[jstayton]

I want to push back a bit on this statement:

> Or can one random person, working in their spare time, build just the right app and reach millions of people?

I agree that it's very difficult for one person; however, to win doesn't always need to be defined as becoming the most popular, or reaching millions of people. What if it's 1,000 people paying $10/month? That is possible for one person going the service-oriented route.

This isn't a nitpick about Sandstorm per se — I do love what they're doing — and I understand they're making this argument because what they do isn't a service. But I do think it's possible to find diversity/decentralization in small, independent service-oriented companies that don't care about being popular on the scale of Facebook or Google.

[kentonv]

> What if it's 1,000 people paying $10/month? That is possible for one person going the service-oriented route.

It may be possible for a single person to pull this off today, but it's quite a lot of work and responsibility, and it's hard to tell if it will pay off until you've put in a lot of upfront investment. Someone who writes a useful application for themselves on a weekend isn't likely to want to engage in this. But if all they have to do is upload the package they already built to an app store, that's a much easier pill to swallow.

Conversely, as a user, I am not very likely to want to put my data into a service run by one person, especially if the data is sensitive or if I don't want it to suddenly disappear. But if I can run the app on Sandstorm where it is automatically sandboxed and firewalled, with the data staying on my own server, and the app can never disappear, I'm much more comfortable.

[nullcipher]

No offense to sandstorm but most of the apps indeed look like they were done in weekend. The blog author grossly underestimates what is required to make a great functional app (even those mobile apps that were done on a weekend aren't popular by any means). If this is the goal, then they should reposition themselves as "OS for distributing your weekend project" (no snark, am serious about this). It's a great niche if that is their initial target.

> and I understand they're making this argument because what they do isn't a service

They do provide a service.

[jstayton]

> They do provide a service.

That's true. I should have worded it differently. They're trying to attract developers to build apps for their platform, rather than launching them as a service (i.e., SaaS).