W3C home > Mailing lists > Public > public-rdf-dawg@w3.org > October to December 2009

Re: ACTION-115: Note on proxy graph URI

From: Steve Harris <steve.harris@garlik.com>
Date: Sun, 11 Oct 2009 19:50:19 +0100
Cc: public-rdf-dawg@w3.org
Message-Id: <BD0B221F-36B4-4D2A-A0E2-6AD0A274B70B@garlik.com>
To: Kjetil Kjernsmo <kjetil@kjernsmo.net>
On 11 Oct 2009, at 17:10, Kjetil Kjernsmo wrote:
> Yup, our discussion may very well not be. I would really like to  
> have a
> straw poll on the topic before we proceed any further. But then, I
> understand that people want a FPWD ASAP...

Writing it into the FPWD doesn't commit us to including it in the PR.

> If we do have enough time for a straw poll, I suggest voting over the
> following questions, separately:
>
> 1) The HTTP Update protocol is required to be RESTful. (+1, yes; 0,
> preferred, but not required; -1, it could be anything)
>
> 2) The protocol must specify how to use POST, PUT, GET and DELETE on  
> a URI
> identifying information resource graph. (+1, yes, required; 0,  
> optional;
> -1, WG shouldn't spend time on it)
>
> 3) The protocol must specify how to use manipulate a graph with a  
> proxy
> graph identifier. (+1, yes, required; 0, optional; -1, WG shouldn't  
> spend
> time on it)
>
> They can be voted on separately, if it is voted in 1) that it must be
> RESTful and we can't find a way to do it in 3), we must drop it,  
> after a
> thorough investigation.

If there's a strawpoll that comes out in favour of 1) and 3), and it  
turns out their incompatible, then we'd have to make a decision down  
the line as to which principle was more important.

I have a strong concern around deciding that "it must be RESTful", as  
no-one in this group really seems to know the exact criteria for some  
API to be RESTful. Particularly there seems to be no hard cutoff  
between RESTful and non-RESTful, which means the "RESTfulness" of any  
design is likely to be highly subjective.

I think that any strawpoll should be "advice to the editor", to  
influence the design, rather that trying to rigidly determine if  
something is or is not RESTful.

- Steve

-- 
Steve Harris
Garlik Limited, 2 Sheen Road, Richmond, TW9 1AE, UK
+44(0)20 8973 2465  http://www.garlik.com/
Registered in England and Wales 535 7233 VAT # 849 0517 11
Registered office: Thames House, Portsmouth Road, Esher, Surrey, KT10  
9AD
Received on Sunday, 11 October 2009 18:50:49 GMT

This archive was generated by hypermail 2.3.1 : Tuesday, 26 March 2013 16:15:40 GMT