[DDI-SRG] ID uniqueness and URN

[Wendy Thomas]

file this as a bug for discussion. I prefer to have the fact that this was 
raised and the result of the discussion documented in Mantis.

Wendy

On Tue, 6 Jan 2009, Pascal Heus wrote:

> All:
> I have an issue with DDI documentation recommendations on ID uniqueness
> and URN generation:
> (1) For identifiers, we require an ID to be unique within it's maintainable
> (2) For URN, we recommend to include the complete object hierarchy, from
> the maintainable to the object itself
>
> The (2) requirement seem to be an overkill given that, since the ID is
> uniqyue within the maintainable, we do not actually need to include the
> full hierarchy to generate a unique URN. Unless there is a particular
> benefit in including the hierarchy in the URN (which I don't see), this
> has the drawback to create unnecessary long URN strings. The
> Manitainable ID + Object ID is sufficient, intermediate object do not
> need to be identified. I'm not sure if we have many identifiable
> elements that are at the 3rd, 4th or deeper level in the schema but I
> would suggest to relax the requirement to include the full hierarchy in
> the URN.
>
> Another way to approach this would be to only require the ID to be
> unique within it's parent identifiable. This would make it easier to
> generate IDs. I would however rather prefer shorter URNs.
>
> Let me know if I should file a bug about this for further discussion
> .
> thanks
> *P
> _______________________________________________
> DDI-SRG mailing list
> DDI-SRG at icpsr.umich.edu
> http://www.icpsr.umich.edu/mailman/listinfo/ddi-srg
>

Wendy L. Thomas                          Phone: +1 612.624.4389
Data Access Core Director		 Fax:   +1 612.626.8375
Minnesota Population Center              Email: wlt at pop.umn.edu
University of Minnesota
50 Willey Hall
225 19th Avenue South
Minneapolis, MN 55455