In the late 90s I made my living as a rock musician. I stepped down in 2000, just in time. Jobs like my old one are disappearing fast; artists with an existing fan base are getting better at exploiting it cutting out various middlemen, while newcomers can get very popular very fast and with near-zero cash investment on Last.fm or Spotify, but they find it next to impossible to build a solid economy. They make music in their free time, the core skill is landing a day job that will pay the rent and allow you to go on tour. I’m told something similar is going on with videomakers and film directors. A deadly cocktail of low cost producton technology and online sharing has thawed massive reservoirs of creativity, turning it from scarce into plentiful. As it did so, it drove a stake through the heart of the music business, which turned into dust like a vampire at high noon (obituary by Dave Kusek). A texbook case of the creative destruction predicted by Joseph Schumpeter.

Fine, but startups are the new rock’n’roll, right? It’s the same narrative: bright, visionary youngsters, obsessed by their own ideas and in sync with the Zeitgeist, becoming millionaires at 23 and inspiring their generation to break away from the dull existence of their parents. And boy, do the young go for it: business plans contest have replaced rock band contests (the latter have migrated onto TV talent shows, maybe low on cred but endowed with a rock-solid business model).

And yet. Recently I found a post by Laurent Kretz, founder of Submate, that describes in merciless detail the hardships of the typical startup entrepreneur’s life. No dream job: it involves living on welfare, sleeping on friends’ couches, getting the boot from girlfriends tired to beg for attention, “being chased by your banker with a chainsaw”. A story all too familiar to my friends of CriticalCity, who triumphed in the end but paid a high price for it.

Mind you, this is not someone looking for a backer. Kretz does have an investor. A seed investor, that gave Mashape just enough money so that four people won’t starve to death as they put in 80-hours weeks for four months. This seems to be all the rage with early stage investor: I remember Joi Ito two years ago saying he only invests in a company that can bring him a working prototype coded by three people who locked themselves away in some cabin for a weekend, and even then he’ll only invest fifty thousand dollars.

Since then Ito has made progress. his latest experience is that a fully functional web services can be launched in three weeks by exactly two people: one designer, one developer. This is because code is modular: you don’t write it from the ground up, you just copy-paste existing routines. This process has been made more fluid by “metaprogramming” tools that stitch together chunks of code coming from different directions into an integrated program.

It does not take a genius to figure out that the startup world has entered a “hey, I can do that too!” phase. Since there is a limit to the market’s ability to absorb new stuff, software development skills could be on their way to becoming plentiful too. Before the job market reacts, we could even have a period in which a software engineer’s time is worth as much as a rock guitarists’, which is to say less than a baby sitter’s. It’s Schumpeter curse: when the market works, it commodifies or obsoletes everything.

Most economists interpret Schumpeter’s careful wording as a process that turns out to be a good thing in the long run, because it makes useful, expensive stuff cheap. I think they are wrong, because Schumpeter’s is not an equilibrium model: if creation and destruction are not in sync, the long run may never come around. What this means, I’m trying to figure out.