W3C home > Mailing lists > Public > ietf-http-wg@w3.org > January to March 2010

Re: rev parameter - LC comment on draft-nottingham-http-link-header-07.txt

From: Mark Nottingham <mnot@mnot.net>
Date: Mon, 8 Feb 2010 17:14:11 +1100
Cc: "Roy T. Fielding" <fielding@gbiv.com>, Apps Discuss <discuss@apps.ietf.org>, HTTP Working Group <ietf-http-wg@w3.org>
Message-Id: <86A801C9-1195-4ABE-9F49-A452993A78A0@mnot.net>
To: Julian Reschke <julian.reschke@gmx.de>


On 05/02/2010, at 11:24 PM, Julian Reschke wrote:
> 
> HTML5 doesn't have rev anymore anyway (well, unless the WG changes this again)
> 
> So why don't we just say that "rev" means what HTML2 used to say, and be done with it?

Because (IMO) implementation experience has shown that rev= is horribly confusing* to developers, and will cause more problems than it solves; it's much better to just register a different relation (even if that does offend those who like tidiness). 

I'd be OK with putting it in the BNF, giving the HTML2 explanation, and deprecating it in prose, though. 


* For those who dispute this, have a look at blog and other discussions of rev=canonical; it tripped up a lot of people.


--
Mark Nottingham     http://www.mnot.net/
Received on Monday, 8 February 2010 06:14:55 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Friday, 27 April 2012 06:51:16 GMT