Could this work?

[netstrider]

Ok so last year I came up with an idea. I figured I'd write it down and save it and come back to it at a later stage. I haven't done any research regarding such a system but I think it might work but might need a little tweaking. Here follows the basic fundamentels of which I've written down that day.

Code2Cash
Ulrich Holtzhuasen
*mail addy removed*
16 October 2007

Goal
To pay developers for their code submissions to project 'X' with cash gathered off
donations and the revenue generated from website adverts. Users also get paid for bug*
reports. Encouraging both more and better code submissions from developers and it also
encourages users to report bugs which in turn will result in a better end*product.

How will it work?
The developer gets paid for the amount of lines he contributes if and only if the
code is accepted into the stable version of the software release.

Users are paid for bug*reports.

Developers earn more cash for resolving reported bugs.

To improve cash income users are kindly prompted at the software's installation
whether they'd like to make a donation. A short description of how the Code2Cash
project works is presented to the user.

Donations are paid by means of PayPal.

A fixed amount must be met for payout. In other words, the developer must reach
five dollars worth of code contribution to receive his/her payment.

Developers get paid an amount that depends on the total cash pool and the
amount of developers.

The Math: CashPool/Developers * Lines of Code. This means that rates will
be continuosly changing.

Users are assigned a fixed amount for each bug report.

[genesus]

It sounds like an amazing idea to me, not sure if there is anything comparable on the market at this time; however, I think something like this would be extremely valuable. Maybe even attract some of those who throw away money on wikipedia donations.

[Thrillhouse]

How will it work?
The developer gets paid for the amount of lines he contributes if and only if the
code is accepted into the stable version of the software release.

I think you would have to come up with a different set of criteria for compensating developers. The number of lines in a piece of code is almost irrelevant. One could space an if statement out into 10 different lines if they so desired (unless you're writing in Python). On the same token, there are often competitions for the best program written on one line of code. I think you'd have to find a different way to quantify a developer's contribution.

Also, how do you assign bugs to developers? If it's a free-for-all and two developers both implement the fix, which one gets paid?

[genesus]

Also, how do you assign bugs to developers? If it's a free-for-all and two developers both implement the fix, which one gets paid?

whoever implements it first...good point on payment per line of code though; however, payment per code, if the first poster could come up with a reasonable way to do it, IS a good idea.

[netstrider]

Thank you for the responses guys.

This is exactly the reason I posted this here to hear your comments and suggestions and whether you think it could work in the real world.

As I mentioned, the system might need a little bit of tweaking. Because I'm not much of a programmer myself your point didn't cross my mind but thanks very much for pointing it out Thrillhouse.

I did mention however that *if and only if* the code gets accepted into the main codebase (stable release). In which case I assume project "X" will have someone who can distinguish good code from bad.

Any other issues, comments or suggestions people?