Re: DBIS commentary

Steven Legg <[email protected]> Fri, 27 Nov 2015 16:40:08 +1100
Newsgroups gmane.ietf.ldapext
Message-ID <[email protected]>
Hi Mark & Jordan,

On 26/11/2015 10:01 PM, Bannister, Mark wrote:
[snip]

>> Footnote 1 says that published classes may not be redefined. Although not all changes seem permissible,
>> some changes (e.g. addition of MAY elements) seem like they could be allowed.  Is there a reference for
>> this restriction?  Note that between X.521-1993 and X.521-2012 the TelecommunicationAttributeSet had
>> an attribute deleted.
>
> I haven't looked for a reference to this restriction, I had learned it from seeing previous exchanges
> on the ldapext mailing list with Kurt Zeilenga.  Here is one such exchange, regarding changing the
> schema declaration of 'posixGroup' in RFC2307bis:
>
> https://www.ietf.org/mail-archive/web/ldapext/current/msg02072.html

The rules of extensibility for object classes in X.519-2008, Clause 12.4 say:

     "Optional user attributes may be added to an existing object class without
      assigning a new object identifier."

Removing an attribute isn't covered, so by omission, it isn't allowed.

I don't recall the LDAP specifications having anything specific to say on the
matter, though I suspect the prevailing attitude is that you can't change an
object class without changing the OID.

The teletexTerminalIdentifier attribute was removed from TelecommunicationAttributeSet
because the service and/or the specifications were defunct, but it was a violation
of X.500's own extensibility rules. IMO the attribute should have just been made
obsolete. To get around possible interoperability problems I treat the new object
class definitions minus teletexTerminalIdentifier as the "original" definitions so
if I encounter a definition that includes teletexTerminalIdentifier it looks like
an allowed extension to the object class.

[snip]

Regards,
Steven