I agree with Alan.

IMHO:
REST abstracts certain features from HTTP, in order to have a guideline to implement better Web applications. It is strongly bound to HTTP (so I do not consider as a problem to have HTTP examples in our doc) but it is not tightly bound to that transport. Section 4 of the 2002 paper by Fielding and Taylor describes the REST architectural elements without reference to HTML. In sect. 5 they use HTTP, together with other protocols, to describe an example, and they explicitely note that "Although the Web’s primary transfer protocol is HTTP, the architecture includes seamless access to resources that originate on many pre-existing network servers..." (in sect. 5.2). And section 7.2 is entitled "REST applied to HTTP".

However the point raised by Alexander is relevant, and should be clarified in the paper.

2015-03-24 19:47 GMT+01:00 Sill, Alan <alan.sill@ttu.edu>:
Um, well, not really. REST is just REST; its a design pattern that can be implemented using any protocol. It's true that REST-over-HTTP is the most popular form, but there is nothing about stating that a design is RESTful that implies the use of HTTP.

Alan



On Mar 24, 2015, at 2:36 PM, Papaspyrou, Alexander <Alexander.Papaspyrou@adesso.de> wrote:

Well, it depends on how picky you are. Albeit, this can be easily fixed by changing REST to CRUD in the core specs – that’s pretty much what we meant when saying „RESTful protocol“.

@Thijs: Correct me if I am wrong (you will, I bet ^^)

-A.

From: <Sill>, Alan Sill
Date: Dienstag, 24. März 2015 14:15
To: "Stanik, Alexander"
Cc: "occi-wg@ogf.org"
Subject: Re: [occi-wg] Documents for OCCI rev 1.2

Hi,

Did I miss something in what you are flagging? This is not a contradiction. REST design is not dependent on being implemented in HTTP -- REST can be achieved with any transport.

Alan

On Mar 24, 2015, at 8:04 AM, Stanik, Alexander <alexander.stanik@tu-berlin.de> wrote:

- In the first sentence OCCI is defined as RESTful Protocol, but the text behind the second bullet says that the Core Model is HTTP independent. I think it would be very beneficial to keep the core model HTTP independent, which is also the advantage of having the Redering specifications in separate documents.

-------------------------------------------------------
      >>> business. people. technology. <<<
-------------------------------------------------------

adesso AG mit Sitz in Dortmund
Vorstand: Christoph Junge, Michael Kenfenheuer (Co-Vors.), Dr. Rüdiger Striemer (Co-Vors.)
Vorsitzender des Aufsichtsrates: Prof. Dr. Volker Gruhn
Amtsgericht Dortmund HRB 20663

 


_______________________________________________
occi-wg mailing list
occi-wg@ogf.org
https://www.ogf.org/mailman/listinfo/occi-wg




--
Augusto Ciuffoletti
Dipartimento di Informatica
Università di Pisa
56100 - Pisa (Italy)