W3C home > Mailing lists > Public > public-rif-wg@w3.org > May 2008

Re: planning RIF's future

From: Michael Kifer <kifer@cs.sunysb.edu>
Date: Thu, 08 May 2008 16:41:20 -0400
To: Sandro Hawke <sandro@w3.org>
Cc: public-rif-wg@w3.org
Message-ID: <30429.1210279280@cs.sunysb.edu>


> 
> > Looks like a good plan, but whether 1 year is enough depends on the
> > speed of the progress.  As you know, sometimes we hit unexpected snags.
> > 
> > I have some questions/comments:
> > 
> > 1.  You wrote:
> >         - BLD, DTB, and SWC  -- finish up what're working on now;
> > 
> >     Was the omission of FLD intentional?  Did you mean BLD/FLD?
> >     What about UC&R?
> 
> I left out FLD and UCR, accidentally, because I don't think if them as
> bearing directly on implementations and user functionality, and that's
> what I was thinking about.
> 
> I'm not really sure how FLD should be handled procedurally.  While it is
> used normatively in defining a dialect, does it make sense to say FLD,
> by itself, is (or is not) normative, or some implementation is/is-not
> compliant with FLD? 

For implementations, there is not obvious sense in which they can be
compliant with FLD, but dialects must be compliant.

> If it's not normative in itself, we may want to
> just publish it as a WG Note, which does not need the full level of
> external review that a Recommendation does.  (We have the same option
> with UCR.)  In both cases, if we want to update and republish a newer
> Note later, that's pretty easy.  This makes sense if people aren't
> actually using it as a spec for interoperability, relying on it being
> stable.  (as I say that, I'm thinking FLD probably does need to be a
> Rec, but I'm concerned about calling it "done" before LP, FOL, etc are
> farther along.)

In that sense, FLD will never be considered "done". But if it is not made
normative then there will be no guard that will prevent non-compliant
dialects, and then the entire RIF framework will come apart.

The same is basically with the XTAN stuff. It must be there, but if future
needs will prove that it is inadequate then it must be extended. But such
changes should not be done without a very good reason.

FLD and the extensibility framework are sort of like a constitution: laws
must be compliant with it, but noone can actually be sentenced based on a
constitutional article (it does not specify punishment). If there is a need
to introduce a law that conflicts with a relevant constitution, and there is
strong support for it, then the constitution is changed.


	--michael  


> > 2.  When you are talking about 1 year, do you mean May 2009?
> 
> Yes, through May 31, 2009.
> 
> >     Is it a kind of 2nd phase?
> 
> Kind of, but I think "Phase I" / "Phase II" distinction in the Charter
> isn't very helpful at this point.  I think it makes sense to just focus
> on what folks need and we can accomplish in the next year.  It may be, I
> suppose, that we end up required to do some kind of more-strategic
> longer-term plan, but I'd rather avoid that.
> 
> > 3.  Getting LP + FOL + PRD dialects to the last call in < 1 year is very
> >     ambitious, but I assume that we can get an extension on that.
> >     (In fact, LP and FOL are mainly done as part of FLD, but significant
> >     work is still ahead.)
> 
> Yeah.  There are several unknowns here.  I think if everything goes
> well, and we keep up the pace, this plan is doable.  I'm concerned that
> if we slow down very much at all (because we're each putting less into
> RIF, because we get lost in unimportant details, or because we get too
> perfectionist), we'll start to lose the participants who really need to
> get something usable out of this (cf Gary's comment [1]).
> 
>        -- Sandro
> 
> [1] http://lists.w3.org/Archives/Public/public-rif-wg/2008May/0078.html
> 
Received on Thursday, 8 May 2008 20:44:06 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Tuesday, 2 June 2009 18:33:48 GMT