The week outside MyPages started with an article in the paper about LiveJournal, based here in Portland, being sold to Six Degrees. Because we consider user involvement so important for MyPages, I'm very interested in hearing what happens with any community-based website. Whether the community is online, as with LiveJournal, or more local like Craigslist.
No outsider is going to know the terms of a deal between two private companies unless something goes very wrong or very right. Either somebody sues or they're written up in Fortune or Forbes. But it seems like a good time to be running a community site.
So it was surprising to see that LiveJournal was sold to Six Degrees.
At the end of the week, I got IM'd by J that LiveJournal had been down for hours because of a power outage at their colo provider, Internap. The first thought was "wow, good time to sell". And the second thought was "wow, glad we didn't colo with Internap". Which we had thought about.
Ultimately, the feeling was one of "it could have been us" instead of Schadenfreude. Based on what I've seen from a glance every six months or so, the folks at LiveJournal seem to have done a lot of things right starting with no money. They came up with smart solutions for difficult problems, like memcached.
Picking the right people to work with is important. Which is a "duh" comment but I certainly wish I knew a magical way to tell that this was going to work or no, it absolutely wasn't. We've had some fits and starts with outside folks on advertising or public relations and programming and web design. We went with hosting rather than putting in our own servers to keep costs down. And having the hosting company short on bandwidth - though it took a while to figure that out - affected the way we felt about the web designers.
In a previous life, I started and ran an internet service provider. One of the reasons I sold was that there were things that I wanted to do, like putting in redundant this or a system to handle a failure of that. I'd been in a few bad situations, where I would have gladly written a big check to get us out of that particular bad situation. But there's some things that you can't easily fix. Converting this system into a cluster or buying a hot swap drive works for some things.
There are the situations you know about, vaguely know about and don't have enough experience to realize that there's a problem ... until it shows up. Then you know about it. Boy, do you know about it. Usually only a few customers are screaming. But all of them are interested in knowing when it's going to be fixed. And everybody gets pretty tired of saying "we're working hard on the problem and we'll have it fixed ... soon".
The people you'd want to hire to prevent those situations won't work for you long-term or at all because you don't have enough interesting work for them to do besides fixing that problem. And if you deal with a contractor, well, they may or may not be available when you need them. So the issue becomes dealing with a company. Which means trying to figure out whether they're going to be around and investing a lot of dollars. Which means that you need to do that very carefully, which means it takes lots and lots of time.
And that's another thing you don't have. Watching the pennies means that you don't have experienced management sitting around to find a business solution (hiring an outside company) to a technical problem. Not being big means you can't hire some people because you can't pay them and you can't hire others because you want them to take webhosting support calls as well as design the ultimate reliable email system.
Once in a while, you get lucky. You get somebody between jobs and his wife is wondering where the mortgage payment is going to come from and you get somebody for a while that you'd never get otherwise. Or you get somebody you've brought along internally who keeps improving. But that also assumes a big enough company that you can hire people as tech support and bring the good ones into technical areas.
So there's a fear issue that got to me after a while. Which is "I know there are icebergs. And I bet there are icebergs I don' t know about. And some of them could put me out of business. Everything I've got is at risk here, 24 hours a day, 7 days a week. And I've got a responsibility to the folks who work here to make sure that doesn't happen. And to our customers who trusted us. And I would like to keep my house and car, too."
I'm guessing that Brad Fitzpatrick decided it was time to let someone else worry about that for a while. Just ironic how good his timing was.