Talk:HTTPURIUseCases

From W3C Wiki
Jump to: navigation, search

In each use case there is a "sender" writing something (usually RDF, for inclusion in email, a triple store, an HTML or RDF document, etc.) that they want to be understood by a "receiver".

I think HTML and RDF are different enough that these should be separate use cases. The use cases shouldn't assume those contexts are the same. In fact, in HTML, a@href, img@src, @xml:xmlns, scripts, fonts, objects, etc. are all different contexts. You don't need a zillion use cases, just make the ones you're talking about more concrete.

The answers may very well be different.

-- JAR: Thanks for this. I think the answers are different and I have suggested talking about RDF first, then seeing if there is any way to generalize (which may or may not be appropriate).

In use case "A", it asks the question "Which is true". But neither are "true". I think the use cases needs to be expanded to explain the consequence of assuming that both statements reflect the attempt of the sender to communicate the assertion to the receiver, in order to work out the actual harm in not allowing both to be asserted.

Please elaborate the use case to 'Sender wishes to tell receiver, in RDF, the weight of the whale in Moby Dick' (or some such)

-- JAR: I have attempted to do this, please see whether what's written there now works.

you _can_ write anything you want

In Example 1, there is a question: Before you can write this do you have to do a fetch and look at the response? (Or, what parts of the response do you have to look at, and how hard?)

I think it would be useful to elaborate the use cases and then use them more carefully in the discussion, since you seem to be discussing alternatives in the use case page.

Question: Before you can write this do you have to do a fetch and look at the response? (Or, what parts of the response do you have to look at, and how hard?)

Answers:

  • A sender can write anything the sender wants, and doesn't "have to" do anything.
  • The response can vary anyway

please tell the use case in more detail. A sender writes something and the receiver wants to read and interpret what the sender writes. Who "has to" do what?

-- JAR: Good point. The goal is for the sender to be understood by the receiver. Given this it is important what they write. What they "have to do" to achieve correct communication depends on their prior agreement on how to communicate (the proposal details).