Re: opaque uris and self-describing resources

On Wed, 16 Jan 2008 19:38:46 +0100, Paul Prescod <> wrote:

> If there is an HTML page at these links then how do you distinguish
> between a GET intended to get those pages from a GET intended to read
> from STDIN?

In what wild circumstances would you usually do an HTTP GET against stdin,  
stdout or stderr in a *nix environment? Using the HTTP URI as an opaque  
identifier is different from using it as a dereferencing HTTP resource  
identifier. The two are orthogonal and don't conflict with each other.

The URI <> is used in XML processing  
software as an opaque identifier. Doing an HTTP GET against that URI  
dereferences an HTML document. That HTML document describes what the  
opaque identifier means. The HTML document itself is irrelevant to the  
identifier and the URI could return something completely different, a 404  
or even nothing at all (not dereferencable). The identifier still means  
the same and still has utility in XML processing software.

The self-documenting nature of using HTTP URIs as identifiers vanishes  
when you can't find documentation at the end of the URI, but it's still a  
valid identifier no matter what an HTTP GET against the URI returns (or  

Asbjørn Ulsberg          -=|=-
«He's a loathsome offensive brute, yet I can't look away»

Received on Thursday, 17 January 2008 23:49:39 UTC