I owe it all to something I read today on one of my favorite blogs, TechCrunch. It was a surprisingly offensive post called “Amateur Hour Over at Twitter?” composed even more shockingly by none other than Michael Arrington, Mr. Techcrunch himself! In the post, Arrington utterly nukes Blaine Cook, who recently parted ways with Twitter after being one of its chief architects since its inception. I am convinced that there’s some major subtext behind this story between Arrington and the subject of his public flogging of Blaine Cook, . What else would explain the level of scrutiny that Mr. Cook endured in such a public manner? How could anyone place so much of the blame for Twitter’s woes solely upon one person’s shoulders? Was Blaine responsible for some of the problems that Twitter endured (as well as all of us)? Sure, that’s a given, but to label the work that he and his team accomplished as “amateurish” is just plain inexcusable.
The Rumpelstiltskin Effect
After doing some more research into this I came across more starling information that changed my own opinion and view of Twitter and any other web service for that matter. Apparently, one of the most successful communications successes of our time (twitter) accomplished their massive popularity with a staggering engineering staff of three (3) people, including the “incompetent” Blaine Cook. Not too shabby for rank amateurs, huh?
Look, as a member of I.T. and over a decade of experience with networking infrastructures, I will admit having a slight bias towards under-manned tech staffs. I know what it’s like to deal with the “spin gold from straw” expectations like these guys have done for the past few years. That is why it’s so disappointing to see anyone get disrespected for a job that really should’ve been praised instead. To learn more about the complexity of the Twitter infrastructure and the tremendous challenges they have dealt with, simply checkout Blaine’s presentation: Big Bird (Scaling Twitter). Most of us won’t fully understand the magnitude of their challenges but I think this presentation will give at least a glimpse at some of the reasons Twitter has experienced the problems they’ve had. As a matter of fact, I’ve come away from all of this in amazement that we haven’t had far more serious problems than we’ve all had!
I spoke to a couple of friends who are rock star Ruby on Rails developers who’ve seen Blaine’s presentation and who use Twitter on a regular basis. Our conversation was incredibly enlightening and revealing. The following image is a basic diagram of a simple Ruby on Rails web application, which is what Twitter was built upon. Just imagine a diagram for today’s Twitter. It would dwarf this design because they encompass countless databases and web servers that span the globe with no end in site. The more of us that sign-up and add others to our network, the larger the Twitter databases and network grows. Obviously, this also means the greater the possibility of performance problems and downtime. I’m not making excuses, just telling it like it is, folks. It’s all part of growing pains.
While writing this I just read a news flash on Techcruch about Twitter’s VP of Engineering and Operations Lee Mighdoll leaving after only three months. No one seems to know what’s going on at Twitter these days. Obviously, there is something major taking place at the corporate level. Rumors are running wild about Twitter scrambling for funding and potential sale of the service, but it’s all speculation. The lack of a visible business plan and revenue generation stream only adds more fuel to the fire of these rumors. All I know is that we shouldn’t be so quick to throw anyone under the bus without first learning more about the facts. I know I’ve been bad about that in the past and I will not repeat those mistakes ever again with Twitter. These nameless, faceless services have living breathing human beings who take pride in their work and careers. Also, there’s always more to the story than meets the eye.