Re: open issues in RIF-BLD, RIF-RDF: metadata, built-ins

Michael Kifer wrote:
> The proposal in [2] is not quite clear to me. If I understand it correctly,
> it is not sufficiently flexible.

Care to be more specific?

> Moreover, I do not understand why are you talking about external data sets
> and not about external rulesets. A data set is a special case of a rule
> set.

I've taken it for granted that there will be a rule import mechanism 
(and indeed an associated module mechanism for rules). I've also taken 
it for granted that that will be part of the rule language. Whereas 
information on source datasets is just metadata to guide an application 
builder and is not currently associated with a formal processing model.

The case where the data is in RIF format is indeed a special case of 
that. Which is why that page says:

[[[
Some datasets may be simply RIF rulesets comprising just facts. In that 
case no further information is needed. Referencing a global dataset in 
that case can be achieved using a simple ruleset import mechanism.

We could stop there and say that how datasets in some 
application-specific datamodel get translated to RIF fact-sets is 
outside the scope of RIF.

However, there seems to be a desire for RIF to be able to reference 
datasets in application-specific formats and at least document what the 
format is.
]]]

I'm not sure that I care that much about this either way. The reason 
that page exists is that Christian felt that dataset identification was 
an important topic at F2F(N-2) and I agreed to put a sketch together. It 
is fine by me if the WG wants drop it. It is also fine by me if someone 
wants to state different requirements not met by this mechanism or 
propose alternative mechanisms.

On the other hand metadata inclusion in rules and rulesets is a 
necessary requirement and still needs to be done. If I get time I'll try 
to put some notes together on that though my time available for RIF 
related work is rather constrained at the moment.

Dave
-- 
Hewlett-Packard Limited
Registered Office: Cain Road, Bracknell, Berks RG12 1HN
Registered No: 690597 England

> 
> 
> 	--michael  
> 
> 
>> i) Metadata in RIF / referring to external data sets, RDF
>> ===
>> it has long been recognized that RIF needs a mechanism for specifying
>> metadata.  In addition, the current proposal for referring to external
>> data sets and data models [2] relies on metadata for this reference.
>> However, the BLD language does not have it.  I propose adding adding
>> metadata to the language as soon as possible.
>> Additionally, the working group needs to agree on the mechanism and
>> vocabulary to use for referring to external data sets. Personally, I am
>> fine with following the proposal in [2]. Fortunately, this is also on
>> the agenda for today's telephone conference :-)
>>
>> I really think these issues need to be resolved before the publication
>> of the next working draft.
>>
> .....
>> [1] http://lists.w3.org/Archives/Public/public-rif-wg/2007Sep/0001.html
>> [2] http://www.w3.org/2005/rules/wg/wiki/Arch/Data_Sets
>> [3]
>> http://www.w3.org/2005/rules/wg/wiki/Arch/RDF#head-cb9271092ef7391c51020efbc1900dfc64edbd08
>> [4] http://lists.w3.org/Archives/Public/public-rif-wg/2007Jul/0030.html
>> [5] http://www.w3.org/2005/rules/wg/wiki/List_of_functions_and_operators
>> [6] http://lists.w3.org/Archives/Public/public-rif-wg/2007Jul/0038.html
>> -- 
>> Jos de Bruijn            debruijn@inf.unibz.it
>>                       http://www.debruijn.net/
>

Received on Saturday, 8 September 2007 09:04:04 UTC