Re: Manchester syntax track

[I'm not sure why you switched from public-owl-wg to public-owl-dev,
Uli, but I guess I'll reply where you sent it...]

> On 9 Dec 2008, at 20:22, Sandro Hawke wrote:
>
> > My sense is that making it Rec Track would amount to the WG saying
> > "everyone who wants a human-readable serialization for OWL 2 SHOULD  
> > use
> > the Manchester Syntax."
> 
> Are you sure? How comes it wouldn't say "everyone who wants a human- 
> readable serialization for OWL 2 SHOULD *consider* to use the  
> Manchester Syntax -- yet everybody is free to invent their own, e.g.,  
> application-specific or language-specific one. Moreover, if you claim  
> you support Manchester Syntax, this is what you should be supporting."

Yes, I'm sure. 

I guess there's some ambiguity here:

    W3C publishes Recommendations when it believes that the ideas in the
    technical report are appropriate for widespread deployment and that
    they promote W3C's mission. 
      -- http://www.w3.org/2005/10/Process-20051014/tr.html#rec-publication

It's not clear there exactly how widespead they mean, but my
understanding is that a specification becoming a Recommendation reflects
consensus within the Consortium that *everyone* who wants to
interoperate (in some area of Web functionality) SHOULD conform to that
spec.  It's all about network effects -- there's great advantage to
having a single language -- and that's what Recs are for.  The
consequence of that, however, is that when you don't have such
consensus, you can't issue a Recommendation.  (And I suspect that's the
case for Manchester Syntax, but I don't really know.)

If we just want people to "consider" it, and to have a stable
specification they can implement if they like, then it's published as a
Member Submission, Team Submission, or Working Group Note.  (The choice
among those three being determined by who is proposing it.  With these
publicatiions, there's no need to assess consensus across W3C that it's
the right solution.)

> I think there are 3 points to this:
> 
> - [for implementors] a  document with describes Manchester syntax  
> exactly, so that implementors can agree.
> 
> - [for users] if you want, e.g., to show snippets of your ontology in  
> a paper, then it would make probably a lot of sense to use Manchester  
> syntax (and this is where I like the "should") - and it should be ok  
> to use it (and not have a "hey, why didn't you use one of the standard  
> ones?" review)
> 
> - [for users ] if you want to build some editor for a medical ontology  
> (or a chemistry one or what have you), then you still could start with  
> Manchester Syntax, but you are free to use whatever use most  
> suitable...?!
> 
> 
> 
> > Even if, in their text, our documents
> > explicitely disclaims this idea, the word "Recommendation" carries too
> > much weight to avoid this reading.    And I don't think the WG wants  
> > to
> > say that, but I could be wrong.
> >
> 
> ...I would want to say that/my slightly rephrased version of it.  
> Cheers, Uli

It seems to me that can all be done, just fine, with a Working Group
Note.

     -- Sandro

Received on Wednesday, 10 December 2008 16:39:07 UTC