Re: Analyzing Use Cases and Requirements (or What is the RIF?)

Allen,

ALthough your message makes some very good points, it also contains some 
quite puzzling statements:

Ginsberg, Allen wrote:
> 
> [...]
> 
> The main issue that needs to be addressed is whether the WG endorses
> the view of the RIF specification as incorporating metalinguistic
> features and interchange-enabling features in addition to rule-language
> features.

At the very least, it should be the reverse! Whether the WG endorses the 
view of the RIF specification as incorporating rule-language features in 
addition to interchange-enabling features might indeed be an issue. But 
the charter is very clear about the RIF being about enabling 
interchange: even the name carefully avoids the word "language" to 
stress the focus on the interchange-enabling features!

>  Is the RIF "just" a rule language, albeit one that
> provides a well-reasoned standard, or is
> the RIF more than that?  And, if so, what is the "more?"
> 
> [...] 
> 
> [...] So at least some of the "more" that could be part of the
> overall RIF specification is whatever features would be necessary to
> enable (some degree of) standard-based mechanized rule translation
> either between non-RIF languages and RIF, or between different RIF
> dialects
> 
> Looking over the use cases that have been submitted, it seems that a
> fair number imply a view of the RIF specification as involving 
> something "more" than a rule language.  Not all of these use cases are
> stated explicitly in terms of the issue of "translate and transfer,"
> but many do talk about properties of rules that would be important to
> take into account in attempting to share rules.

I am not sure how to understand "non-RIF languages". If you mean 
"existing or future rule languages" other than the RIF, then the "more" 
is actually the RIF: a major deliverable is indeed "A W3C Recommendation 
providing technical specifications of the interchange format, suitable 
for implementers of rule engines and rule language translation 
software". "Rule translation" being between the rule languages 
implemented by the "rule engines" and the RIF, to enable rule 
interchange between rule engines using different rule languages 
(including ones with the same of similar semantics, btw, as a major use 
case).

And, anyway, I would rather say that the RIF is "less" than a rule 
language, if anything (actually, I would just say that it is something 
different, hence my question at the kick-off whether we all agreed that 
the WG was about specifying a standard rule interchange format and not a 
satndard rule language for the Web [1]).

> The point is that there is a big difference between a rule-language
> feature, e.g., use of classical negation, and a feature that 
> allows one to represent or talk ABOUT such features.  The latter are
> "metalinguistic" or "meta-level" features, since they are not part
> of the rule language as such.   

Indeed. A RIF does not even (necessarily) need a semantics (as a rule 
language), but it needs to be able to specify unambiguously what is the 
intended semantics of the interchanged rules (one way to do this being 
to give it a precise semantics of its own, see the discussion on RIF VS 
Rule language on the public mailing list).

My understanding of phase 1 is actually that it (phase 1) is restricted 
to Horn logic expressiveness precisely to allow us to focus on the 
basics (syntax, extensibility, conformance, RDF/OWL compatibility, data 
access etc) without being burdened by difficult or contentious questions 
about the semantics of most useful features of relevant rule languages 
(and how it can be expressed in the RIF). Also see my comment at the 
kick-off re the phase 1 RIF not being really useful [2]: if we do our 
job well in phase 1, we will have a sound basis on which to specify the 
more widely useful features, in phase 2.

My intention was to keep that message short. Sigh!

Christian

[1] http://www.w3.org/2005/12/08-rif-irc at 19:49:28-34
[2] same place, at 19:10:44

NB: A paragraph from the first version of the draft charter that was 
publicly discussed, back in July 05 (version 1.64; the charter is 
version 1.7), stated that: "The Working Group is to specify an 
interlingua, a common format into which existing rule languages can be 
mapped. This interlingua may itself be considered a rule language, and 
may be supported natively by some rule systems in the future, but 
interlingua features are more important than features which make the 
language itself easy to use directly."

Sandro, it might be interesting to know why that paragraph was dropped: 
is it because there was disagreement to that statement, or for other 
reasons (e.g. it was seen as confusing or redundant or whatever)?

Received on Thursday, 5 January 2006 18:24:07 UTC