This weekend, I did with Mehdi a blitzweekend session, the first of its kind. The goal was to make a product in 2 days, from concept idea, design, development to deployment. This is a real challenge in itself, something that is possible now thanks to technologies like ruby on rails.
We began Saturday 8.30 am, each having our own web application idea. I wanted to build a game, and his idea was to make something like crunchbase for Montreal.
If we fast forward to Sunday evening, I can say this was both an intense and memorable experience. Time constraints is a huge challenge, and forces you to execute, both in good and bad ways. Here are my notes, some of them are subjective, some are not that important, but we are writing them all for the next session, in hopes to make it better:
- it’s essential to write specs and sort them by priority. I had features of the application grouped into batches, from required to nice-to-haves. Looking at it again, half of the planned features were developed.
- this is obvious but one needs to be deconnected from RSS, twitter, email etc. This was super easy. No family/friends either. Somehow this was arranged. There were a couple of things I couldn’t have not gone to, but this is by far the most productive and intense days I experienced in months. I estimate I spent about 15 full hours during the two days for the exercise.
- There is no time left to look up for documentation, tutorials, reference docs etc. You just have to know your tools. If you don’t, it’s best to implement a feature with something you already master
- Time flies. Knowing that you have to deliver in a few hours shuts down the whole world and inputs. Before you know it, 4 hours have passed.
- I had another look at the code this morning. All I can say is that it’s not the most elegant code you will find out there, but hey it works. I also had a look at the results, and I can’t believe I took 2 days to do it. I guess I lost time in design concepts and design.
- coffee is a great beverage!
Mehdi finished his crunchbase application. The result for me is at startupweb20.com; the game was planned at the beginning to be a parody with what’s happenning in the “web2.0″ world. You can sign up and launch a startup*, and try to win millions before the whole ecosystem crashes. This was deployed at 9.30pm, which is 37 hours after we began. Not sure now what I am going to do with it, I guess it will scraped for the next blitzweekend session.
* note: the usual path is signup, create a new startup, hire 3 people for your startup, then develop a feature.
Comments
Denis Canuel October 08, 2007
Seriously, pretty interesting and this makes me think about an article from the guy who started “Plentyoffish.com” where he said that even though you have no inspiration, it’s amazing what you can achieve once you sit down and start coding.
That’s what I usually do and it works.
Heri October 08, 2007
although I didn’t begin developing until I had a very very clear vision of what the website should be about.
by the way, ideally, we should have blitzweekends with different teams of 5 and there should be a guy doing the marketing and business plans. but we were just testing the concept this time.
Denis Canuel October 08, 2007
I wonder if having a week long version would produce better results? For my current project (YulNews), it took about an hour or so to jot down the big features but it took me many days to refine everything until I was happy. Even now (as I code) I change some pieces here and there (mostly depending on my mood and user feedback).
Sometimes, small ideas can go very far. For example, refactormycode.com, which BTW could be applied to many mainstream domains like cooking, designing your house, your girlfriend’s makeup or dressing habits, etc. I only took about 10 seconds to think about it but the point is, there are many domains left wide open and sometimes small projects like this can fulfill a niche and be very profitable. I think that the refactormycode guys just opened a nice door to many new collaboration based ideas…
Heri October 08, 2007
In Montreal, we might do another test in 2 weeks, one before the “real” one where everyone would be invited. I guess you can come at the one in 2 weeks.
The interesting part is limiting yourself to 2 days. It’s the pressure that counts. If you give yourself 7 days, everybody knows you will give the best of yourself only the last 2 or 3 days. Also, I don’t think we will get a group of 30 to 50 people avalaible for 1 week.
the refactormycode guys is 1, and that should be marc-andré
by the way, I wasn’t inspired by refactormycode for this. this was planned from 2 months ago.
Denis Canuel October 08, 2007
However, and I will stick to what I said earlier, I think that the whole picture can’t be achieved in such a short time. You need more time to really think what you will be offering, consult with others, etc. unless you’re doing a very small project :)
Quebec Valley » Blog Archives » Upcoming StartupWeekend in Montreal? October 08, 2007
[...] from MontrealTechWatch, mentions about a possible StartupWeekend / BlitzWeekend in a few weeks. He shares his experience and it seems like it was very [...]
Open invitation to a blitzweekend meeting | Montreal Tech Watch October 16, 2007
[...] may recall the first post mentionning an event called blitzweekend, where the goal was to make a product within a very limited timespan, i.e. 2 days. We are now going [...]
[official] Blitzweekend | Montreal Tech Watch November 07, 2007
[...] you follow Montreal Tech Watch or if you met me around Montreal, then you’ve probably heard of [...]
Lakeesha Besemer January 24, 2011