RE: ISSUE-4: Versioning, namespace URIs and MIME types

On Wed, 18 Feb 2009, Larry Masinter wrote:
> 
> I'm discussing whether there is to be a standard way of indicating which 
> mode is intended or desired.  I agree that in some circumstances, 
> allowing someone to say something about the mode intended can have 
> unintended side effects which also have to be mitigated, but the 
> possibility of those side effects isn't sufficient justification for 
> avoiding having a standard method.

Doesn't HTML5 already define the standard methods for user agents to 
determine which mode to use? I'm confused.


> > Then my original e-mail on this thread is relevant in that it shows 
> > that we only have two possible axes for versioning, namely the 
> > namespace and the tag name for whatever elements we want to version.
> 
> Well, no, those aren't the only two *possible* axes, those are the two 
> that you may prefer. DOCTYPE, version attributes and other mechanisms 
> are also *possible* axes, and no version mechanism by itself may be 
> sufficient.

The DOCTYPE axis and version attributes can't be used for triggering 
modes, since they're not available in the example I gave in:

   http://lists.w3.org/Archives/Public/public-html/2009Feb/0375.html

(Unless you are talking about modes that are intended to be non-conforming 
and not used, in which case they don't need to be available. This is the 
case with the quirks and limited-quirks modes we have currently.)


> > Versioning in general is bad idea IMHO, as discussed almost two years ago:
> >    http://lists.w3.org/Archives/Public/www-archive/2007Jun/0024.html 
> 
> Version designations inside web document formats when there are multiple 
> versions is a fundamental distributed network design, certainly over 30 
> years old. The earliest discussion I can readily find related to the web 
> is only 16 years old, though. See:
> 
> http://ksi.cpsc.ucalgary.ca/archives/WWW-TALK/www-talk-1992.messages/216.html
> 
> The principles of versioning discussed then have been underlying all 
> MIME registrations and format discussions in the intervening years.

I wouldn't consider MIME registrations a success story. I certainly 
wouldn't consider them evidence that versioning is a good thing. They are 
at best inconclusive on the matter, IMHO.

-- 
Ian Hickson               U+1047E                )\._.,--....,'``.    fL
http://ln.hixie.ch/       U+263A                /,   _.. \   _\  ;`._ ,.
Things that are impossible just take longer.   `._.-(,_..'--(,_..'`-.;.'

Received on Thursday, 19 February 2009 02:19:17 UTC