Hi,

REST is too important to leave to chance, especially if we want to see an
eco-system of open source and other reference implementations and POCs. I
suggest a layered approach. Define the abstraction in the main spec. The REST
implementation/binding could be defined in an appendix (slight preference) or
separate document. I suspect there won’t be a rush to define WS bindings. :-)

Just one man’s opinion.

Thanks,

Paul

Paul Lipton

VP, Industry Standards and Open Source

Member, CA Council for Technical Excellence

Phone (preferred number!): +1 215 867-9231

Mobile: +1 267 987-6887

Email: paul...@ca.com <mailto:paul...@ca.com>

THIS MESSAGE MAY CONTAIN CONFIDENTIAL, PRIVILEGED OR OTHER LEGALLY PROTECTED
INFORMATION. IT IS INTENDED FOR THE ADDRESSEE ONLY. IF YOU ARE NOT THE ADDRESSEE
(OR SOMEONE THE ADDRESSEE AUTHORIZED TO RECEIVE THIS MESSAGE), YOU ARE
PROHIBITED FROM COPYING, DISTRIBUTING OR OTHERWISE USING IT. PLEASE NOTIFY THE
SENDER AND RETURN IT AT OUR COST. THANK YOU.

From: Stav...@uk.fujitsu.com [mailto:Stav...@uk.fujitsu.com] Sent: Thursday, June 17, 2010 9:40 AM To: sa...@lists.oasis-open.org Subject: [saf] Some work on the interfaces

Hi all,

Attached a very preliminary draft of a REST-like interface. Some questions to
the TC:

- Should we mandate a specific interface approach (e.g. WS, REST) or
shall we just abstractly define necessary operations and then suggest how these
can be implemented in REST or WS in appendices?

- Should this interface text be on the main spec or in a separate
document?

I haven't added this to the document folder yet, waiting for discussion first.

With something like this in place, we can build a straightforward REST-like
interface for quick POCs and demos on top of our SAF implementation (which is
currently WS-based and a little bit hard to use :-))

Cheers,

Stavros

--

Fujitsu Laboratories of Europe

Fujitsu Laboratories of Europe Limited Hayes Park Central, Hayes End Road, Hayes, Middlesex, UB4 8FE Registered No. 4153469