Re: LDAPv4? (Was: Updating "core" specification)
"Ed Reed" <[email protected]>
| Newsgroups | gmane.ietf.ldup |
|---|---|
| Message-ID | <[email protected]> |
I think this depends on "how much change is too much". There has been so much variation in various vendors implementations of LDAPv3 that I don't have any trouble at all in asserting, in LDUP, that certain practices or conventions are expected from LDAPv3 servers that support LDUP, even if such vary, in minor ways, from the canonical core specifications. To provide a "reasonable interpretation" of LDAP data consistency constraints so that the LDUP protocol and URP can deliver utility to their users seems "within scope". Adding a SQL-like "JOIN" operation (which the X.500 community are looking at, or have done something about), dictating universal support of "families of entries", and so forth seem "out of scope". LDUP should not make such major expectations on solution providers, and should not dictate their use, or changes to their definitions. In fact, I don't even worry too much about whether LDUP can support them. Transaction support, as we've discussed, belongs in the experimental arena, in my opinion, so changes to the LDAPv3 core specifications to support them seem, for instance, out of scope for LDUP to me. I think the consistency model and URPs ability to support it, or the changes to it necessary for URP to work, are an important area of discussion, here. The Model (Architecture) document includes a long addendum in which we have attempted to address such questions. Its review would be welcome. Bottom line - LDAPv4 is out of scope. The subtle changes to LDAPv3 to work in a multi-master environment, with the necessary adjustments to data consistency expectations clients should have when working in multiple-writer/multiple-replica environments are in-scope. With that in mind, Kurt's text... This group may extend the LDAPv3 "core" specification as necessary to deliver LDAP Replication. However, updating the LDAPv3 "core" specification is beyond scope of this working group. makes sense to me, and is consistent with what I think I'm working to accomplish. Ed Parenthetically, I can see the importance of reaching agreement on this in light of Kurt's other responsibilities to LDAPv3bis. ================= Ed Reed Reed-Matthews, Inc. +1 585 624 2402 http://www.Reed-Matthews.COM Note: Area code is 585 >>> "Kurt D. Zeilenga" <[email protected]> 11/16/01 05:56PM >>> (This is a serious question, I am still trying to figure out what is beyond scope of this working group.) Is engineering a version 4 of LDAP within scope? At 10:32 AM 2001-11-16, Chris Apple wrote: >That's simple enough to do. We'll add... >> This group may update the LDAPv3 "core" specification >> as necessary to deliver LDAP Replication.