Hara Ra wrote:
"
>Establishment of a committee on a world level for the management of a
>data base counting the various problems
>of scientific nature which one can solve by methods of massive
>parallelization. **time-out** the committee
>himself occupy also to carry out à a classification by command
>preferably of these problem, define the tactics of
>distribution automatic of code de computer with processor unoccupied
and
>to count the result and them put
>accessible with public.
Ah, the centrist approach. I'll sell my cpu time at a market rate, provided:
What do I need a central committee for? "
Just to propose all the applications available and to gather the results in
the most efficient way...
No need, of course, for only one central committee. But it could help to
reduce redundancy and to have a de facto homogenous system (less
incompatibilities). It could also work like a central market where the
peoples could be aware at any time of all the applications available and
choose those they want to be computed on their machines according to their
utility functions. The
last is probably not very scientific, but it's a way to reward peoples for
the use of the infrastructure they pay for.
If I find the applications interesting I will certainly not ask for money in exchange. I think I could even want to pay a little to have the satisfaction that my machines are also working on good projects ;). Security of course is an absolute necessity.
delriviere
christophe