Several posters have taken it as a given that software ought to live for a long, long time. That's a philosophy worth considering, but many people just want to

Several posters have taken it as a given that software ought to live for a long, long time. That's a philosophy worth considering, but many people just want to make money. If you lived in Atlanta during the '96 Olympics, you might have rented out your house for $10K for the month. That's a great "business" even though you know it will fail a month from now. However, you would not be rational to note that rents are now $10K/month and buy up a bunch of houses using that rent as a long-term factor in your business plan. There's nothing necessarily wrong with building iFart, etc. But, it's absurd to pretend that revenues for fads like these will continue. I am reminded of an article I read about the business of opening up trendy nightclubs. The experienced owners think that it's impossible for a club to remain trendy for more than two or three years. So, they open them but build a business plan that doesn't anticipate long-term revenue with the expectation that they'll build their next one while the current one is still doing well. Note that I believe Facebook/iPhone apps can be a great extension or component of a long-term business model, but that model can't be predicated on the continued popularity of those platforms alone.