From Phoenix to Mercury

Sigh.  The "next Multiply" activities are very discouraging.

First, we have anotherblogspot.com.  I have to applaud herbearjc for taking the bull by the horns and making something happen, but I don't think it will last.

He took an existing FOSS application (good plan), put it on a commercial host (for starters, another good plan), and is hacking away at it trying to make it look like Multiply.

Unfortunately, the starting point is a LONG way from Multiply.  I doubt very much he will be able to get close enough to make it appealing.  There is something about Multiply that makes it so much nicer than Facebook or Blogspot, or Wordpress, or livejournal, and it is hard to put your finger on just what that is.

But there is another problem.  The site is already quite slow, and as it develops and gets more users, it will get slower.  If it keeps expanding, the bandwidth costs are going to become unmanageable.  That's the thing about the off the shelf hosting providers; they are quite affordable for low bandwidth sites, but as you add users and increase the bandwidth demand, the costs explode.

Plus, as basically one individual doing this as a personal project, at some point he is going to get fed up with it.  I give it two years, tops.

Paul and Kimberly are going at it from entirely the other direction.  They realize they need a business plan so they are working that angle. Now, Kimberly clearly has a LOT on the ball.  Paul hasn't really provided me enough data to be able to judge, but there is at least some evidence he hasn't got a clue.

Their problem is that they are convinced they need to build their own datacenter. Not only is this an expensive undertaking, but it entails long term costs that are far more manageable when shared with others through some commercial host or the cloud.  The probability of them ever getting off the ground seems slim, although they are at least looking at some of the hard problems.

Their main concern seems to be privacy, which is really kind of worrisome.  They have this illusion that if they control the hardware it will be more secure than if a team of experts is adding to their own security.  This is a fairly common fantasy among folks with relatively little experience.

If they ever get it off the ground they are going to have to charge big time to support their private infrastructure. I don't see it becoming a reality.

herbearjc may have a more economic model, if he can 1) find some income source and 2) recognize the need to change models over time.  He can buy bandwidth and cycles, and at some point move to the cloud or to a large scale provider, perhaps first the cloud and then to an Akami or somebody like that.

But he somehow needs to build a sustainable business structure, and I see no evidence of that happening.