Ask YC: Is Y Combinator itself a startup?
Having said so many things about startups, to what extent is it recursive?
I very much disagree, though with the specifics rather than the general point. It's very difficult to keep a full mental model of both the technology you're using and the problem you're solving in your head at once. Either you veer over to the technology side, building something that's cool but not as useful to users, or you veer over to the business side, coming up with great ideas for users that can't be implemented in any feasible time period. This is one major reason why startups can fail to build something people want.
You need at least two people that trust each other to keep everything centered - one to build things and one to critique what's been built and suggest ways that it can be made more user-friendly. It doesn't have to be hacker + businessperson: hacker + hacker with each of them writing code and then looking at what the other has written for improvements is probably more efficient, and gives you a swing person if you need something implemented fast. But it's very rare that it can be done by a single person.
I've found that separating out the technology and business concerns (at least in the companies that I've worked at) leads to, occasionally, bad solutions, tension, and a lot of the creativity is squelched. Sometimes, good ideas look uninspired from merely a business perspective, uninteresting from a technical perspective, but in reality they're great ideas.
A good example is the DOS licensing deal. Making a CP/M compatible operating system (poorly) isn't that bright. And from their business perspective, IBM couldn't see it's platform being commoditized, because they failed to understand how portable software would become. But Bill Gates understood.
I completely agree. Infact I am a single person and can deal with both parts. But this idea is in conflict with what YC has said: "We're reluctant to accept one-person companies, though we have funded a couple."
my comment was from my perspective not a vc. Of course when I put efforts on the business side of things I feel I am as effective as doing the development. But as someone looking at a VC as an option instead of self funding, I think that I could reduce my efforts on the business side and do more on development.
The original comment was "we worry if we had a lot of money..." my question/comment was if more money is bad then offer more expertise.
"Ideally you want between two and four founders. It would be hard to start with just one. One person would find the moral weight of starting a company hard to bear. Even Bill Gates, who seems to be able to bear a good deal of moral weight, had to have a co-founder."
That does seem to be true. But I wouldn't think that implies a need for specialization, nor that it would help.
For further comments on the subject, Philip Greenspun has written substantially about how small teams of two or three consultants handling everything associated with a project would outperform, in terms of time, money, quality, and customer and consultant satisfaction, more specialized, supported teams.
That's contingent upon picking the right people, and training them the right way: communicating the right values, building up a well rounded skill set, getting people thinking about the customer and how to them to pay for stuff. When VC's took over and the structure was rearranged to encompass business specialists, and hr experts and so on, the company rapidly collapsed. There were other factors too: Joel Spolsky writes that the consulting market collapsed in general, but ArsDigita had big customers, which actually didn't go away, they just started doing things in house because ArsDigita got worse.
He explains some of this in the following link: http://itc.conversationsnetwork.org/shows/detail94.html
I would probably apply design-by-contract methods. HR, finance and others do play a good role in a company but as long as they are contained :)
I guess startups would eventually want to become a real company and there are costs in that as well. you can minimized that but you can not completely avoid them
I bet other hackers dislike working in 'real companies' too, so it might even be a magnet for more independently minded hackers. The most independently minded hackers are likely to go off and start their own, though.
"awesome" by the way your word not mine.