Comments on: What is SOA? http://72.249.21.88/nonintersecting/2007/10/05/what-is-soa/ Life and Technology (non-intersecting) Fri, 05 Dec 2008 02:56:16 +0000 http://wordpress.org/?v=2.6.3 By: mpu http://72.249.21.88/nonintersecting/2007/10/05/what-is-soa/#comment-11119 mpu Sat, 29 Nov 2008 17:24:22 +0000 http://wanderingbarque.com/nonintersecting/2007/10/05/what-is-soa/#comment-11119 End of statement? End of statement?

]]> By: Huh? ‘SOA’ voted most ‘confusing acronym of the year’ « The Whole Enchilada http://72.249.21.88/nonintersecting/2007/10/05/what-is-soa/#comment-10875 Huh? ‘SOA’ voted most ‘confusing acronym of the year’ « The Whole Enchilada Sat, 10 Nov 2007 03:16:08 +0000 http://wanderingbarque.com/nonintersecting/2007/10/05/what-is-soa/#comment-10875 [...] Pete Lacey’s Weblog :: What is SOA? [...] [...] Pete Lacey’s Weblog :: What is SOA? [...]

]]> By: Bob McCormick http://72.249.21.88/nonintersecting/2007/10/05/what-is-soa/#comment-10840 Bob McCormick Tue, 23 Oct 2007 22:56:19 +0000 http://wanderingbarque.com/nonintersecting/2007/10/05/what-is-soa/#comment-10840 If SOA is really an architectural style, then we really don't need a new term (like ROA) to describe replacing the WS-* bloat with REST. And IMHO REST is perfectly capable of implementing "services" for any sane definition of services. Rather than have to fight the whole mindshare fight of convincing people than SOA is yesterdays news and ROA is the new "right" way to do things, why not just embrace and extend? REST is just SOA done right. It's SOA improved. SOA 2.0. Instead of calling an obscure OpenNewAccount service, we "create" a NewAccount. What could be more natural? What could be more "Business Focused"? WS-* and the obtuse bloat than the vendors and committees have turned it into is the problem, not SOA. If SOA is really an architectural style, then we really don’t need a new term (like ROA) to describe replacing the WS-* bloat with REST. And IMHO REST is perfectly capable of implementing “services” for any sane definition of services.

Rather than have to fight the whole mindshare fight of convincing people than SOA is yesterdays news and ROA is the new “right” way to do things, why not just embrace and extend? REST is just SOA done right. It’s SOA improved. SOA 2.0. Instead of calling an obscure OpenNewAccount service, we “create” a NewAccount. What could be more natural? What could be more “Business Focused”?

WS-* and the obtuse bloat than the vendors and committees have turned it into is the problem, not SOA.

]]> By: Not a TLA more, Not a TLA less. « Being just me http://72.249.21.88/nonintersecting/2007/10/05/what-is-soa/#comment-10830 Not a TLA more, Not a TLA less. « Being just me Sat, 20 Oct 2007 09:52:11 +0000 http://wanderingbarque.com/nonintersecting/2007/10/05/what-is-soa/#comment-10830 [...] Not a TLA more, Not a TLA less. Filed under: SOA — Sankar Khrishnamurthy @ 9:51 am Pete Lacey has written an entry regarding What is SOA? And the blogosphere has responded as well,  Sam Ruby, Stefan Tilkov, to just mention two.  Here are my 2 cents. [...] [...] Not a TLA more, Not a TLA less. Filed under: SOA — Sankar Khrishnamurthy @ 9:51 am Pete Lacey has written an entry regarding What is SOA? And the blogosphere has responded as well,  Sam Ruby, Stefan Tilkov, to just mention two.  Here are my 2 cents. [...]

]]> By: Ian Bruce http://72.249.21.88/nonintersecting/2007/10/05/what-is-soa/#comment-10827 Ian Bruce Wed, 17 Oct 2007 20:42:34 +0000 http://wanderingbarque.com/nonintersecting/2007/10/05/what-is-soa/#comment-10827 Eloquent and fun, a great post Pete. As one of the marketing types that may or may not be responsible for some of the confusion and hype, I like Bosworth's pithy definition of web services, which might also apply to the ethos behind the whole SOA idea. Eloquent and fun, a great post Pete. As one of the marketing types that may or may not be responsible for some of the confusion and hype, I like Bosworth’s pithy definition of web services, which might also apply to the ethos behind the whole SOA idea.

]]> By: Pete http://72.249.21.88/nonintersecting/2007/10/05/what-is-soa/#comment-10826 Pete Wed, 17 Oct 2007 13:45:16 +0000 http://wanderingbarque.com/nonintersecting/2007/10/05/what-is-soa/#comment-10826 <strong>Chris:</strong> When I wrote "stateful resources" I was referring to resource state, not session state. Confusion abounds because the word state is overloaded. See this later <a href="http://wanderingbarque.com/nonintersecting/2007/10/08/towards-a-better-network-programming-taxonomy/#comment-10809">comment</a>. Chris: When I wrote “stateful resources” I was referring to resource state, not session state. Confusion abounds because the word state is overloaded. See this later comment.

