Re: Naming ports vs. naming documents

Hi Jeni,

On 5/3/06, Jeni Tennison <jeni@jenitennison.com> wrote:
> The web architecture Rec says [1]:
>
>    "Although many URI schemes (§2.4) are named after protocols, this does
>     not imply that use of such a URI will necessarily result in access to
>     the resource via the named protocol. Even when an agent uses a URI to
>     retrieve a representation, that access might be through gateways,
>     proxies, caches, and name resolution services that are independent of
>     the protocol associated with the scheme name."
>
> I think it's absolutely fine to use an HTTP URI to refer to a document
> that is actually retrieved via a resource manager: the URI is just an
> identifier for the resource.

I agree: when getting the document for URI
http://www.google.com/xhtml, there may not be a server at Google that
receives an HTTP request. Instead the document might be returned by a
proxy, gateway, or cache. But in general those will return a document
that is equivalent to the document you would get if there was an
actual HTTP request going all the way to a Google server. I think we
can agree that in most cases it would not be a very good idea to use
the URI http://www.google.com/xhtml to point to a document completely
different that the usual Google search page. We can let authors choose
what URI to use, and consider that if they use URIs that make their
code confusing (like http://www.google.com/xhtml), this is their
problem. But there are maybe other alternatives, which I would like to
explore.

On 5/3/06, Norman Walsh <Norman.Walsh@sun.com> wrote:
> | 2) I am not convinced that reusing schemes is a good idea.
>
> I am certain that it is. As Jeni points out, WebArch discourages the
> invention of new schemes.

Norm,

Alright, then I will rephrase that: I do not feel comfortable allowing
authors to use URIs with schemes that come with an expected semantic
(like http://www.google.com/xhtml) to make references to documents
produced in the pipeline. And at the same time I agree: we should try
not to invent our own scheme.

Can we conciliate these two constraints? In a MIME encoded message, an
image can be embedded and referenced through a URI. But for those
images we don't use just any URI, but we use URIs with the cid: scheme
[1]. Would the cid: scheme make sense in our case? What other schemes
could we consider?

[1] http://www.rfc-editor.org/rfc/rfc2392.txt

Alex
--
Blog (XML, Web apps, Open Source):
http://www.orbeon.com/blog/

Received on Wednesday, 3 May 2006 20:47:43 UTC