It's not really a Catch-22; if you can't sell your idea well enough to get a coder interested in being your cofounder, then the fault lies in either your idea or your pitch, not in some insurmountable market force.

There are a lot of talented developers out there who haven't come up with a billion-dollar idea of their own yet, and who would be happy to hitch on to someone else's if they're getting a fair piece of the action.

However, if you're trying to convince them to do work for hire (i.e., give up any rights to their own work) at far less than market rate with some vague promise of more compensation later, you deserve to fail. The developers who will fall for that argument are going to be inexperienced and/or just plain incompetent; the rock stars (who you really want to work with) know the value of their work, and are unlikely to jump on a bad deal.

I know a number of non-developers who have had an idea, pulled together a team with no funding and built pretty impressive products, which they've then used to raise funding.

The reason they pull this off is because they don't expect 50% just for having an idea. That's ridiculous. If you're not a developer, you better figure out what else you're providing and what you'll be doing so everyone feels like you're justifying your cut.

There are tons of other things to do, like raising capital, finding free office space, getting content, public relations, finding more team resources, etc. Most will rely on connections -- if you have no connections and you can't code, well, then it's going to be more difficult. Better hope you have a lot of charisma :)

(I'm a developer and I end up advising a lot of aspiring entrepreneurs)

I think the problem is that even if suggested idea is good and executable, usually these idea-people know little about nothing how to execute the idea. Typically they know nothing of industry, running a business, web, planning, making products, design and never nothing technical. Programmer should just figure out all the features, design and business plans while programming.

My advise is that you have to sell the idea to programmers same way you would sell the idea for investors. You have to do your homework, have some plans, and make clear what you expect from them and what you can bring to the table(eg. money, connections, expertise). You won't go to investors asking for millions with a idea written on a napkin.

Aye, the thing that a non-technical founder can bring to the table is industry experience and connections.

One small company I worked at was run by former investment bankers. They knew VBA obviously, but that was it (the app was C++ on Solaris). But they knew the foreign exchange business inside out, and "sales" consisted of them having lunch with their mates (which BTW bagged us 7 of the top 10 FX banks as clients.... no programmer could have done that).