Re: Issue-12 and the next UCR draft

[snip]
>> There are already a large number of in-use or proposed semantic  
>> web rule languages (CWM, Euler,

These are not rule languages, but rule engines. The language they  
support is N3.

>> WRL, SWRL, SWSL,

May wish to call out SWSL-Rules vs. SWSL-FOL explicitly.

>> JenaRules etc)[refs]. It will be possible to transmit some  
>> fragment of these rule languages via RIF. However that on its own  
>> does not provide interoperability - a SWRL engine will not be able  
>> to execute an arbitrary CWM ruleset, nor vice versa.
>> The working group is currently divided on whether this is  
>> sufficient or whether some further step is needed to give guidance  
>> to semantic web rule developers and implementers. We seek feedback  
>> from the community on this issue.
>> By phrasing this goal as "provide the basis for ..." we are  
>> indicating that there is unlikely be a single semantic web rule  
>> language and that RIF will not propose one. However, it also says  
>> that RIF should go further than minimal compatibility and try to  
>> bring some order to the chaos of semantic web rule languages.

Good luck :)

>> For example, this might take the form of a recommended profile or  
>> small number of dialects, with recommended sets of builtin  
>> predicates and functions. This would not suppress the continued  
>> invention of new rule languages for semantic web applications but  
>> would provide a common denominator that developers and  
>> implementers could agree upon as a useful core.

My current experience with SPARQL, on the relatively simpler issue of  
query languages and one where we pick a winner, makes me run  
screaming into the night ;)

Pick a winner that already has a clear syntax and semantics and well  
understood theory and just go with that, if you are going to take  
this route. Oh, let me be more clear, pick a winner that *already has  
a detailed well worked out specification*.

There are deep issues with BNodes that have hurt us in SPARQL, and I  
think most extant SemWeb rule languages largely punt on them. It  
would be good to deal with them properly. (e.g., are BNodes scoped to  
the document? Even when they appear in rules?)

Cheers,
Bijan.

Received on Wednesday, 13 September 2006 13:08:07 UTC