Monday, December 02, 2013

Tips for successful hackathons

As many of you may be aware, Salesforce.com hosted a hackathon at the recent Dreamforce in SF. All did not go well.

http://blogs.kqed.org/newsfix/2013/11/25/salesforce-hack-scandal/

Oops. Its never good when reputable press wonders whether favoritism or cheating was involved. You can be pretty sure your hackathon wasn't a home run when that happens.

I've helped set up and helped run a few hackathons. Even with prizes, altho not $1m, for sure. I'd say a few things are important to keep in mind. Most importantly, the hackathon is for the developers who commit their time to participate. If you're sponsoring a hackathon, and you remember that the number 1 goal is to make sure developers walk away feeling like they learned something, maybe met a few new people, and had fun, you won't go too far wrong.

Judging should not be in secret and the rules of the hackathon should be so clear and simple there is little chance of screwing up. Ideally, in my opinion, the developers who participate in the hackathon should vote openly to decide who wins.

Probably having a $1m prize, while perhaps appealing to your corporate sponsor's marketing department, is an anti-pattern.

Its probably wise to discuss intellectual property up front. Pick a license for the code developed, or have each team or dev pick a license.

These guys have some good suggestions for successful hackathons:
http://thenextweb.com/dd/2013/09/19/five-things-we-learned-about-organising-great-hackathons/

A lesson I learned the hard way as a hackathon organizer is to make sure you have a vegetarian-friendly set of food choices. It was an awkward moment and a hard-learned lesson.

Anyway, Salesforce reviewed and clarified the results of their hackathon:
http://blogs.developerforce.com/developer-relations/2013/12/update-on-salesforce1-hackathon-feedback-review.html

The ever-astute Barb Darrow of Gigaom captures the essence, as she so frequently does:
http://gigaom.com/2013/12/02/salesforce-com-hackathon-do-over-nets-basically-the-same-results/


No comments: