W3C home > Mailing lists > Public > public-html@w3.org > April 2007

Re: Separation of versioning concerns

From: James Graham <jg307@cam.ac.uk>
Date: Wed, 25 Apr 2007 14:45:43 +0100
Message-ID: <462F5B87.9070606@cam.ac.uk>
To: Eric Daspet <eric.daspet@survol.fr>
CC: public-html@w3.org

Eric Daspet wrote:
> 
> 
> Le Mer 25 avril 2007 13:45, Sam Ruby a écrit :
> 
>> [...] This group needs a way to say that this
>> particular representation of this particular page was produced
>> specifically for IE8.0 (or whatever) and may not render as intended by
> browsers that do not implement a rendering mode that is compatible with
> that particular version of that particular browser.
> 
> This assume that authors that rely on a specific version know that they
> are doing so. In my opinion, this is not the case. Most authors may rely
> on a bug and do not know that it is a bug. They won't use the explicit
> lockdown as they think they are "standard" (think of many people that say
> "I'm standard : look, it runs well on IE").
> 
> This suggestion is fine only if authors are aware that they are exploiting
> bugs and specific behaviour. I do not think we are in a such situation.

I accept that it's hard to know if one is relying on a bug in the general case. 
However, if we accept a little loss of precison, it is likely that we can do 
rather well. If a page meets some of the following criteria then the author 
would be well-advised to opt in to a single IE version:

1) The page only works in IE (arguably such a page is not part of the public web 
but part of a vendor specific walled-garden).
2) The page displays significantly differtently in IE and other major browsers.
3) The page uses some way of sending different code to IE and other browsers 
(e.g. CSS hacks, UA sniffing in javascript code, etc.).

Not that this list is rather conservative; an author might send different 
javascript to IE and other browsers but use object detection to ensure that if 
IE did gain support for the relevant features the site would continue to work 
unmodified.

We must also remember that people are capable of testing their sites during beta 
periods leading up to new releases of major browsers; if there is some simple 
switch that allows them to opt into the old rendering mode they can apply it 
then or fix the site.

This is not 2001 anymore, IE does not have a market share that allows people to 
build IE-only sites on the public web and I don't think we should operate under 
the assumption that it does.

-- 
"Eternity's a terrible thought. I mean, where's it all going to end?"
  -- Tom Stoppard, Rosencrantz and Guildenstern are Dead
Received on Wednesday, 25 April 2007 13:48:05 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Wednesday, 9 May 2012 00:15:53 GMT