Re: X.500, Naming and the Internet

[[e-mail]]

> >True, even with oids as DNs, the Directory can fulfill its mapping function
> >since the 'search' operation is not limited to examining distinguished
> >attributes only. However 'search' is not the only legal operation
> >allowed in the Directory. A significant use of the Directory today
> >is the browsing of the data in there.
> 
> What about "search for ObjectClass=Organisation, return Organisation-Name"?

The problem with this is that browsing involves navigation of the 
tree. It is not sufficient only to 'list', but also necessary for
the user to be able to view what comes back as the names of nodes
in the trees so that the user can further navigate the tree.

Also, nondistinguished attribute values don't have to be unique. What
happens when multiple entries come back looking the same because they
happen to have the same (nondistinguished) organization name? Then you
need to return more information on the entry. If this is carried to its
extreme, the entire entry needs to be returned, which is undesirable.

> Looks like "list" is a subset of search...

No arguments here. The 'list' and 'read' operations are pretty much
subsets of 'search'.

> >> 	<DN: INT=1; INT=3; INT=250; INT=2; CN=Christian Huitema;>
> >
> >What can I say? Aaaarrrrgggghh!! :-)
> >
> Ok, I admit this is extreme in the directory context -- although it is not
> in an "OID directory" world. And presented in the wrong order too. But, then:
> 	<DN= CN=Christian Huitema; O=INRIA; O=FNET; O=Internet>
> looks perfectly reasonable...

I agree, this looks perfectly reasonable, assuming that this is what the
civil infrastructure in France looks like.

Actually, it doesn't really matter whether it looks 'reasonable' or not,
as long as the infrastructure created by the appropriate registration
authority is reflected in the DIT. In the case of the civil infrastructure,
this means that 'reasonable' is whatever the actual civil infrastructure
in France looks like.

At the risk of being excessively pedantic, the point I'm trying to
get across is that the Directory should reflect the infrastructure
created by the appropriate existing registration authorities. ISO does
not have the right to create civil infrastructure in France, the French
government does. Conversely, ISO *does* have the right to create oid
infrastructure, so the ISO oid tree should be reflected if we ever store
oids as objects in the Directory.


Wengyik