Reminder of San Diego and Minneapolis discussions

["Timothy Hahn" <hahnt@us.ibm.com>]

Hi all,

Since San Diego and Minneapolis, we have only made a little progress.  This note is an attempt to remind everyone what we had decided to work on:

Description: procedures for merging and updating schema, including a discussion on removing existing schema and understanding when schema elements can be deleted
Responsible:  Ludovic, Poitou, Tim Hahn
Status: I-D produced: draft-poitou-ldap-schema-update-00.txt

Description: determine the allowable changes to existing schema elements, define "do no harm" operations, include a discussion of implications for existing data
Responsible: Mark Hinkley, Tim Hahn to get Bob Moore's document ( I have gotten the document but have not built the I-D)
Status: no progress

Description: define extensions to attribute type and object class ABNF to allow for specification of "unique" and "referential integrity"
Responsible: Jim Sermersheim,        Ludovic Poitou, Roger Harrison
Status: no progress

Description: define procedures for partitioning, i.e. show how different schemas can be applied to different areas of the DIT
Responsible: Tim Hahn
Status: I-D produced draft-hahn-schemapart-00.txt

Description: define a way for ensuring unique attribute type and object class names (not just OIDs)
Responsible: Mortezza ? Bob Joslin
Status: no progress

Description: discovery of attribute type options that are allowable in a server (was an oversight in LDAPv3 RFCs)
Responsible: Jim Sermersheim, Mark Wahl
Status: no progress

Description: grouping of LDAP schema pieces, packages of schema, listing dependencies between schema packages, versioning schema packages
Responsible: no one assigned as there exists an Informational RFC 2927 to describe one approach
Status: closed

Description: define how to describe schema as "first class objects" instead of "structured types".
Responsible: Roger Harrison, Brian Jarvis
Status: no progress

Description: define additional attributes for subschemasubentry. Attributes such as ditContentRules are ill-defined in the current RFCs. This work would clarify their definition and usage
Responsible: Steven Legg
Status: no progress

Description: guide LDUP WG on how to replicate schema
Responsible: joint ELSE and LDUP workgroup followers
Status: ongoing

Looking forward to hearing from you all on whether any progress has been/will be made here.

The more interest/work we get done here the better chance we will have of either a) getting a separate workgroup started or b) having the work covered under LDAPext.

While I'm here - by posting to the list - what are peoples' feelings on whether LDAPext should "sponsor" this or we should create our own workgroup?

Regards,
Tim Hahn

Internet: hahnt@us.ibm.com
Internal: Timothy Hahn/Endicott/IBM@IBMUS or IBMUSM00(HAHNT)
phone: 607.752.6388     tie-line: 8/852.6388
fax: 607.752.3681