W3C home > Mailing lists > Public > public-ws-chor@w3.org > September 2005

Re: reasoning about interaction patterns

From: Roberto Lucchi <lucchi@cs.unibo.it>
Date: Tue, 13 Sep 2005 13:59:01 +0200 (CEST)
To: Steve Ross-Talbot <steve@pi4tech.com>
Cc: Claudio Guidi <cguidi@cs.unibo.it>, public-ws-chor@w3.org
Message-ID: <Pine.LNX.4.62.0509131346530.12913@sarastro.cs.unibo.it>

Dear Steve,

thanks for the reference, it will be a pleasure for me to know related 
work of Honda, Yoshida and Carbone.

Best regards,

On Tue, 13 Sep 2005, Steve Ross-Talbot wrote:

> Roberto,
> are you in touch with Honda, Yoshida and Carbone at all? They have been doing 
> work not unrelated but certainly a different facet to yours in looking at 
> liveness and looking at both a global and an end-point calculus which form 
> the basis of the formal semantics for CDL both now and in the future. You can 
> get their details from the WS-CDL membership list.
> I shall take a look at the paper.
> Cheers
> Steve T
> On 13 Sep 2005, at 11:31, Roberto Lucchi wrote:
>> Dear all,
>> I would like to point out that a paper which deals with choreography 
>> languages and in particular on the interaction patterns when alignment 
>> property is considered has been published in the proceeding of 2nd 
>> International Workshop on Web Services and Formal Methods WS-FM'05, LNCS 
>> 3670.
>> Here we report the abstract, we hope this can help and stimulate the 
>> discussion about the right interaction patterns that should be used at the 
>> choreography level.
>> Abstract:
>> Choreography languages provide a top-view design way for
>> describing complex systems composed of services distributed over the
>> network. The basic building block of such languages is the interaction
>> between two peers which are of two kinds: request and request-respond.
>> WS-CDL, which is the most representative choreography language, supports a 
>> pattern for programming the request interaction and two patterns
>> for the request-respond one. Furthermore, it allows to specify if an 
>> interaction is aligned or not whose meaning is related to the possibility 
>> to
>> control when the interaction completes. In this paper we reason about
>> interaction patterns by analyzing their adequacy when considering the
>> fact that they have to support the alignment property. We show the 
>> inadequacy of the two patterns supporting the request-respond interaction;
>> one of them because it does not permit to reason on alignment at the
>> right granularity level and the other one for some expressiveness lacks.
>> The paper is also available at the following URL:
>> http://www.cs.unibo.it/~lucchi/pubbl.html
>> Best regards,
>>  Roberto
Received on Tuesday, 13 September 2005 11:59:07 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 6 January 2015 21:01:08 UTC