zerosum dirt(nap)

evolution through a series of accidents

zerosum dirt(nap)

Constraint-Driven Development++

February 08, 2008 by nap · Comments

There was a nice writeup on ReadWriteWeb yesterday about Montreal’s upcoming Blitzweekend event. Like the Rails Rumble that we organized last September, Blitzweekend is another attempt to give people a “weekend to build a functioning startup from scratch”. Unlike the Rumble, it isn’t a competition, and it places added value on physical proximity.

Josh Catone (of ReadWriteWeb and also RailsForum) dubs this sort of thing “organized ultra-rapid development” and notes that, against all odds, this strange weekend-startup phenomenon seems to be trending up. This is really just constraint-based development taken to the next level, and the reason it’s exciting is simple: it pushes you to deliver results, immediately.

Results, of course, are what makes application development exciting; seeing a movie get automatically reencoded after upload, a search query produce a number of restaurants in my area, plotted on a map, a transaction get authorized and recorded by the third party API you’re using. And later, seeing your end to end workflow in place and your first user signup. When we see results, we get motivated to work harder in order to see even more impressive results. When we don’t see results, we get depressed, we step back, we overthink. Our motivation falters.

Events and contests like Blitzweekend help give developer-entrepreneurs that extra motivation kick that they often need to get over the motivation speedbump, and an opportunity to gather a team together and just go for it. If you haven’t tried it, you really should. In the process you’ll find out if you can work with your team in the long haul, write some (hopefully) great code, and figure out whether or not the idea has legs to stand on. If it does, you’ll have visible proof of this and will probably be more jazzed than ever to continue working on it. If it doesn’t, you won’t have to spend any more time wondering “what if”. All in 2-3 days.

(It’s important to have realistic expectations of course. Are you going to be able to build the next Flickr in 48 hours? Or the next Facebook? Nope. But you sure can get a nice first iteration out there and working. Every application has to start somewhere, and if the first iteration of your idea isn’t ugly and lacking important features, then there’s a good chance it took you too long. Short iterations can pack a lot of result-fu ninja punch.)

Anyway, I’ve rambled on for a bit and I’m sorry. We’ll do another Rumble, or something like it, later this year. For sure. But the reason I’m writing is this: You don’t have to wait.

That is, you don’t need an actual event to prove that this brand of ultra-mega CDD works.

Take the last 3-4 days of this month, of next month, of whenever. If you’re a freelancer, just tell clients you’re unavailable during those days. Get 3 other people to do the same thing. Agree on an idea, flesh out the basics. And just do it, Nike-style.

I’m seriously thinking about doing just this, starting at the end of February, maybe at the end of each month. Anybody wanna be on my team?

blog comments powered by Disqus