On 6/4/07, Bob Mottram <[EMAIL PROTECTED]> wrote:
[...] Judging by the volume of text generated so far
on this subject I expect that anyone joining this sort of venture will
waste a lot of their mental energy determining precisely who owns what
and arguing over the details of the mechanism for allocation of
profits based upon contributions (does number of lines of code equal
contribution, and if so is this likely to lead to a good AGI design?).
Yes, measuring contributions is the crux of the problem.
To measure *each* item of contribution,
self-rating +
(optional) peer-rating +
(optional) peer complaint +
(optional) managerial board arbitration
is the only feasible solution I can think of. Peer-rating should be made
optional as we cannot expect all members to rate every contribution -- too
time-wasting.
Trying to measure a member's *overall* contribution over a period of time
is problematic because peers can't have photographic memory of the myriad
contributions made by an individual in a period.
The problem is compounded by the problems of assessing:
1. short-term income-generating ability
2. risk of the contribution being ripped-off by theives
Interestingly, self-rating +... can solve all these problems, provided that
we have a significant number of well-behaving and honest members.
Using a non-existent AGI to rate contributions... is not a realistic idea.
So if someone can suggest a better, realistic solution, I'm willing to adopt
it. If not, I still believe that self-rating +... is worth giving a try.
YKY
-----
This list is sponsored by AGIRI: http://www.agiri.org/email
To unsubscribe or change your options, please go to:
http://v2.listbox.com/member/?member_id=231415&user_secret=e9e40a7e
