IBM's Mills: 'Find me a company not interested in SOA principles'

TalkBack 1 of 6:

Next »

• Thread View
• Flat View

SOA is DOA

Yes, check back a few years and you will see that it WAS me that coined that phrase. I should have said that SOA is RPC.

[I challenge anybody in the room to bring forward a company anywhere in the world who?d say that they?re not trying to make their IT operation more efficient, more effective, and that?s inclusive of their application assets]

A loaded statement meant to quash dissension before it starts.

[Companies all over the world say they want to optimize their business processes. They want to have the appropriate number of application assets to support their business processes. They don?t want redundancy. They don?t want 30 ledgers. They don?t want dozens of supply chain systems. They don?t want five CRM systems. They want one set of application assets that effectively and efficiently serve their business needs and requirements. And they want to use those assets to the fullest.]

Of course they do! Now for the sales pitch.

[Thats what service orientation is about. I can?t name a single company in the world that does not have interest in this. And I can?t name a company that would not say that they?re not on some kind of journey to improve the service orientation, the structure of their application environment for better economics and better efficiency.]

Blah, blah, blah. An IBM SOA sales droid that can't name a single company . . . The media frenzy in SOA is unavoidable - and top management types (that are NOT technical) are always looking for shiny and new.

[Because the alternative is that all your applications are disconnected from each other. Your processes don?t work effectively. Your business ratios are the worst in your market segment, and your company is on the verge of bankruptcy.

So there?s no other choice. There?s no alternative to this idea. It doesn?t matter if your buzzword at the end of the day is service oriented architecture, or business process integration, or enterprise application integration? It all leads to the same place.]

B.S. SOA is not the only way to connect applications. Even if it's scripts and ftp - one application can connect and feed into another. If this wasn't in place - no one would be in business. Saying that SOA makes this EASIER is much more believable than saying you can't do it without SOA.

[Nobody gets up in the morning with the idea that they want to accumulate more redundant IT assets. What they want to do is they want to run their business more effectively. CEOs don?t get up in the morning thinking about acquiring more IT for the sake of IT. they don?t get up in the morning thinking about IT architecture, they get in the morning thinking about, how do they run their business more efficiently and effectively. They turn to the experts in their business, they guys that know how to do that. Notions of sharing are obvious in very company are obvious in every company. Governance, shared service, shared capabilities, shared resources.]

Once again, loading the statement with "NOBODY" does this - keeping anyone from speaking up against this. CEOs have nothing to do with accumulating IT. I have seen many a manager accumulate IT - to make himself more important in the company. As for turning to the experts, he means high-priced consultants - which OF COURSE talk up SOA. If you asked some of the experts in the trenches, you may get much better advice (nah! They aren't the experts, since they have been doing this for the same company for many years . . .).

[This is a dominant conversation globally today. Because of the fundamental lack of available capital dollars to run a company. So everybody is on a path towards greater efficiency. Everybody is on a path to share things of all kinds. Everybody is on a path to share services, which is the very essence of what service oriented architecture is about. There?s no alternative. [The 'SOA is dead' debate] is a useless debate.]

This is HIS dominant conversation today. Then he tries to frame this in the "SAVE MONEY" category - except that the short term cost of ripping out everything and re-architecting and re-deploying new technology has a phenomenal pricetag. Then he goes into the "everyone is doing it" schpeel to once again quiet dissent.

Very opinionated. Very hard sell. He might give Perry Mason a run for his money.

In the end, everything hinges on the reuse aspect. Reuse has been attempted in just about every IT endeavor for the past 30 years. Why haven't we gotten it right? The biggest issue is one of "where do I find this?" when trying to create applications. A true SOA would have hundreds or thousands (or more) "services" that would need to be "mashed up" for that next killer app. What if the developer can't find that one service he needs - and (unknowingly) duplicates it. What if a developer thinks he can do better and knowingly duplicates. What if a service provides MOST of what the developer wants but not all? What happens when a new service obsoletes an older one?

The problem with SOA is that there is no SOM (Service Oriented Maintenance). And because maintenance is 90% of the cost of software - how could this IBM guy say that you will save money?

Posted by: Roger Ramjet   Posted on: 05/05/09  (Edited: 05/05/2009 @ 11:18) You are currently: a Guest | Members login | Terms of Use

Alert moderator to an offensive message

Subscribe to this discussion via Email or RSS

SponsoredWhite Papers, Webcasts, and Downloads

• The True Costs of Virtual Server Solutions VMware In an economic environment that is repeatedly heralding the message "do ... Download Now
• Three Steps You Need to Know to Stop Data Loss Varonis Sensitive data exposed to misuse or loss... it is the stuff of nightmares ... Download Now
• Why Isn't Server Virtualization Saving Us More? A Few Small Changes May Dramatically Increase Your Efficiency VMware Companies have rapidly adopted server virtualization over the past few ... Download Now