How Linden Labs becomes 'the next Google', by reducing downtime.

Well, with Second Life being down for the second day in a row, I thought I spend some time writing down my views on what Linden Labs needs to accomplish to become 'the next Google'.

1. Reduce downtime
2. Reduce Lag! Change script / prim limit handling.
3. Open up the server more
4. Internationalize!
   
5. Allow delegation of most functionality to web services
6. Be careful with creating new land
7. Unite mainland and Estate concepts
8. Change tier fee handling
9. Change TOS to accept more responsibility
   
10. Make it browser based.
    

I plan to write about all of the above ten points that i listed above, more or less in order of priority, in future posts. In this post I'll concentrate on the first point, reducing downtime, which is appropriate now that SL is, well.. down.

Reducing downtime is so important, because in order for Second Life to evolve towards the 3D internet dream, uptime has to be an almost guaranteed 100 % for the grid, just as it is with the web today. And just like the web, individual simulators (websites) may be down now and then, but even short downtime can cost you lots of visitors for a long time. If Google would be down, say once every month, for say, 20 minutes, I'd probably not use it, especially if the outages would be unpredictable. At the moment Second Life is in a phase where a lot of people are discovering it. People that are not die hard fans of virtual worlds like the old user base, but that are just checking it out, because they heard about it on the news. Some of those people will become new regular inhabitants, but too much downtime will chase a lot of them out again. Scheduled maintenance is better then unexpected downtime, but in the long run the system will not be deemed reliable, but reliably unreliable. You know for sure you can't rely on the service always, at any time. You know for sure it will be down.