I don't get it. At all. Let's use me as an example: I'm currently in my 20s, a software developer with his mind and heart set on Mars. I'm Elon's target market, basically. So let's say I spend my money to go to Mars and fulfill my life's dreams and goals. You're suggesting that I shouldn't go, because 'git push' and 'git pull' are more expensive? Hah. Ridiculous!
> There's no way a guy on Mars could compete on costs. I think you're also underestimating how much of a premium bandwidth will be too.
No, I don't think so. Latency is very high, but data rate isn't going to be all that bad - in fact, it might be spectacular as the medium of transit is the near-vacuum of space. Though I agree to a certain extent that costs will generally be higher when building software for Earth.
But slow down here, because this was just one example of a segment of a potential Martian economy. We clearly are not going to predict all the details this far out, and we're already debating about the data transmission costs. I think the point here is to take a step back and realize that Martians would certainly have a lot to do, would have their own small (though growing) economy, and that exports to Earth will likely be mostly digital due to shipping costs for physical objects.