OASIS Mailing List ArchivesView the OASIS mailing list archive below
or browse/search using MarkMail.

 


Help: OASIS Mailing Lists Help | MarkMail Help

wsbpel message

[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [List Home]


Subject: Re: [wsbpel] implicite links of the runtime engine (was: Implicit<sequence> macro)


David RR Webber - XML ebusiness wrote:

>Message text written by Assaf Arkin
>  
>
>So as a BPM vendor who has to support a million and one different 
>protocols and means of communication, WSDL is definitely my preferred 
>choice.
><<<
>
>Assaf,
>
>Right - but it should not be your ONLY choice.
>
>Big difference.   It's OK to borrow a model as a means
>to express your neutral interfacing - but direct linkage
>and dependence on physical features is not good
>engineering practice.   That guarantees are very
>short life-expectancy for implementations.
>  
>
>Of course if your company wants to make it its only 
>choice that's fine - but as an OASIS standard I think
>it needs broader appeal, eh?
>
The question here is model vs interoperability.

If you are looking at a model -- and my implementation is based on a 
model -- then you are perfectly right. We can sit down separately and 
define a model that is very generic and not directly dependent on any 
XML or Web services technologies, maybe just elect to use some form 
process calculus.

Since my implementation is based on a model it can support several of 
the languages out there, not all of which are concerned with WS or even 
XML (e.g. modeling and notation). Models are good and we can decide to 
sit down and figure out the correct model to use. But models fall short 
in one respect -- they don't provide you with sufficient constraints to 
get any level of interoperability. We might as well exchange UML 
diagrams -- our tools would represent the same process definition but 
there's no way we could get our products to talk to each other.

So my interest in this working group is strictly in providing a language 
for the purpose of interoperability, and my implementation would amongst 
other things also support this language when it facilitates 
interoperability. And that means that models aside, we need to put some 
stake in the ground and decide to use some core set of technologies.

Deciding to use WSDL is not a good constraint for a model, but it's a 
perfect constraint for a language that deals with interoperability of 
Web services.

arkin

>
>Thanks, DW.
>  
>




[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [List Home]