Multiplicators
If you've been following my twitter feed you see that I've been doing a lot of infrastructure setup lately. We are now laying the groundwork to for some real work later. Some of the work we're doing at the moment is setting up puppet. The Goal: Automated and reproducable server configuration and setup right from the start. We're also figuring out the optimal configuration for Hudson to have properly tested builds. Plus we are spending time with Python - specifically Python Paste, Pylons and web.py, to find an architecture we're super-comfortable with. Oh and we're also fiddling with the way we do Scrum as we're all new to it.
All this takes time and keeps us from building our actual product. So why do we do it - especially as a startup?
I like to think of these technologies as multipliers. Each of them - Puppet, Hudson, Python/WSGI/Pylons/web.py and Scrum - will multiply our productivity. And it gets even better: Bind together enough of these innovations and you multiply the multplier. The result of these tight bonds? Raw speed.
Speed is of essence as startup. That's the one of the very few advantage we'll ever have against the big guys in the beginning.
Case in point: By spending some quality time digging through the Paste Deployment documentations and understanding setuptools, I was able to find a great configuration for a modular Pylons setup. Something I had to manually in Okapi with a few weeks of work and fiddling. Now it's done in a blink of an eye.