]]> By: Chris Marino http://72.249.21.88/nonintersecting/2007/10/05/what-is-soa/#comment-10824 Chris Marino Wed, 17 Oct 2007 03:12:36 +0000 http://wanderingbarque.com/nonintersecting/2007/10/05/what-is-soa/#comment-10824 Pete, came back to this post tonight and saw for the first time that you say: "Resource Oriented Architecture (ROA): A technical approach to NOC that has an addressable, stateful resource as its principle abstraction. Today, REST/HTTP is the chief implementation approach" Don't want to fan the flames more than I have to, but the definition of REST requires that interactions be *stateless*. See Fieldings paper http://www.ics.uci.edu/~fielding/pubs/dissertation/rest_arch_style.htm Was this a typo, or are you saying that statefulness is essential to NOC? I am not trying to be pedantic, nor am I dogmatic about this. I simple want to understand your position. IMHO, stateful resources aren't really all that different from RPCs. You might as well put the SOAP messages in the URI or HTTP header (something I've seen being used). It might be NOC, but I wouldn't call it REST Pete, came back to this post tonight and saw for the first time that you say:

“Resource Oriented Architecture (ROA): A technical approach to NOC that has an addressable, stateful resource as its principle abstraction. Today, REST/HTTP is the chief implementation approach”

Don’t want to fan the flames more than I have to, but the definition of REST requires that interactions be *stateless*. See Fieldings paper

http://www.ics.uci.edu/~fielding/pubs/dissertation/rest_arch_style.htm

Was this a typo, or are you saying that statefulness is essential to NOC?

I am not trying to be pedantic, nor am I dogmatic about this. I simple want to understand your position.

IMHO, stateful resources aren’t really all that different from RPCs. You might as well put the SOAP messages in the URI or HTTP header (something I’ve seen being used).

It might be NOC, but I wouldn’t call it REST

]]> By: SnapLogic Blog » ESBs and the Last Mile of Integration http://72.249.21.88/nonintersecting/2007/10/05/what-is-soa/#comment-10814 SnapLogic Blog » ESBs and the Last Mile of Integration Thu, 11 Oct 2007 16:57:57 +0000 http://wanderingbarque.com/nonintersecting/2007/10/05/what-is-soa/#comment-10814 [...] To me, the real question is how do you get the benefits of flexibility and dynamic languages, while avoiding ad-hoc interfaces? I think the solution is a services architecture - the problem is that we still have debate over what web services and a services architecture really is. I like Adam Bosworth’s definition of Web services from a 2005 interview in Queue: KIRK MCKUSICK (KM) People sure talk a lot about Web Services, but it’s not clear they’re all talking about the same thing. How would you define “Web Services”? ADAM BOSWORTH (AB) The term Web Services refers to an architecture that allows applications to talk to each other. Period. End of statement. [...] [...] To me, the real question is how do you get the benefits of flexibility and dynamic languages, while avoiding ad-hoc interfaces? I think the solution is a services architecture - the problem is that we still have debate over what web services and a services architecture really is. I like Adam Bosworth’s definition of Web services from a 2005 interview in Queue: KIRK MCKUSICK (KM) People sure talk a lot about Web Services, but it’s not clear they’re all talking about the same thing. How would you define “Web Services”? ADAM BOSWORTH (AB) The term Web Services refers to an architecture that allows applications to talk to each other. Period. End of statement. [...]

]]> By: Labnotes » Rounded Corners - 160 (Resourced) http://72.249.21.88/nonintersecting/2007/10/05/what-is-soa/#comment-10805 Labnotes » Rounded Corners - 160 (Resourced) Tue, 09 Oct 2007 16:02:32 +0000 http://wanderingbarque.com/nonintersecting/2007/10/05/what-is-soa/#comment-10805 [...] Resourced. This whole ROA vs SOA debate, where are we going with that? [...] [...] Resourced. This whole ROA vs SOA debate, where are we going with that? [...]

]]> By: Martin Eggenberger http://72.249.21.88/nonintersecting/2007/10/05/what-is-soa/#comment-10799 Martin Eggenberger Tue, 09 Oct 2007 04:33:13 +0000 http://wanderingbarque.com/nonintersecting/2007/10/05/what-is-soa/#comment-10799 Pete Lacey I totally agree with the naming confusion and I would put it as follows: What comes to mind very quickly is that Architecture as in SOA doesn't make a running system but describes a system that uses Service Oriented Computing principles. I guess some Marketing folks didn't understand the difference between Architecture, Development, Operations and Business. But why is it advantageous to use the SOA versus the SOC acronym. I guess, some big consulting companies do like to sell only the architecture piece without the actual development and operational support necessary to make an architecture work. This is brilliant, as most executive management doesn't understand the difference either. Let's sell an architecture (e.g. blue print) and have them figure out how to build it and in the meantime we can sell them services for implementing that blue print. Most people are probably not offended by this issue but as a software professional I demand a name change. And the winner is anything but SOA. I vote for SOC as in service oriented computing which encompasses everything from business services, service compositions, service runtime and governance; to cross cutting concerns. Pete Lacey
I totally agree with the naming confusion and I would put it as follows:

What comes to mind very quickly is that Architecture as in SOA doesn’t make a running system but describes a system that uses Service Oriented Computing principles. I guess some Marketing folks didn’t understand the difference between Architecture, Development, Operations and Business. But why is it advantageous to use the SOA versus the SOC acronym. I guess, some big consulting companies do like to sell only the architecture piece without the actual development and operational support necessary to make an architecture work.

This is brilliant, as most executive management doesn’t understand the difference either. Let’s sell an architecture (e.g. blue print) and have them figure out how to build it and in the meantime we can sell them services for implementing that blue print. Most people are probably
not offended by this issue but as a software professional I demand a name change.

And the winner is anything but SOA.
I vote for SOC as in service oriented computing which encompasses everything from business services, service compositions, service runtime and governance; to cross cutting concerns.

]]>