4 ms·

You're right, it really should be expressed the other way: integrate skill with respect to time and you get the company's business model. Presumably skill is a

You're right, it really should be expressed the other way: integrate skill with respect to time and you get the company's business model. Presumably skill is a vector-valued function that traces out the company's core competencies over time, and naturally the "Business Model Canvas" is a surface that represents the accumulated area along this function. The point the article is making is a corollary of this. You want to choose co-founders that are linearly-independent in skill-space and maximize the magnitude of each of the desired components over time. Over a large enough interval, this is the same as optimizing on the first derivative of skill, or maximizing in skill-learning space (preferring higher values at t=0, ceteris paribus).

beachstartup 12y ago
uh, right. in english, is you want to find people you can actually work day to day with, who are actually capable of learning things and doing things on time while communicating problems. in reality this tends to be the following kind of people: * generally not your childhood friends, though i've given contract work to childhood friends and it turned out fine. * professional acquaintances / previous coworkers / someone you've seen work on a day to day basis * people you've talked with on irc for years and who aren't insane * major contributors to open source software you use * a reference from someone in an entrepreneurial network
- nostrademons 12y ago
  The interesting thing I've noticed about cofounders is that basically every rule about cofounders has been broken by at least one billion-dollar startup: "Don't be a single founder"? E-bay. "Don't do a shotgun wedding. The founders have to have known each other for a while and worked on projects together before they jump into the stress of a startup"? DropBox (met through a mutual classmate's introduction after Drew's first YC application) and GitHub (met at a networking meetup). "Don't do an uneven equity split"? Microsoft, Facebook, YouTube, Amazon, Netscape. "Don't be an asshole"? Apple, Oracle, SnapChat. "Make sure everyone's willing to quit their day job or not return to school"? Apple (Steve Wozniak initially wasn't willing to leave HP until their VC convinced him), Google (Larry and Sergey tried to sell the company and finish their Ph.Ds, but nobody was willing to buy at the price they wanted, < $1M). "Get a technical cofounder"? This one's harder, the only companies I can think of are Auctomatic (which had technical cofounders after it merged with the Collison brothers), Digg (which didn't do so well), and Travelocity (which was a spin-off of American Airlines). This doesn't mean the rules are invalid, but it does point out how many exceptions there are. I think that usually a good test is simply that it feels right, that you want to work with that person.

beachstartup 12y ago

uh, right. in english, is you want to find people you can actually work day to day with, who are actually capable of learning things and doing things on time while communicating problems. in reality this tends to be the following kind of people: * generally not your childhood friends, though i've given contract work to childhood friends and it turned out fine. * professional acquaintances / previous coworkers / someone you've seen work on a day to day basis * people you've talked with on irc for years and who aren't insane * major contributors to open source software you use * a reference from someone in an entrepreneurial network

nostrademons 12y ago
The interesting thing I've noticed about cofounders is that basically every rule about cofounders has been broken by at least one billion-dollar startup: "Don't be a single founder"? E-bay. "Don't do a shotgun wedding. The founders have to have known each other for a while and worked on projects together before they jump into the stress of a startup"? DropBox (met through a mutual classmate's introduction after Drew's first YC application) and GitHub (met at a networking meetup). "Don't do an uneven equity split"? Microsoft, Facebook, YouTube, Amazon, Netscape. "Don't be an asshole"? Apple, Oracle, SnapChat. "Make sure everyone's willing to quit their day job or not return to school"? Apple (Steve Wozniak initially wasn't willing to leave HP until their VC convinced him), Google (Larry and Sergey tried to sell the company and finish their Ph.Ds, but nobody was willing to buy at the price they wanted, < $1M). "Get a technical cofounder"? This one's harder, the only companies I can think of are Auctomatic (which had technical cofounders after it merged with the Collison brothers), Digg (which didn't do so well), and Travelocity (which was a spin-off of American Airlines). This doesn't mean the rules are invalid, but it does point out how many exceptions there are. I think that usually a good test is simply that it feels right, that you want to work with that person.