Web Projects as an Deep Water Oil Rig

Imagine a world where large critical projects never catastrophically fail. No huge BP oil spills. A world where T-mobile Sidekick users didn’t wake up one day to find their phones had lost (literally) their life into the great bit bucket in the sky.

Why is this so hard to get right? In the software arena, missed project deadlines, or worse buggy software, are chronic problems that have plagued the industry since day one.

In the web space, projects suffer even worse. Often, unlike their software brethren, web projects don’t have dedicated QA (quality assurance – testing), network and deployment topology is a neglected after thought, and the under-resourced team has to live with a schedule that basically consists of “get it up ASAP, we will fix it later”.

Some days this is cause for despair. If a big company like BP can’t get a drilling project right, how can we keep a large web application smoothly running, given all the cards stacked against it in the first place?

It is the very transient nature of the web that saves us. Sure the code was slapped up there ASAP and sure it was hacked and tinkered with to massage it to ever changing requirements (”well now that I see it … it needs to …”), but web infrastructure often doesn’t last. Web sites are continually evolving and changing. A project is doing well if it lasts more than 3 years without a major rewrite. Vendor platforms get swapped in and out like a new pair of shoes.

With the web we can tear down the foundations and rebuild at the whim of the latest VP. It just takes money. We don’t have to get it right the first time, unlike BP. And given the chaotic nature of the web .. that’s a good thing.