There's an online debate right now over whether the economic climate might adversely affect SOA implementations.
Brenda Michelson first raised the question last week at SOA Consortium Insight, in a blog post that shared how various members of the consortium responded to the question during a conference call. Surprisingly, most of the callers thought a weak economy would actually spur SOA adoption.
ZDNet'sJoe McKendrick piggy-backed on this discussion -- which, it turns out, he had broached way back in November. And it snowballed from there.
My take is that this is a conversation of more importance to SOA consultants and vendors than your average IT shop. That's not to say it's irrelevant. But let's be honest. When it comes to building your service-oriented architecture, this topic only matters if it's your SOA at risk. If the economy doesn't impact your SOA, then who cares if 65 percent of the other SOA implementations are put on hold?
An argument could be made that a trend to halt SOA implementations might give you a bit of bargaining leverage when you're hiring a SOA consultant ... but, probably not.
So rather than summarizing the entire "will it or won't it" debate, let's cut to the chase. What should you do if the economy forces company cut-backs, which, in turn, threaten your SOA? Here's a list of tips, compiled from the various SOA experts who've commented on this topic:
1. Focus on short-term ROI. Despite his insistence that SOA shouldn't be a project, David Linthicum knows that's how it's still functionally viewed at many organizations. So, in this post, he offers three excellent tips on what to do if your SOA faces budget cuts. At the top of the list, of course, is making sure everyone understands the ROI of your SOA.
But as Alastair Bathgate, who runs Blue Prism, an enterprise software company points out, this isn't the time for long-range ROIs. Your business case needs to focus on delivering value in the short term. Excellent advice.
2. Think tactical, not strategic. Remember those arguments over whether you should even discuss the concept of SOA with the business? Even if you feel the business can and should understand SOA -- as I did -- now is not the time to start explaining. The OnStrategies Perspective blog points out that the recessions of the early 1990s and 2000s caused companies to target special projects -- which, in practice, translated into strategic or architectural projects, including the shift to client/server architecture.
Jason Bloomberg of ZapThink LLC said in a SearchSOA.com article that the best strategy in 2008 for companies that have not yet embarked on SOA is to focus on projects where SOA can offer tactical cost savings.
3. Tie your SOA to Web 2.0 and social networking projects. As one IT director pointed out during the SOA Consortium discussion, telecommunications and other industries are playing catch-up with social networking and Web 2.0 technologies -- and SOA enables the back-end for these technologies.
4. Tie your SOA to emerging business changes. In the same discussion, a chief architect mentioned that a weakened economy in the U.S. will prompt businesses to look at emerging and overseas markets to diversify. Another change he expects is more interest in peer-to-peer business interactions. Both of these business shifts could open up new opportunities for using SOA.
5. Focus on the dead or dying. Again, from the SOA consortium post, an industry practice leader points out that if the platform's dying, then you might see companies invest in SOA. But otherwise, he expects many will settle for putting services on top of old applications. Not the best move, but a likely scenario.
6. Reroute funds from other projects. This one is another from Linthicum's post. Sometimes, you've got to rob Peter to pay Paul.
7. Try Guerrilla SOA. Some people have criticized this bottom-up approach, saying it's not the best way to build a SOA. But sometimes you've got to forget what's best and do what works, right? McKendrick's post references a Webcast with Jim Webber, ThoughtWorks SOA practice leader, which I've written about before. Webber offers a lot of great advice about tying SOA to business projects and deciding whether or not you need a vendor's solution.
For more on this topic, you should also check out "How SOA may thrive in a recession," published on SearchSOA.com yesterday. The article offers tips from Linthicum, McKendrick and four other SOA experts.