On 10/25/05, GerardM gerard.meijssen@gmail.com wrote:
Hoi, Given that it is only a few wikipedias that can be considered to somehow cover subjects adequately, the plateau that may arise is still in the future. To suggest that we have to take the emergence of a pleateau into consideration, you have to identify why it is likely to arise. Personally I find it easier to find arguments why we are NOT going to reach this plateau in the near future..
So the best we can do with your "sound accounting practices" is to have multiple scenarios available to us. Not because of accounting practices but because they impact what we do and how we do it. Given however the many projects that need funding and the wishes we have to have an impact in the languages where we are still weak, it does not seem reasonable to expect an end to the growth that we have experienced over our first few years.
Thanks, GerardM
I've been speaking with various people off list and the data we have isn't nearly as bad as I first suspected. It's actually more a problem with communication of the data to the public that's lacking. I have my doubts that the hardware model can hold over the long run, whether there is a growth plateau or merely less exponential growth. I believe there are multiple mistakes being made which are to some extent cancelling each other out (on one end there is the fact that old hardware is being reused and on the other end there is the fact that the system is not scaling perfectly linearly). But as far as predicting the actual hardware requirements on a quarter-to-quarter basis I don't think there's a problem. Also while looking at the data I've been quite impressed at how well the system does seem to be scaling up. Maybe I was just plain wrong when I looked at the system a year or two ago, or maybe the situation has improved more than I was aware of. Probably a little of both, but either way, I think a lot of people are doing a good job.