Mon, 28 Jan 2008 13:22:15 +0100
To: "Frank Ellermann" <hmdmhdfmhdjmzdtjmzdtzktdkztdjz@gmail.com>
Cc: public-html-comments@w3.org
Message-ID: <op.t5m3bdvj64w2qv@annevk-t60.oslo.opera.com>

On Mon, 28 Jan 2008 12:42:55 +0100, Frank Ellermann  
<hmdmhdfmhdjmzdtjmzdtzktdkztdjz@gmail.com> wrote:
> Anne van Kesteren wrote:
>> The text/html syntax allows you to use XHTML in as
>> far XHTML 1.x as text/html allowed you to use it.
> I want missing explicit tags as clear errors, not as
> optional.  Case-sensitive, unforgiving, that is as it
> should be from my POV (=> easier to debug).

Then use XHTML5 with the proper media type. XHTML 1.x as text/html didn't  
have this feature either.

>> no prefixes, but <br/> works
> <br/> fails miserably with Netscape 3, I think you
> mean <br />.

Either way, Netscape 3 is sort of irrelevant at this point. (I know you  
used it until the summer of 2007 :-))

>> I suppose xml:lang is the only difference
> Right, some "transitional" (~ forever) side effects
> are ridiculous, fortunately xml:id wasn't around :-)
> BTW, your diff memo could mention that lang="" (or
> xml:lang="") are now permitted for XML compatibility.
> That wasn't the case in XHTML 1, it was forced to
> stick to NMTOKEN (non-empty, as in HTML 4).

Added a small note to my local copy that I should probably mention that  
and also IRIs. Thanks.

Anne van Kesteren
Received on Monday, 28 January 2008 12:19:31 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 6 January 2015 20:26:24 UTC