WEBDAV Requirements Changes

At our working group meeting on Monday I presented a list of open issues
related to requirements. There was a brief but useful discussion.  In
addition, Yaron and Jim gave me some written comments on the requirements.
Before I actually make any changes, let me solicit everyone's opinions on
the issues I listed:

1. Attribute Search:  In or out of scope? 
2. Partial-Resource Locking:  Do we need changes / extensions to the HTTP
protocol or not? 
3. Lock Types:  What types of locks are needed? 
4. Semantics of Copy / Move:  New types of objects like collections, version
graphs, and metadata make the semantics of copy / move more complex.  What
should the requirements say about them?
5. Clarifications of HTTP DELETE / PUT methods:  New types of objects like
collections, version graphs, and metadata make the semantics of PUT and
DELETE more complex.  What should the requirements say about them?
6. Link Integrity:  Much of WEBDAV is architected around links.  This makes
issues of link maintenance and integrity pressing.  Should the requirements
say anything about it?
7. Structured Documents: Not currently mentioned in the requirements, but a
proposal for structured documents is under discussion.  Should structured
documents be added to the requirements?
8. Reservations and Versioning: Should reservations be discussed separately
from versioning? Do we want to support reservations on resources that are
not version tree handles? 
9. Diff / Merge: Do we want these or not? 
10. EMail / Disconnected Operation: At the moment we just say that the
WEBDAV spec will not do anything to prevent disconnected operation or EMail.
Is this enough?
11. Mandatory / Optional Functionality: Should the requirements say anything
about which functionality is optional / mandatory for WEBDAV compliance? 
12. Simplicity vs. Flexibility: As a design principle, how should these be
balanced?

In addition the group in the room pressed us to add requirements for
authentication, access control, and interoperability with existing security
protocols.

Yaron and Jim asked for the following changes: 

1. Remove all requirements about the semantics of copy and move.  Give a
simple, one-sentence definition of the operation and leave it at that.
2. Remove all discussion of implications for existing HTTP methods.  Simply
say that we are extending HTTP, and let the reader conclude that it is our
responsibility to specify any required changes / extensions to existing HTTP
methods.
3. Do not say which functionality is mandatory or optional.
4. Attribute search should be IN scope.
5. Change "relationship" to "link" throughout.
6. Remove the statement that attributes are resources.
7. Say nothing about multi-resource locking or partial-resource locking.
8. Provide better rationale for making server support for locking optional,
or else remove the requirement that this support be optional.
9. Remove the statement about who can do an UNLOCK.
10. Combine the discussion of reservations with the discussion of locking.
Treat reservations as advisory locks. Also talk about shared vs. exclusive
write locks.
11. Get rid of the paragraph in 5.5.2 that talks about multi-step source
processing.
12. Get rid of the requirements that copy and move leave audit trails.
13. Get rid of 5.7.3 Delete.
14. Remove the clause in 5.9.1.3 that allows servers to restrict the WEBDAV
versioning model.
15. Remove requirement 5.9.2.8 that the client be able to request that the
server generate a version id for a new member of a version graph.
16. Change 5.9.2.13 to say that it must be possible for a client to
*request* (not *get*) a server-side diff.
17. Remove requirement 5.9.2.14 for server-side merge.

Please comment on any of these issues and recommendations.  I'll try to take
Jim's suggestion that we revise the requirements monthly between now and
August, when they will be submitted as an informational rfc.

--Judy
Name:			Judith A. Slein
E-Mail:			slein@wrc.xerox.com
Internal Phone:  	8*222-5169
External Phone:		(716) 422-5169
Fax:			(716) 265-7133
MailStop:		128-29E

Received on Thursday, 10 April 1997 14:47:28 UTC