Buying a lottery ticket and tweeting demands at Paul Graham are almost certainly feasible methods for obtaining several million dollars by the end of next week. However, neither method is viable because that's not the way things work: to a first approximation, lottery tickets lose and the evidence is that Paul Graham doesn't give millions of dollars to people he doesn't know when they tweet at him.

Unfortunately, it's easy to conflate feasibility with viability because anything that is infeasible is also nonviable: e.g. buying a ticket for last weeks lottery or tweeting demands at Charles de Gaulle. Fortunately, the Brikman's article does not fall into this confusion.

It's easier to determine what is feasible -- two app developers can build a restaurant app and put it in the app store -- than what is viable: build a restaurant app that people use. To be feasible, it only has to be technically possible for the developers to build an app. To be viable, it has to be useful enough that people use it.

One might say that viability is checking feasibility against non-technical goals. If the reason for building the app is to make money, feasibility is based on projected prices, costs, and sales. Viability about assessing those projections.

Brikman's position appears to be that a product is whatever can be put put in front of customers or potential customers. It's reasonably consistent with a world in which many things people tend to consider products are free or fundware. It's plausible.

But I can see why someone might choke on that, so YMMV.