Doing Something Big
by Mike Levin SEO & Datamaster, 08/12/2009
My “something big’s” seem to come in 5-year intervals, because the result of that big something appears to have about 5-years of life in it before I feel the world has changed sufficiently to warrent a large overhaul of my methodologies and workflow. Now is such a time. In 2000, it was a generalized system akin to Ruby on Rails for the digital signage folks, Scala, Inc. In 2004, it was HitTail, the real-time topic suggestion engine for search optimized publishing for Connors Communications.
Today, it’s… well… keep reading.
I still vigorously believe in my last “something big” as few folks have actually even tackled the real-time Web analytics reporting in the way that real-time actually implies. The “cloud” has worked AGAINST real-time due to how it discourages tightly coupled highly optimized server work. You will be seeing MORE real-time stuff from me that meets ACTUAL user expectations of the term real-time, which flies in the face of cloud conventions.
But my next something big extends this into the world of Google services, and API’s at large in a big enough way that it constitutes a complete shift in my approach to development work. Along with giving up my old Microsoft safety blanket of SQL Server (still a kick-ass product), and IIS, I’m taking up the much more Googly Linux/Python/Google Docs/API’s as my platform. So, is this a cloud approach or a local server approach? Well, it’s a hybrid that gets incredible user interfaces and data storage capabilties from the cloud, and incredible performance and breaking of arbitrary security contexts from a (multiple) local servers.
I think I’m truly onto something big, getting the best from what both the cloud and distributed computing have to offer. They each have their relative advantages and disadvantages, and building a meainingful project that extracts the best of each is like solving simultaneous equasions in your head while swinging on a trapeze.