In the startup world, the biggest benefit to being able to do some level of technical work seems to be being able to more quickly attract a truly high quality technical co-founder. If you just have an idea and promise equity, you'll have a tough time attracting top talent. If you have an an idea, promise equity, and are willing to pay a ~market salary, you might do okay - but the idea better be really great since your money will run out.
If you have a prototype/demo and some level of traction, you can much more easily attract better technical talent.
As a programmer, I would be more interested in working with a non-technical founder who has these skills:
- domain experience
- a good network in the target market / sales leads
- ability to manage customer relationships and close sales
- good communicator
- familiarity with software development process, even if as an outsider
This set of skills would balance well with mine. I don't really care about working with highly technical founders as the second or third person at the company - that's my role. I want to work with someone who can sell and make sure that the product gets in front of customers.
The caveat here is that my interests lie in business systems, not in doing the next Facebook/Twitter/4sq clone.
the biggest problem with non technical founders, from my perspective as a programmer who has dealt with many of them, is that they don't have the foggiest idea how long it should take to build something. I've heard far to many say something like, "I told an investor we'd have this built in 3 months".
I generalize of course, the best person I've ever worked for was a non technical founder and he had an uncanny ability to understand this stuff at some level despite no direct experience in it.
If you have a prototype/demo and some level of traction, you can much more easily attract better technical talent.