I'm in. Can I suggest as its an SOA group that we have an SOA stack
rather than just REST/SOAP so we look at the different layers and then
we can map REST/SOAP onto a single model. I'd also suggest using the
SOA RM as the basis for that stack.
Steve
2008/8/6 Mark Baker <distobj@...>:
> On 8/5/08, Gervas Douglas <gervas.douglas@...> wrote:
>> You do not have to read the REST/WS*-SOAP debates on this Group in
>> great detail to get the persistent impression that there is a
>> significant mismatch of comprehension between the two camps. My idea
>> of a stack comparison was that it would aid in seeing the difference
>> of capabilities/roles/approaches of the two technologies and thus help
>> to reduce misunderstandings. Any volunteers from the ranks of
>> technical genius in this Group??
>
> I think we've tried this before without success, but if you'd like to
> serve as moderator to keep the discussion on track and technical, I'm
> game to try again.
>
> Mark.
>
I'm in. Can I suggest as its an SOA group that we have an SOA stack rather than just REST/SOAP so we look at the different layers and then we can map...
Steven, If with a combination of your experience and brains and those of the RESTafarians with whom you are wont to joust, we could produce a definitive SOA...
... +1 ... I don't think the SOA RM has much to offer a discussion about stacks. I'm happy to use OSI in full or in IETF-abbreviated form, or any other stack...
... The RM has a clear split between the technology piece (Execution context) and the Service and Capability piece, so one key question is how clearly these...
... I don't even know what that means, but it doesn't sound like it has much to with issues of stack design, so I really doubt it'll be very helpful as a tool....
Let me explain.... If we assume that a stack is (strawman) Business Context Business Service Logical Model Policy Software Service/Client ... context below...
... Steve, that's largely voodoo to me. It appears we don't even agree what a stack is. To me, a stack describes a pipeline for message processing, where an ...
... The problem is that this, IT convention, limits the discussion to purely the technical part of the discussion and the runtime rather than the service...
... Ok, I actually agree with this, but I'm buggered to come up with a way for us to properly agree on a stack, no matter the size of it. Should we just start...
... I have to say that I love your command of colloquial English, Alex - I just hope that our more religious US colleagues persist in their incomprehension...
OK, I'll bite. I can't edit the wiki (I don't have an invite key) but you can see my first attempt [1]. I did a quick search for "SOA Stack" came up with [2]...
You can find Alan's initial contribution (thanks, Alan) at http://service-orientated-architecture.pbwiki.com/SOA+Stack. If you want access to add to this...
Gervas, count me in, too :-) IMHO the critical issue is to come to agreement about the goals one tries to achieve regarding the design and realisation of some...
Steve, ... You're the one saying it has no meaning. What you mean is, perhaps, that you don't see any meaning. ... That's nonsense. Of course technical...
... Morning. ... Is it? I've never thought of Foreign keys as being a good start for communication. ... Not really, sometimes it is actually important to list...
... Sorry, my bad, so let me rephrase; But as a kick-ass semantics-shifting cache-enabled resource-oriented mechanism for communication, that's a pretty good...
... Or to put it another way.... plumbing. Yes letters are a good mechanism for communication but only when syntax and semantics are applied. ... But that is...
Time for me throw my tuppence into the thread again. Let's clarify what constitutes the uniform interface in REST. It is: 1. The request method 2. The content...
... How are the bits below self-descriptive? Take the order (of type transaction+xml) how do I know when I write the code that this is the document that I...
Steve, The application protocol specifies these things. The application protocol is more than a message schema, it will also describe behaviours including but...
... Yup. ... AtomPub gives a basis for data but I'm not sure where it does the application semantics. If I want to say that a given link results in a set of...
... I did try to make clear that I feel that trying to fit everything into AtomPub is a fools errand. From my POV, e-commerce requires a new MIME type and app...
... My apologies for missing it. ... I have one its based on a two instructions DO and DO NOT both are idempotent. The only slight issue is that I've assumed a...
Having problems with message search?
Fill out this form to ensure your group is one of the first to be migrated to the new message search system.
Offline 
Send Email
