Re: Web Rule Language - WRL vs SWRL

On Jun 22, 2005, at 7:55 PM, Michael Kifer wrote:

> Bijan,
>
> It looks like we agree about almost everything,

Maybe, maybe not :)

Part of the problem is we punted back up to a higher level (which I 
instigated). I think a "multiple stack" view is radically different and 
a fair bit of thought needs to be put into that and how to make things 
play well together.

Or, to put it another way: your half of the commonality supports, 
afaict, standardizing soon; my supports standardizing later.

> and I am afraid that
> this thread is loosing its entertainment value. :-)

Oh well. I can call you saprophytic , if that would help :)

> I just want to comment on one of your points:
>
>> The question is on the integration. Do we know enough to design an
>> integrated solution. If you look at my rules workshop position paper, 
>> I
>> argue for integration, but I don't think we're quite there yet
>> technically. Hence, I'd like to wait a bit.
>
> I think the (nonmonotonic) rule-based side of the stack is very mature 
> (I
> dare say more mature than the "other" side :-) and if W3 doesn't do it
> then somebody else will.

Interesting. Perhaps. I do think that a lighter weight standards body 
could be a good place to go. (I am interested in knowing what the 
Prolog vendors think of all this. Also, my impression from the rules 
workshop is 1) RuleML hasn't done what's needed for the business rules 
folks, which worries me and 2) the business rules folks aren't 
particularly interested in the stuff like SWSL-Rules (which does seems 
a shame, but is an issue)

> There is a good story about minimal integration
> using the black-box architecture, and this is very appropriate for
> standardization.

I'd like to take some credit for drawing attention to such approaches 
(for over a year now). Overall they've seemed to have had little 
uptake. I hope that our group will release some tools that will get 
people exploring them a bit more. There are a lot of issues, though. 
Whether research or standardization is a bit more difficult to judge (I 
tend to think of them as more researchy).

Cheers,
Bijan.

Received on Thursday, 23 June 2005 02:06:45 UTC