Ask HN: How long do you stick to your idea before calling it quits?

My friend and I have spent a lot of time in the last year building things which we hope will be a good match for the needs of our respective markets. In the same time period (9 months) that I'm still working on one idea (http://www.plentyoftweeps.com), my friend is now on his fourth(!) one.

I've been doing a lot of thinking about comparing our approaches, namely:

1. Taking one idea and running it into the ground, trying every which way to find a product/market fit.

vs.

2. Spending 2-3 months on an idea, releasing it, seeing if it "takes off", showing consistent growth on some scale early on, and if not, moving onto the next one.

I used to think #1 was superior, but at this point, I'm not so sure that I'm in a much better position than had I taken approach #2.

Which approach do you guys prefer? Do you do a lot of experimenting with little projects to see what sticks, or do you tend to devote a long period of time (1 or 2 years, say, in lieu of massive, obvious success) to something before you draw conclusions about its validity as a business? And in either case, how do you know when to quit?

2-longish term 5+ months incubation/development, and 1 throw it together in a weekend and see what happens.

The thrown together site got much better traction than the well prepared.

I'm launching another 3 day project (hopefully tomorrow), so I'll know more than.

One thing I have found, and I think your language points to it as well, is that when you are completely dedicated to an idea, you're trying to make it fit. Whereas if you are in the mode of seeing if their is a market (re: faster projects), your model of 'success' is if it gets early traction.

I thought that a 5 month project was reaching for minimum viable product. Now I'm realizing that minimum viable may be less than viable as a product. Really more like minimum viable market test.