Comments on: The XML Grinch http://jbrugge.com/blog/2007/05/24/the-xml-grinch/ this and that and some of those Wed, 7 Jan 2009 09:41:49 +0000 http://wordpress.org/?v=2.0.4 by: Mike O'K http://jbrugge.com/blog/2007/05/24/the-xml-grinch/#comment-12301 Sat, 20 Oct 2007 01:44:04 +0000 http://jbrugge.com/blog/2007/05/24/the-xml-grinch/#comment-12301 John, I stumbled across your blog, and Glean, while looking at various Style Sheets for Cobertura. I recently did a little review of some of the ideas behind ESBs, when considering the issues with (possibly) replacing a very key, but very ancient (20 year old) socket app, written in assembly. What technology to use for the next 20? I think I somewhat understand the benefits of an ESB: isn't that like my USB device on my computer? Or Microsoft's printer driver? A little perusing the internet blogs, starting from Vinoski's post, helped "distill" the ideas about ESBs into something I could understand even easier: Smart Plumbing. Not sure if Turner coined that term himself, but I think it really fits. After all, it seems we are moving away from centralized system, since the days of Multics. For other examples, note the success of peer to peer vs. Napster, or the decentralized repository scheme of BitKeeper vs. the centralized nature of CVS. ESBs: Wrong approach to adopting SOA by Bobby Woolf (Think about SOA first, then use ESB if necessary, or a simpler approach) http://www.ibm.com/developerworks/webservices/library/ws-soa-esbarch/ Steve Vinoski: ESBs force canonical protocols (exactly the problem we have of extra unnecessary mappings) "In practice, nobody ever needs to connect everything to everything else." http://blogs.iona.com/vinoski/archives/000071.html ESBs - EAI all over again (instead of open standards, like WS-*, we are tied to a proprietary software- webMethods). http://blogs.iona.com/vinoski/archives/000164.html Integration at the edges: http://www.ipbabble.com/2005/12/translation_does_not_mean_hub.html This is the best one, that best, in my opinion, distilled the concept of an ESB: The dark side to ESBs: "What value, therefore, is an ESB? To my mind, an ESB is smart-plumbing to which to attach dumb nodes." Interesting just for the fact that it's a Microsoft guy talking about vendor lock-in: http://blogs.msdn.com/richardt/archive/2005/03/23/401146.aspx#403310 http://blogs.msdn.com/richardt/archive/2005/04/28/413159.aspx As an aside, one certainly must wonder whether companies might reign in their leading minds - note the hilarious quote at the start of Bobby Woolf's article above: "Since its publication, Bobby’s article has drawn a lot of interest, and we appreciate the discussion surrounding this article. Unfortunately, it gave some readers the impression that IBM® no longer values the ESB. Rest assured that nothing is further from the truth. See the sidebar below by Greg Flurry and Kyle Brown for a clarification of the issues involved. " Vinoski, while at Iona, seemed to be in a similarly odd position. "But Steve, if ESBs are EAI all over again, why should I get Iona's offering?" His answer was not too convincing... John,

I stumbled across your blog, and Glean, while looking at various Style Sheets for Cobertura.

I recently did a little review of some of the ideas behind ESBs, when considering the issues with (possibly) replacing a very key, but very ancient (20 year old) socket app, written in assembly. What technology to use for the next 20?

I think I somewhat understand the benefits of an ESB: isn’t that like my USB device on my computer? Or Microsoft’s printer driver? A little perusing the internet blogs, starting from Vinoski’s post, helped “distill” the ideas about ESBs into something I could understand even easier: Smart Plumbing. Not sure if Turner coined that term himself, but I think it really fits.

After all, it seems we are moving away from centralized system, since the days of Multics. For other examples, note the success of peer to peer vs. Napster, or the decentralized repository scheme of BitKeeper vs. the centralized nature of CVS.

ESBs: Wrong approach to adopting SOA by Bobby Woolf (Think about SOA first, then use ESB if necessary, or a simpler approach)
http://www.ibm.com/developerworks/webservices/library/ws-soa-esbarch/

Steve Vinoski: ESBs force canonical protocols (exactly the problem we have of extra unnecessary mappings)
“In practice, nobody ever needs to connect everything to everything else.”
http://blogs.iona.com/vinoski/archives/000071.html

ESBs - EAI all over again (instead of open standards, like WS-*, we are tied to a proprietary software- webMethods).
http://blogs.iona.com/vinoski/archives/000164.html

Integration at the edges:
http://www.ipbabble.com/2005/12/translation_does_not_mean_hub.html

This is the best one, that best, in my opinion, distilled the concept of an ESB:
The dark side to ESBs: “What value, therefore, is an ESB? To my mind, an ESB is smart-plumbing to which to attach dumb nodes.” Interesting just for the fact that it’s a Microsoft guy talking about vendor lock-in:
http://blogs.msdn.com/richardt/archive/2005/03/23/401146.aspx#403310
http://blogs.msdn.com/richardt/archive/2005/04/28/413159.aspx

As an aside, one certainly must wonder whether companies might reign in their leading minds - note the hilarious quote at the start of Bobby Woolf’s article above:

“Since its publication, Bobby’s article has drawn a lot of interest, and we appreciate the discussion surrounding this article. Unfortunately, it gave some readers the impression that IBM® no longer values the ESB. Rest assured that nothing is further from the truth. See the sidebar below by Greg Flurry and Kyle Brown for a clarification of the issues involved.
”

Vinoski, while at Iona, seemed to be in a similarly odd position. “But Steve, if ESBs are EAI all over again, why should I get Iona’s offering?” His answer was not too convincing…

]]>