James M Snell wrote:
Brian Smith wrote:
[snip]
My point is that your example is using atom:category/atom:id in a
totally different way than I was using it, and it seems really hard to
predict when that will happen.
[snip]
Correct; this is the crux of the problem. Sam and I went around on these
kinds of issues before with the Feed validator. While using atom:id or
atom:link in the ways discussed in this thread and not technically in
valid, they are questionable given the fact that there is no spec text
anyone can point to. Whose usage is correct? There's no way of knowing.
I don't agree there's a distinction.
In one case I embedded atom:id in a category and POSTed that to a server
to say categorise these things. In the other case Brian embedded a link
in category to say "The category may be uniquely identified by the IRI"
on GET. What would break?
Let me it another way - do we need to make categories resources and
provide them with atom:ids? If so, let's do that.
Bill
