Linden Lab is a wonderful company that's working miracles every day. But it doesn't have experience managing a booming business, and that shows. It needs to bring in people who know how to take a company from emerging startup to billion-dollar giant.

I've been thinking this for some time now, but I was moved to write by Akela Talamasca of the blog Second Life Insider, who responded insightfully to our interview with Linden Lab co-founder Philip Rosedale.

Talamasca questions Rosedale's devotion to expanding Second Life at all costs. Rosedale told InformationWeek: "Our mission is to get this technology to everyone in the world as fast as we can, and we'll make sacrifices in marginal revenue to maximize that goal. We don't want to limit anyone's access to Second Life."

Talamasca responds that Linden Lab needs to balance the drive for growth with the need to provide a stable system for its users.

Second Life is buggier'n the mattress at a cheap Las Vegas motel. Any time you log in to Second Life, you'll find a better-than-even chance that something important is broken. Just today, I found myself unable to instant message other people in Second Life, and therefore, had to delay a business meeting 20 minutes until we could connect over the regular Internet.

Read the blogs of Second Life users and you see a constant catalog of bugs. Most serious are the ongoing lost-inventory problems, where in-world virtual property just disappears for a while. As far as I know, those disappearances are just temporary, but it's tough to build a business based on a service as buggy as Second Life.

The relentless drive for growth exceeds the platform's ability to support its user base, and that leads to a very low retention rate. Linden Lab itself says that only 10% of people who try Second Life stay with it.

Talamasca says:

They reached 6 million plus accounts, but how many of those stay? Very little. Is it enough to get the tech out to everyone in its current, slightly broken form? Perhaps LL should take a cue from Blizzard, the company that makes World of Warcraft: release nothing until it's done, regardless of how long it may take, or how much pressure you feel the fans are putting on you. Blizzard is, without a doubt, one of the most envied, beloved, and successful game companies in the world -- and its active userbase is vastly greater than SL's. They take the time to get their products right the first time, their UI is pretty easy to understand, and in the months and months that I've been playing it, I haven't once experienced a single crash or enforced downtime after an update has been released.

Linden Lab also needs to communicate better. Rosedale told InformationWeek that Linden Lab is committed to facilitating communication between people above all. Talamasca responds that the statement is ironic:

"[M]any residents feel that their concerns aren't being met or even acknowledged at all. This is very much a top-down system, where LL lets some knowledge drop, the residents get it, speculate wildly for a while, then in the absence of any firm confirm/denials, eventually give up and get on with their lives, hoping for better times.

I'm sympathetic to both sides of this. Linden Lab has an unruly customer base. Second Life user forums on the Web and the comment threads on the official Linden Lab blog are overwhelming. I seldom read them, and when I do, I'm crushed by the sheer volume of complaints, some significant, others petty and childish. I can see where Linden Lab might see its time as too limited to wade through all that.

Moreover, Linden Lab expends enormous effort communicating with its user community. The official blog is pretty detailed, and it has many town hall meetings in-world every week.

On the other hand, the user community's complaints are also reasonable. It really is hard to tell what Linden Lab is doing. It statements on upgrades and new features are often quite cryptic, and clarifications sometimes confusing.

Case in point: The expected launch of voice Wednesday. I began to get a sense that wasn't going to happen midday Tuesday and started calling around and making a pest of myself in the company until I finally got confirmation at 7:30 p.m. EDT. That's crazy. Serious technology companies don't make a mystery out of when they're going to deploy a major feature. They let people know.

I expect Linden Lab will learn this kind of thing over time. But why bother? There are already people at Web 2.0 companies that know how to take a business from breakout startup to billion-dollar giant. Hire one of them, make him or her CEO or COO, and leave the people at Linden Lab to do what only they can do, which is to continue developing and operating Second Life.

In hiring this new CEO or COO, Linden Lab would need to avoid making the mistake that Apple made in the 1990s, when it ousted Steve Jobs and replaced him with professional management. None of Jobs' successors really understood Apple's strengths, and the company floundered because of it. Instead, they need to follow the lead of Google when it hired Eric Schmidt, or eBay when it hired Meg Whitman, and hire outsiders who get the company vision. Linden Lab's new management needs to understand not just how to run a fast-growing tech business, but how to run Second Life.

P.S. The people at Linden Lab are pretty smart -- I wouldn't be at all surprised to learn that they're already one step ahead of me on this.