... I am going to make an argument otherwise. In section 5.2.1.1 of your dissertation, you open with: "The key abstraction of information in REST is a...
... The URI identifies the resource, not the information that is sent in response to a GET on that resource. As such, I don't understand what you mean by...
... I used the term "information" because that is the term you used in the quoted section of your dissertation. ... If HTTP Auth is used to make a...
... It isn't the *only* information in REST. The identified information is an abstraction. If I have a link to http://bigbank.com/services/myaccount then it...
... In this thread, we are interested in the granularity of the abstraction identified by the URL. Let's use a different example to make the point, and then...
... You are very close to the point. You just demonstrated an important use of giving user-specific resources their own URI. I am arguing in favour of this. I...
... For clarity, make this: I am arguing against URIs like <http://www.acmebank.com/accounts/payment/12/>, where this URI refers to the 12th payment made by...
... ah. thanks for that clarification. i think your point (that such URIs are RESTless) can be more clearly made by stating that the only way that the...
This discussion needs to be much clearer about what the resource in question is. AFAICS the example is an 'account' resource. The URI need not be the only...
... If it's a URI, then there's nothing at all RESTless about it. That URI is perfectly fine, even if what you GET from it differs depending upon who invoked...
... ..., or HTTP Auth, which is conceptually nothing more than a specially tagged cookie. Thank you for adding this clarification to the argument. Tyler...
... It is not the URI on its own that I am arguing against. It is the absence of the user-specific resource URI that I am arguing against. I am arguing against...
However, there are issues with the durability of identity. In the example of the court case, the URI that identifies "the bank account of some person...
Tyler, Mark, Roy, a. Cookie dependent URIs are inherently restless -or- b. (http-)authentication is inherently restless (401s considered RESTless) -or- c....
... Trying to decide how many resources there are behind a URL is kind of an angels-on-the-head-of-a-pin question. But it is demonstrably necessary to make...
... dependent on a cookie/auth. There is the 'Vary' response header. This is used to indicate what request headers were used to determine the representation...
... Unfortunately, REST only says "important" resources should have a URI. This leaves us with a subjective test. I take it from your response that you believe...
... Not really. I was arguing against a claim that simply is not true about REST. ... That is not a likely scenario. Applications working within an account...
Ok, is use of Vary RESTful and if so is it good REST style? "Vary" means that resources are addressable via a URI+mimeheaders and not just a URI. I had been...
I think we are talking about three different things: * Availability * Representation * Personalization I think Availability and Representation are RESTful...
... Unfortunately this is unavoidable. Similarly in an OO language you would say that important resources should be represented as objects (as opposed to being...
... You're presuming that each personalized representation is its own resource. That's the point that Walden is trying to get across. The Slashdot home page is...
Paul, I buy all that, and yet I still have an issue with Tyler's example of the bank account URI that identifies anyone's account at that bank, depending on...
... Yes, it seems clear to me that one needs to drag _digitally signed messages_ into courts. Let's say you have a very precise (specific) and URI for a single...
The relevant test here is, I believe, whether the state of the resource is relevant to the meaning of the message/data, or whether its identity is all that is...
