W3C home > Mailing lists > Public > w3c-dist-auth@w3.org > July to September 1998

Re: Baby steps

From: Sam Ruby <rubys@us.ibm.com>
Date: Fri, 7 Aug 1998 06:32:50 -0400
To: <w3c-dist-auth@w3.org>
Message-ID: <5040200017935772000002L022*@MHS>
I now understand and appreciate how we got to the current proposal.

This being said, I would like to propose this as our first scenario - even
thought it isn't recognized as such yet  ;-).

I would also like to extend the scenario to involve deletes and renames.

I also would like to explore the following aspects for each scenario:

1.  Is the spec complete?  Is there enough precision in the specification so
that users of a WebDAV  implementation provided by vendor A could switch to
vendor B?  I'm a bit concerned about the assumptions that the caller must make
about the collection.

2. Is the solution practical?  I'm concerned not about about the total number
of network round trips, but the quantity of data required for each.  If a
PROPFIND is required which returns back information about every element, then
the solution won't scale.  The inevitable solution will be that each vendor
will invent their own value-added interface which performs the necessary
filtering on the server.  This dillutes the value of a standard.

3. Is the spec consistent?  In this case, I don't see one, but those that
attended the last conference will remember my "garden path" dealing with the
implication of frozen safe references.
Received on Friday, 7 August 1998 06:22:49 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Tuesday, 2 June 2009 18:43:47 GMT