[Date Prev][Date Next] [Chronological] [Thread] [Top]

Re: Proposal: A possibly feasible method to invent OIDs for incomplete entries



At 01:44 PM 11/15/99 +0100, Julio Sanchez Fernandez wrote:
>
>Well, let me summarize the rationale.  We absolutely *need* to have OIDs
>on all schema data that require it and that we publish.

And we need to publish schema for all items to which a client modify.

>For attribute
>types and object classes whare no OID is known, we have several options:
>
>	- Reject them at definition time

This is the only reasonable choice.
>	- Not publish them

As we MUST publish schema to clients which can modify entries
defined using that schema.  And we SHOULD publish schema for
entries which are readable.

>All these options are hard on administrators and/or clients making use of
>the information.  Publishing broken definitions, even if it seems a common
>"solution" is really against the standard and may break clients.

Yes. We really must publish schema and that schema must conform to
standard-track specifications.

>We may
>be able to parse that ("be liberal") but we cannot inflict it upon others
>("be conservative").

Exactly, we must provide clients with schema per the specs.

>The current situation for 2.0 is that compliance is left to the
>administrator.

Yes.  And in general, administators will always be able to
deploy non-compliant servers.  We should view our job as
providing the tools necessary to deploy complaint software
but not necessarily restrict administrators from (ab)using
the software.

>If she wants to have a compliant server, she must find the
>needed OIDs or assign them herself.

Yes.  We need to encourage OID assignment.

>We may leave it at that.  But maybe we might help.

Yes.  We should provide detail instructions on how to obtain
an OID arc for her enterprise.

>First, let's see what are the relevant requirements for assigned OIDs:

I do not want to get into this business.  Getting a OID from
IANA is as easy as filling out a web form and waiting for the
e-mail response.

I would rather we just bitch when OIDs are not provided...
maybe providing the URL to the administrator each time.

>Opinions?

I do not believe it wise to programmatically generate OIDs.

Kurt



----
Kurt D. Zeilenga		<kurt@boolean.net>
Net Boolean Incorporated	<http://www.boolean.net/>