W3C home > Mailing lists > Public > www-webdav-dasl@w3.org > January to March 1998

RE: DASL Charter

From: Jim Whitehead <ejw@ics.uci.edu>
Date: Tue, 3 Mar 1998 14:43:54 -0800
Message-ID: <01BD46B2.CB1F6FC0.ejw@ics.uci.edu>
To: "'Surendra Reddy'" <skreddy@us.oracle.com>, Alex Hopmann <hopmann@holonet.net>, "www-webdav-dasl@w3.org" <www-webdav-dasl@w3.org>
Unfortunately, I think the text "other interoperable HTTP applications" is 
*way* too broad.  This could be construed to mean, among other things:
  - Internet Printing Protocol (IPP) properties
  - any RPC-like protocol bolted-on to (passed through) POST

I had an "aha" moment when Alex wrote:

> Also, those two queries look identical. Please keep in mind that DAV
> has a totally open ended property framework, so <d:dma-guid> is a
> property just like any other.

The emerging consensus on this list appears to be that DASL should support 
searching on repositories whose data model is equivalent (or can be made to 
look equivalent) to the DAV resource data model (i.e., a body plus 
properties which are name, value pairs).  So, my stab at rewriting the 
Charter is to modify the first sentence, which now says:

> The goal of this working group is to provide searching
> interoperability among the variety of underlying storage
> systems whose data can be exposed by the WEBDAV extensions to
> HTTP.

Change this to read:

The goal of this working group is to provide searching
interoperability among the variety of underlying storage
systems whose data can be exposed (viewed? are compatible with?)
through the HTTP plus WebDAV data model.

So, if, as you and Jim Davis have suggested, the properties of a repository 
can be expressed in XML as name, value pairs where the name is a URI (which 
can be an encoding of the property namespace of an underlying repository), 
and the value can be encoded as well-formed XML, then the properties are 
compatible with the HTTP + WebDAV  resource data model, and are amenable to 
searching via DASL.

One concern I have with cross-repository searching is the problem of 
typing.  As near as I can tell, WebDAV is the only system of those 
mentioned in this thread which uses XML for its typing.  Others, like DMA, 
have a type system for properties, and it seems to me this type system 
might creep into the search syntax if we're not careful.  On the other 
hand, by using XML to encode search syntax and search results, perhaps we 
can avoid this entire issue.

- Jim

On Tuesday, March 03, 1998 10:07 AM, Surendra Reddy 
[SMTP:skreddy@us.oracle.com] wrote:
>      Alex,
>
>         |:The goal of this working group is to provide searching
>         |:interoperability among the variety of underlying storage
>         |:systems whose data can be exposed by the WEBDAV extensions to
>         |:HTTP. The working group will define HTTP extensions that enable
>         |:server-executed queries to locate resources based upon their
>         |:property values and text content.
>
>     I would recommend the following changes to DASL charter.
>
>     "The goal of this working group is to define and develop an
>     extensible  DAV Searching and Locating protocol as
>     extensions to HTTP.The working group will define HTTP
>     extensions that enable  server-executed queries to locate resources
>     based upon their property values and text content as exposed by
>     DAV as well as other interoperable HTTP applications."
>
>      With this change incorporated, I am full agreement with DASL 
charter.
>
>     Appreciate your comments.
>
>  Best regards,
>
> Surendra
Received on Tuesday, 3 March 1998 18:19:13 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Sunday, 22 March 2009 03:38:03 GMT