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] Question on Profile Usage Re: [wsbpel] Issue 82 - resolution]


Quoting Satish Thatte <satisht@microsoft.com>:

>This is an interesting discussion but I have lost track of
>>the thesis/antithesis dialectic here that we are debating
>to >achieve synthesis ;-)
You seem to disagree with the following parts of my "thought
experiment":

1) The merits of translating abstract BPEL into some high
   level language module that can be further be modified and
   compiled into something deployable.

2) That if we make certain assumptions about the abstract
BPEL, perhaps we can make the simplify the translation
process. The fundamental assumptions in the "abstract-BPEL-
as-template-method-implementing a protocol-model," one
should minimise the BPEL constructs to the bare essentials
needed to implement a protocol and produce the correct
public visible behaviour. I feel candidates  for
exclusion are constructs that are mainly internal (i.e.,
flow, calls to web services the abstract BPEL patron does
not control, compensate).

Your argument is this "lightweight BPEL" approach can
only handle "trivial" processes. If I understand you
correctly, trivial processes are so because they do
not use the features you associate with a long running
transaction, for example, compensation, or atomic
transactions.

I am speculating that this approach could yield "light
weight" BPEL executable than can solve *some* classes of
real problems for individuals that otherwise would not
use WS-BPEL (i.e., SMEs). Think quicker (and cleaner)
than a pure GPL solution (in say Perl with REST) but
dirtier than a full blown WS-BPEL solution. This is not
to say that this simple solution will supplant the need
for full-blown engines. Perhaps having these light weight
implementations may actually help grow the WS-BPEL
ecology by creating a critical mass of engines (and
exposure to WS-BPEL concepts) that make the demanding
sites more amendable to using heavy duty BPEL engines.


Cheers,
Andrew






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