gods wear hats

I’ve been playing with Google’s App Engine as part of their beta and I started thinking about what benefit Google got from it. Presumably, there will be some kind of monetary payback as those apps which exceed their “free” limit then start paying, but at the prices they are charging it seems unlikely that this will be a big money-spinner.

Google like to use Python. They purchased Jaiku over Twitter because Jaiku was Python and Twitter was Ruby (I’m sure there were other reasons too. A lot of their APIs are in Python, as are many of their own apps.

Google only releases big scale apps. They released Google Calendar and it allegedly had over 1 million signups in the first week. Google Analytics likewise had stupid numbers of signups (so much so that they couldn’t cope with the volume and had to shut down registration). In fact, this kind of “Instant-One-Million-Users” (that needs its own acronym) means that Google can’t afford to release anything unless it is massively scaleable.

Calendar was a home-grown app for them, whereas Analytics was an acquisition. Maybe this is one reason why the former was able to cope with IOMU and the latter wasn’t. With this in mind, enter App Engine.

Now you can write apps that meet Google’s criteria: they are Pythonic, massively scaleable and best yet: already run on Google’s infrastructure. So, when someone writes the next big web app, it just got a whole lot easier for Google to acquire them and then avoid death by IOMU.