[revorad]

I bet Drew Houston wasn't spending 8-10 hours a day sharing files, neither was Zuck looking up student profiles all day long.

I don't agree with the notion that you have to understand the ins and outs of something to start working on it. See http://paulgraham.com/schlep.html

Do you think all of Basecamp's customers are software companies because that's the only domain 37s knows deeply about?

Funeral directors use Facebook too.

I wouldn't think "wow, i've got a need/want for better funerals". I would probably notice some specific problem and think how to solve it. If software can help, that's great, otherwise I'm not starting an undertaking business any time soon.

[mgkimsal]

You end up needing to be willing to commit to learning a hell of a lot about an industry, and that's time you won't be developing. The tech, in most cases, is secondary to understanding the business needs and workflow of an industry. One of the most successful guys I know as a small business software guy took a year to take mortuary classes and work in the funeral industry to learn their needs before developing services for them.

[BadassFractal]

I agree with this. If you want to be an entrepreneur, it doesn't matter if you start a construction company, a funeral service or flower delivery. You might not ever write a line of code and still have a perfectly profitable fast growing business.

The prob is that we're all tech geeks here and all we want to do is turn our fetish for coding into the next big thing, so we always try to think in terms of software solutions.

[AznHisoka]

Facebook is an outlier. About DropBox, I'm sure Drew Houston spent every living hour digging into the intricacies of other solutions, the risks, what people wanted, etc when he decided he wanted to make DropBox.

You can pick up the domain knowledge while working on your product or before working on it, but you still have to get it somehow.