The thing about close friends is that the friendship will help you a lot when discussing the business, there are a lot of business that break out because a bad relationship among founders.
More broadly speaking, I suggest starting with "what kind of person do I need?" and go from there. For example, if your idea is to build a social networking site for environmentalists, I would see if I had any friends studying Environmental Science. Or if you're a business guy in need of a coder, I would hang around the Computer Science department.
My present founding team consists of people I knew, trusted and was friends with prior to starting work on my product. But they did not start working with me at the same time I started out.
The co-founder relationship will form naturally when/if both of you are ready to start a company. You are in a great place to meet good hackers. Who knows, maybe some of the friends you are meeting now will become future business partners.
One of my co-founders was just a friend in school who I worked together with on projects. After school, we began working at the same company before starting to talk about leaving to do a startup. But if I would have asked him in school I'm sure he would have said "are you crazy".. Some people just need some convincing :)
Another co-founder is a cousin but we've been talking about this stuff for over 10 years so that choice was obvious.
I knew his technical skills, but he was more than that, a great guy and a family friend.
Though there were over 20 developers whom I managed, but he was the chosen one :)
only after a 2 year period, we went on to start our company.
my suggestion, get someone whom you dont mind meeting at anytime of the day, any place in the world, with whom you share many things in common.
startups are stressful, dont get someone who has financial and personal problems( more than you).someone who can commit minimum one year of his/her life for the Idea and you
You should come to the next Devhouse http://superhappydevhouse.org/SuperHappyDevHouse21. It's on Nov 10th in SF.
Also try to find someone that can complement you, i.e. if you know your weaknesses then find someone that is strong at those (of course things that matter to your business)
Cofounders at first sight, incorporated forever.
It turned out so right,
For strangers in the night.
1) you're looking for someone who has similar values yet contrasting /knowledge/skills/abilities (KSA)
The similar values makes them a 'potential friend' and that's the best you can aim for. In the past I made the mistake like so many co-founder do, of connecting with ppl because of our common KSA. At first it's very easy to get along with someone who is equally passionate about, say, linux. Over time your conflicting philosophies will lead to a poopy_work_environment(tm)... at which point you'll mostly like start the "why I hate this person" list to help validate breaking apart the company. I assume it's very similar to a marriage/divorce.
2) Understand that nothing is more important than the 'who'
The person/ppl you choose to build a company with will make you who you're going to become. If your values are similar you'll agree quickly on possible problems and get up into a high level of creativity when brainstorming solutions. The process will be rapid and you'll alter your ideas with nothing more than an "um" from the other person/co-founder... You'll know exactly what they're thinking. You're a product of your environment so nothing is more important to you than the 'who'.
Let's say you get 100 ppl interested in working with you, how do you choose? I recommend tracking down someone who's studied Industrial and Organizational Psychology. I was lucky to find two post-grads from a local university (UW) here in Waterloo and they suggested I read the textbook "Human Resource Selection, 5th ed., by R.D. Gatewood and H.S. Field). They also assisted with a few filters. The best I'd say is to create a 'structured interview'. This will help you interpret a persons values.
4) pick a small 'what'
Once you've gathered your 'potential friends', choose a small problem/project to work on -- even if it's for free. Make sure the project is meaty enough (6months?). The purpose is to alpha-test your startup. Especially watch how/if you all resolve conflict. I suggest agreeing before hand on the most pragmatic approach possible... something like, if we reach an impass than we'll flip a coin... watch how the person who didn't get his/her way responds (especially if that's you).
5) try find someone who has already run the race -- successfully!
Imagine this 'potential friend' was someone like pg. Their insight will handicap your odds of success.
6) Do not rush it. At worst get a job at Burger King and wait.
I hope that's coherent. I apologize for the length of this post, it wasn't intended. If you have further questions, you can email me at firstname.lastname@example.org
Seems like we have a fairly prim audience here...