I agree...it took a while on the phone to get to the heart of what the fellow was asking about, and I've arbitrarily called the types of business "community" and "value" since that's the only thing I could think of that really fits the difference between these very different businesses. Auctomatic and Zenter are obviously not even a little similar, and yet when one asks "did their value come from their community or the utility of their software?" the answer is clear. Since none of these things are really black and white, I just fudged a bit to try to bring to light the "ah ha!" moment I had during the conversation.
I'm obviously not suggesting that people shouldn't build community-based startups or that YC shouldn't fund them. Just suggesting that maybe a useful application is a safer bet than an application that has to reach millions of users to be a valuable acquisition target (or to make money). But, if we wanted safe, we'd be working for Microsoft or Google, right?
Let me suggest we might call them "tool" or "service" businesses.
I think this is a useful distinction, but it's also one that breaks down at a point. Reddit, for example, was a useful way to kill time even before it had a big user community, just because it had a few good posts each day. If kn0thing and spez sent you that list by email rather than maintaining it on the web, it wouldn't feel like a community but like a service. Finding a service that can be delivered usefully without a big community, but can also become an identifying activity for a growing community, seems to be the secret sauce of many successful community sites.
Let me suggest we might call them "tool" or "service" businesses.
I think this is a useful distinction, but it's also one that breaks down at a point. Reddit, for example, was a useful way to kill time even before it had a big user community, just because it had a few good posts each day. If kn0thing and spez sent you that list by email rather than maintaining it on the web, it wouldn't feel like a community but like a service. Finding a service that can be delivered usefully without a big community, but can also become an identifying activity for a growing community, seems to be the secret sauce of many successful community sites.
I'm obviously not suggesting that people shouldn't build community-based startups or that YC shouldn't fund them. Just suggesting that maybe a useful application is a safer bet than an application that has to reach millions of users to be a valuable acquisition target (or to make money). But, if we wanted safe, we'd be working for Microsoft or Google, right?