SOA 2009: Do we need architects or firefighters?

ZDNet blogging colleague Ian Finley just posted an interesting question that reflects ever-shifting priorities in today's rough-and-tumble economy: "When the building's on fire, who calls an architect? ...

ZDNet blogging colleague Ian Finley just posted an interesting question that reflects ever-shifting priorities in today's rough-and-tumble economy: "When the building's on fire, who calls an architect? ...No one does. They call the fire department." He advises enterprise architects and vendors alike to put down their plans and ESBs, put on firefighters' hats and boots, and get in there and help battle the flames.

It makes perfect sense that priorities need to be focused on marshaling resources on the "right projects." Ian says the current economic situation is actually a great opportunity for enterprise architects and vendors to show the mettle of SOA, and I agree. This is the time to demonstrate how SOA can deliver greater efficiencies, agility and faster time to market than the old way of doing IT. And, indeed, we see SOA returning to its more entrepreneurial roots, with more projects addressing pressing tactical issues, led by smaller, more agile teams.

Interestingly -- and I never considered this -- Ian compares the current time to the Y2K crisis of a decade ago. As he puts it, "You have to go back to the Y2K panic to find a time when so many companies had the same #1 requirement."

However, just to put things in perspective, there's nothing new about the crisis "firefighting" mentality regarding IT deployments. In fact, that's been the problem all along. This mentality has stretched IT resources thin for years now -- through good times and bad. In fact, firefighting may even be more intense in fast-growth times, because companies are scrambling to add new lines of business and increase market share, creating plenty of digital waste and stretching IT resources way beyond their limits.

Yes, enterprises need to be faster moving and more entrepreneurial. But IT planning looks two to three years down the road, as it must. When you drive, you're not looking down at the space in front of your bumpers -- you're looking 100, 200, 300 yards ahead to the horizon, to anticipate what's going to happen.

Remember, economic slumps don't last forever. It's likely the recovery will kick in sooner than we think. And when it does, it will be fast and furious with pent-up demand, especially as business technology wishlists come back online. And the firefighting will continue.

A deliberate, more long-term architectural approach is needed as well -- throwing hardware, software, resources, and consultants at problems willy-nilly in firefighting mode is expensive and gums up agility, and that's what keeps happening.  And that's what SOA is supposed to tamp down.