W3C home > Mailing lists > Public > ietf-dav-versioning@w3.org > April to June 2001

RE: What is a supported property?

From: Clemm, Geoff <gclemm@rational.com>
Date: Fri, 22 Jun 2001 16:06:21 -0400
Message-ID: <3906C56A7BD1F54593344C05BD1374B1018E24CD@SUS-MA1IT01>
To: Ietf-Dav-Versioning <ietf-dav-versioning@w3.org>
   From: Mark Chu-Carroll [mailto:mcc@watson.ibm.com]

   On 22 Jun 2001 14:03:17 -0400, Clemm, Geoff wrote:
   > How about the following:
   > 
   > -------------
   > 
   > 3.1.3	DAV:supported-method-set (protected)
   > 
   > This property identifies the methods that are supported by the
   > resource.  A method is supported by a resource if an application of
   > that method to that resource has the semantics defined for that
   > method by the features supported by that resource.
   > 
   > 3.1.4	DAV:supported-live-property-set (protected)
   > 
   > This property identifies the live properties that are supported by
   > the resource.  A live property is supported by a resource if that
   > property has the semantics defined for that property by the
   > features supported by that resource.

   From following the whole discussion, I can see what you're trying
   to get at, but that wording is incredibly confusing. If I saw that
   in a specification with no further explanation, I'd have no idea of
   what it meant.

Well, that wouldn't be a good thing (:-).  All suggestions for
how to make this clearer/better are greatly appreciated!
However the "great resourcetype debate" falls out, minimally
we need to clearly define what the supported-*-set properties
mean.

   It's even misleading: "a live property is supported ... if that
   property has the semantics defined for that property ...". It reads
   like the property must exist for the resource - and yet, what we're
   talking about is the set of properties which *can*, but may not
   currently, exist for the resource.

I used the phrase "has the semantics defined for that property"
as an attempt to *not* imply that it must exist (but apparently
failed :-).

   I *think* that the meaning of a supported property is the following
   (which is very close to what you said, but unfortunately re-introduces
   that dangerous "type" word). 

           A live property is supported for a given resource if, according
           to the semantics of the resource type, that property may
           exist for that resource.

Unfortunately, this just defers the question to another undefined
phrase "according to the semantics of the resource type".  Adding
values to DAV:resourcetype wouldn't solve this problem, because the
protocol would then have to state when a resource has a particular
value in the DAV:resourcetype field.  We'd be tempted to say "when it
supports the semantics of all the methods and live properties defined
for that resource", but then we'd have closed the circle again (:-).

   Assuming that I'm right about what it means, that's not a difficult
   concept, but it's hard to find a precise way of wording it that 
   doesn't appear circular.

All suggestions are greatly appreciated!  One comment: the definition
probably has to include a reference to the set of features supported
by a resource.  For example, the activity feature adds the
DAV:activity-set to a version, so it is not the case that all versions
support the DAV:activity-set property, but rather that all versions
that support the activity feature support the DAV:activity-set
property.  Happily, the concept of "feature" *is* well defined, so
we're OK using that term.

Cheers,
Geoff
Received on Friday, 22 June 2001 16:00:40 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Tuesday, 8 January 2008 13:57:41 GMT