Re: Prelim. DAV spec.

In message <96Oct31.105634pdt."415911"@mule.parc.xerox.com>, Larry Masinter wri
tes:
># road (b), but I want to argue that choice (a) is the choice you should
>Yaron responded:
>> I say we take the weasel way out and support both. The draft already allows 
>for this.
>
>But actually, I meant to make a stronger case that you should NOT try
>to support choice (b): that is, don't try to allow for 'versioning' to
>apply to resources that don't have URLs.


Let me make an even stronger case, based not on anectodal evicence,
encouragement or rhetoric, but plain logic:

A representation (or entity -- same thing) is immutable, the way
integers and URLs are immutable. How many versions of the number
2 or the string "http://www.w3.org/" are there?

A representation is a sequence of bytes (the body) and a set of
name/value string pairs in the header; all together, a sequence of
bytes. Look at it as a big base-256 integer. It's still an integer,
just like 2. No state. No changes. No versions.

A resource, on the other hand, has state, attributes, related
resources (i.e. linked resources, or variants, or versions, ...)
etc. It's not directly observable: the only thing the world can know
for sure about a resource is (1) it's URL, and (2) how it responds to
requests like GET, and from that (3) which other resources are linked
to/from it. From the outside, files, database entries, program
computation results, etc. are indistinguishable.

It makes perfect sense for an origin server to say (or for anybody
to claim, for that matter) that:

	http://www.w3.org/file1;version-1.2

is the URL for a resource that represents version 1.2 of the
resource at:

	http://www.w3.org/file1

But it makes no sense to say that the representation:

	Content-Type: text/plan

	abc

is version 1.2 of the representation:

	Content-Type: text/plan

	def

any more than it makes sense to say that the number 4 is version 1.2
of the number 2.

Dan

Received on Thursday, 31 October 1996 14:26:44 UTC