- Management scared to try and tackle big engineering challenges, sticking to tiny iterative changes and expecting to "move the needle" with tiny often silly "improvements"
- Inability to address engineering technical debt - management is simply scared of it
Second Life was created more like a cutting-edge research and development project than a traditional commercial product. The delivery of visionary capabilities took precedence over long-term scalability, performance, usability and reliability. Given their inability over the past decade to resolve foundational technical limitations (lag, crashing, etc.) the root causes are probably woven so tightly into the legacy code and architecture that they can't be unwound.
My guess is that Linden Lab calculated that even if they solved all current issues by investing a million or two on reengineering legacy systems, revenue would still decline over time. OpenSim and other emerging platforms are offering land at a tenth the price of Second Life. As the competition's communities, economies and capabilities mature, Linden Lab's customer base will continue to erode. A dramatic cut in land pricing isn't an answer because that would create a corresponding cut in revenue. There is no sign that the market for sandbox worlds like Second Life will significantly increase in the foreseeable future. So there's really not a lot of upside for them. It's no wonder that they made the decision to take advantage of their current cash-rich situation to invest in new products that offer, at least potentially, a much higher return on investment.
If my analysis is correct, what does the future hold? I think we'll see a continuation of incremental improvements to the client, small fixes for issues that don't require a large investment and new features to encourage the use of Second Life as a gaming platform. But a few years down the road, unless Linden Lab can pull a magic new feature-set out of their hat, a declining Second Life and an emerging OpenSim will tilt the balance and we'll see a fast decline in both land ownership and active users.