RE: [Ltru] Charter Discussion

> From: John Cowan [mailto:cowan@ccil.org]
> Sent: Wednesday, June 21, 2006 8:48 AM

> > I'd be stronger about it. Unless someone can present some good
> > reasons beforehand for even wanting to consider [ISO 11179], we should
> > definitely not put it in the charter.
> 
> The reason for *considering* it is that it's an international standard
> that is germane to what we are doing.  If we put it in the charter
> and no one presents sufficiently compelling arguments for using it,
> we simply say that n default of such arguments we have not used 11179.

There are probably two reasons for implementing a product to conform to an ISO standard:

a) it is a de facto industry standard in widespread adoption and for that basically is essential to success of the product (e.g. ISO 9000) 

b) it is a requirement for the product to be acceptable to governmental agencies

It will rarely be the case that the reason for implement a product to conform to an ISO standard is

c) even though it's not widely adopted it holds fantastic innovations with cool benefits

In our situation: 

- I see no indication that (a) holds wrt ISO 11179. 

- Re (b), it's possible that some agencies such as UNESCO might get hooked on ISO 11179, but at present I see no indication that IETF standards-track and best-practice documents are considered insufficiently propped up for governmental acceptance except perhaps in a very limited set of currently-hot areas, such as DNS -- and even in those cases I don't see adoption of something like ISO 11179 being a particular point of government concern.

- Re (c), I don't see any strongly-felt need in our work that ISO 11179 provides a solution for. IIUC, it defines a model for creation of meta-data registries -- both a conceptual model for the information system that constitutes the registry (the infrastructure of the registry as well as the content therein) as well as processes for the maintenance and administration. We already have a registry in place, one that has a fairly-well defined conceptual model (in spite of current debate over certain details) and also established processes for maintenance and administration -- processes with respect to which we have limitations in our flexibility to change. Now, I'm sure that if we studied ISO 11179 we would gain a better understanding of the nature of our endeavour and that we would find ways to improve upon what we have. But, I cannot say whether the cost/benefit ratio would be on the benefit side. The most I can say is that I have not seen reasons to set an expectation of greater benefit than cost.

So, I'm not in principle opposed to use of ISO 11179 and would certainly be open to someone providing a concrete explanation of how this endeavour would benefit from it. I just am not aware of any compelling benefits at this time. 

(Which is basically what I said about 18 months ago when a certain individual was insisting RFC3066bis needed to incorporate conformance to ISO 11179. Oh, and as I also said then, I don't think ISO 11179 actually defines any conformance criteria, so I don't think one can really talk about "conformance to ISO 11179" in any formal sense.)

Peter Constable

_______________________________________________
Ltru mailing list
Ltru@ietf.org
https://www1.ietf.org/mailman/listinfo/ltru