the 80% pipe

> We are talking about an exchange language, right? As I could understand
> from Sandro and others, the goal is to have a language L such that "80%" of
> the rule languages out there could be mapped into L in a semantically
> preserving way.
> 
> Then, with luck, if we have a rule set, R, in a real language L1, and we
> want to send those to a guy using another real language, L2, we would send
> the L2 guy the message map1(R) (which is a formula in L), and the L2 guy
> would then unmap it into R' = map2^{-1}(map1(R). map2^{-1} here is a LaTeX
> for the reverse mapping of map2.
> 
> Now, map1, map2 are supposed to be semantically-preserving, so running R in
> L1 should do the same as running R' at L2.

That sounds about right.  Part of the trick will be making sure that
people who want to can live within the 80%.  Thus the need for use
cases and user participation in the process.

> This is the vision and is very nice. Except that this is still a hard
> research issue, not clear whether it has a solution, and thus is not ready
> for standardization.

Good point.  There's probably not a lot we can say about this,
though.  Each organization will make its own judgement about how
practical this work is, weighed against how important it is.  

> Using FOL for the language L is wrong, as was pointed out repeatedly by
> several people. The reason why people propose FOL is because FOL seemingly
> has all the needed syntax. But this is a mistake. We need a
> *semantically-preserving* mapping, not just a syntactic mapping, and this
> is where FOL falls through.

Assertions have been made about the failings of FOL as an interlingua,
but I haven't seen anyone convinced.  Yes, FOL lacks the
MMS/CWA/UNA/etc of LP.  Yes, LP semantics are historically much more
common in rule lanuages.  Against that, I see Business Rules experts
saying they want FOL (eg Don Chapin on behalf of SBVR at the
Workshop), and I see Semantic Web experts saying they need
monotonicity (eg TimBL and Pat Hayes).  I'm hoping we can find a
win/win consensus here somewhere, but it's not yet clear to me where.
The must-have requirement on one side seems to be Defaults and on the
other to be Merging, and each side claims it can address the other
side's problem, but maybe not well enough.  I also may be getting
those requirements wrong; I'm just reading between the lines as best I
can, since this debate is still crystalizing.

It may make sense to write the charter to be agnostic on this question
and let the WG decide.  I used to think that was a terrible idea, but
now I'm starting to think it really is a highly technical and
not-so-political issue, so maybe the WG can handle it.

> Now you seem to be swerving away from the "interchange language" vision to
> a different topic -- a Web rules language.
> I think we better don't mix these two. They seem to me two different domains
> to be dealt with by two different groups.
...
> I think that the interchange WG should be different from the web rules
> language WG. Moreover, within the web rules language there are several
> stacks (see the aforesaid paper): A nonmon rules language, OWL-based stack,
> SPARQL. These are best handled as separate WGs, as there is enough work for
> all.

If there were two WG's, which would you do first?  The "web rule
language" one makes sense, but there doesn't seem to be enough support
for it.  It looks like there may be enough support for the interchange
WG, even though it faces some rather hard problems, I guess because
the payoff looks so good.  

     -- sandro

Received on Sunday, 28 August 2005 00:29:38 UTC