Re: [tangle] getting the semweb exactly wrong

Jan--

Comments below.

Jan Algermissen wrote:
> Frank,
> 
> there's a lot of value in your reply which I hope to be able to  address 
> tomorrow. Just a quick comment now:
> 
> On Jan 3, 2006, at 11:12 PM, Frank Manola wrote:
> 
>> Anyway, I think the trade-offs involved in using RDF in the  Semantic 
>> Web are reasonable ones.  But I think sometimes that the  differences 
>> with prior work (including work prior to the relational  model) are 
>> sometimes exaggerated.
> 
> 
> I think that the most important question to be answered is this:
> 
> What problems that I today have with the relational model would be  
> better addressed if I used RDF+OWL?
> 
> IOW, what is the killer argument for abandoning my RDBMS and using an  
> RDF store instead?

Be careful here.  The arguments about what *model* to use are one thing, 
and those about what *DBMS* or *storage format* to use, while related, 
are different.  RDF is a terrific integrating model, but I don't like 
the idea of forcing everyone to store everything that way.  I want the 
ability to take data in any form people find useful, for whatever reason 
(relational tables, XML, whatever) and *interpreting it* as RDF, without 
the need to necessarily store it that way.

> 
> I think the (killer?) argument is evolvability and the standard use  
> case is a company having some hundreds of report creating 'scripts'  
> stuffed with SQL statements and thus a deadly dependency on the  
> relational database schema in use. Assuming 1 to 5 man days to  migrate 
> each report evolving the schema becomes next to impossible  leading to 
> zero evolvability.
> 
> How would the situation be different if the hundreds of reports  
> depended on RDF schemas (and OWL ontologies) instead?
> 
> Or: how would the use of RDF/OWL (as opposed to an RDBMS schema)  enable 
> what one might call 'late binding between query and data  model" so that 
> the data model could evolve wiothout breaking the  queries?
> 
> Jan
> 

I think this would be a great case (one of many!) to analyze in detail, 
so as to be able to back up this argument with some concrete data;  and 
also to be able to identify the limits of the advantages provided by RDF 
(for instance, there's going to be a limit on the ability of the "data 
model" to evolve without breaking the queries, RDF or not).

--Frank

Received on Tuesday, 3 January 2006 23:25:28 UTC