Too much 'whining' and not enough action on SOA?

CIO's Nicholas Petreley had a few choice words about all the alleged handwringing that has been going on around SOA lately: Stop whining and fussing, and just make the stuff work.

Petreley invokes the expressions "QQ," which means to complain or cry about something, and "pew pew," which is supposed to be the sound a laser beam makes, signifying aggressive action.

"...in short, "Less QQ, More Pew Pew" roughly translates to, 'Stop whining and bitching and get off your keister and do something.' As much as the expression irks me, QQ is an appropriate description of what many people seem to be doing about service-oriented architecture (SOA). 'It's too expensive. There's too much of a learning curve. It's too much of a culture shock. We have to retrain our developers. We have to hire a squad of mercenary consultants. Studies have shown that SOA causes dreaded combination skin....'

"...there's money to be made with SOA. More important, you can save 50-gallon drums of money with SOA. But you can't realize those savings by QQing. You have to take the first step and start implementing SOA."

Frankly, I don't think there's been a lot of complaining in lieu of action on SOA. I've seen plenty of evidence that there are a lot of SOA-inspired or SOA-related efforts taking place out there. Many are under the radar, or pilot efforts. Most are more likely JBOWS (Just a Bunch of Web Services) than anything close to fully functioning SOAs. But that's okay, because it's a start.

In addition, there's nothing wrong with a good rousing debate on what SOA means for an organization. Perhaps there should have been more debates like this before companies invested millions in ERP and CRM systems in the past. Yes, SOA provides organizations the opportunity to save substantial amounts of money, and we've documented cases over the years here at this blogsite. But expectations have often been built up beyond what JBOWS-type architectures can really deliver at this stage. (Still, it's interesting that 37% of companies in one recent survey considered their SOAs to be meeting expectations.)

Petreley also echoes thoughts expressed in this blogsite, which is, basically, don't spend a lot of time planning to start, simply start. Start now. You don't need a grand master five-year plan -- Just do it. Petreley also says there's plenty of tools out there to make initial SOA implementations relatively easy.

The big challenge is the organizational headwinds that SOA proponents -- the ones that may be positive and confident about the approach -- encounter. It's difficult to move out of the JBOWS stage with resistance from the enterprise. It takes a series of incremental wins -- with demonstrable savings or gains -- to get the business's attention. As the business signs on, JBOWS evolves closer to true SOA.