Re: How do you POST to a "document"?

Roy T. Fielding wrote,
> Miles apparently wants me to remove the definition of resource
> because it arbitrarily constrains other models.  I disagree.
> There are no models that I know of for which the definition of
> resource is not a superset of what they wish to identify.

Unsurprisingly, I disagree.

In fact I think that on your definition the set resources is not merely 
not a superset of the set of entities recognized in (some) other 
models, it's completely disjoint.

When I think of the web in terms of processes and communiction channels 
I'm quite definitely not thinking in terms of abstract entites with the 
characterisitics you claim for resources. Where you see states and 
names I see flows and processing.

I have no desire to exclude your model. On the contrary, I'm happy to 
agree that it can be a useful organizing principle and has been used 
successfully to build significant systems. But I think that much the 
same can be said for one based on processes and flows, and I would have 
thought that it would be hard for anyone who's spent much of their time 
working with network protocols and software to disagree.

But if neither, nor any of several other equally legitimate models, are 
to be excluded, then the Web Architecture must be neutral between them. 
And if those models share common terms, eg. URIs, but assign them 
different roles or meanings, then neutrality in the Web Architecture 
implies that the Web Architecture should stay silent about them too.

Cheers,


Miles

Received on Thursday, 23 January 2003 19:38:21 UTC