[sc34wg3] Documenting merging rules in TMDM

[Kal Ahmed]

Patrick,

On Sat, 2004-03-13 at 16:22, Patrick Durusau wrote:
> Steve,
> 
> Steve Pepper wrote:
> > * Patrick:
> > 
> > | Just a quick reply with more to follow.
> > 
> > Answers to my questions in your capacity as leader and
> > spokesperson for the US National Body, I hope? :-)
> > 
> Too early in the weekend for that sort of responsiblity. :-)
> 
> > | There is a difference between:
> > | 
> > | 1. Merging rules written in a known syntax (your case)
> > | 
> > | and,
> > | 
> > | 2. Merging rules that are embedded in an application.
> > | 
> > | The first allows a customer to make a reasoned judgment of cost and risk 
> > | in migrating from one topic maps software application to another.
> > | 
> > | The second does not.
> > | 
> > | Does that help?
> > 
> > Help in which way? I agree with this entirely, so it establishes
> > that we are on the same page in that regard: Documenting merging
> > rules is preferable to not documenting them.
> > 
> > But are you saying that you would accept Kal's approach, even
> > though it delegates responsibility for defining how to document
> > merging rules to TMCL?
> > 
> No.
> 
> Kal's approach is forced by the limitations of the TMDM that you note 
> below.
> 
> If one follows the TMDM, then Kal's suggestion may be a good fix. (still 
> have the problem of vendor lock by embedding merging rules in software, 
> even if Kal's suggestion is a good theoretical answer to the 
> documentation problem)
> 

I think you are assuming that the TMCL rules would be somehow hidden
from the user - that is not what I am suggesting. I am suggesting that a
conformant implementation of the TMDM with a conformant implementation
of TMCL could use rules expressed in TMCL to perform merging, assuming
such functionality is made a part of TMCL. As for vendor lock-inm as
long as the rules are known (or can be deduced), then one can always
replicate them with another conformant application.

> If one departs from the TMDM, then Kal's suggestion, to the extent that 
> TMCL follows the TMDM, is less useful.
> 
Thats true, TMCL would not constrain RDF or XML or MS Word or any other
information format. But I don't think thats a major restriction. I was
talking about constraining topic maps as described by the TMDM.

Cheers,

Kal
-- 
Kal Ahmed <kal@techquila.com>
techquila