[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: Listing request names



Paul Hoffman / IMC said this:
> At 08:43 PM 11/23/97 -0500, Michael Mealling wrote:
> >Only one OID is required. LDAP, X.500 and friends use OIDs to identify
> >their schema but other directory services do not. This service is meant
> >for all directory services via any protocol so there will most definitly
> >be the case where someone is registering a schema where there are no 
> >OIDs at all.  
> 
> Boy, I'm confused then. The current drafts sure make it look like the
> registry is going to assign an OID for everything, not just LDAP schemas.
> That is, I assumed the "serial number" was in fact an OID.

It is. But it never really needed to be. The LDAP folx wanted it that
way. I was happy with it just being a URN, or simply a filename with
version and filetype in it.

> >No one should be using the service's OID as the primary name for schema.
> 
> Then why create our own OIDs at all? It's always bad to have two names for
> one item if the two names are controlled by different people. If people
> agree that the originator shouldn't be using our OIDs, I'd say we should
> just register what people send us and not create an "internal" OID, or even
> an internal serial number. What is the need for the serial number, which is
> just a shadow name for the name they gave?
> 

The serial number was there to serve as version control, not as a _name_ in
the sense that LDAP uses OIDs. We needed something to be the serial
number for that reason. The LDAP folx wanted everything to use OIDs and
we didn't see a reason not to so we went with it.

Don't think of it as another name. Think of it as more of the RFC number.
E.g. the Internet White Pages Schema RFC is RFC3456 (I can't remember what
it is off the top of my head). By calling it RFC3456 we're not saying that
the IWPS has two names. Just that the IETF calls it that for the sake of
keeping its RFC numbers straight.

You could not call it an OID and it still works just fine. I do agree
that the draft needs to tone itself down a bit when it comes to calling
that thing a NAME for the schema. Sure, its a name in the technical sense,
just not in the practical sense.

-MM

-- 
------------------------------------------------------------------------------
Michael Mealling	| 505 Huntmar Park Drive       | Phone:  (703)742-0400
Software Engineer	| Herndon, VA 22070	       | Fax:    (703)742-9552
Network Solutions	| <URL:http://www.netsol.com>  | michaelm@xxxxxxxxxx

----- End of forwarded message from michael -----

-- 
------------------------------------------------------------------------------
Michael Mealling	| 505 Huntmar Park Drive       | Phone:  (703)742-0400
Software Engineer	| Herndon, VA 22070	       | Fax:    (703)742-9552
Network Solutions	| <URL:http://www.netsol.com>  | michaelm@xxxxxxxxxx