Re: [Usability] Allow donations for precise bugs

[Jon Manning <desplesda gmail com>]

We appear to be talking about two different street performer protocols.
The one I'm talking about is the protocol outlined by John Kelsey and
Bruce Schneier at http://www.firstmonday.dk/issues/issue4_6/kelsey/ .
Your proposal is very similar to their idea, except that the idea for
creating the {product/patch/feature} comes from a user, rather than a
{programmer/artist/author}.

Their paper also discusses likely problems with the protocol, such as
the author running off with the funds and not delivering (solved by
using escrow). Escrow's also used to handle situations where the
donation level doesn't reach the target.

My question is, what happens when two or more developers say that they
are both willing to do the task? I see one of two things happening:

1. The first developer gets a 'lock' on the task, and nobody else can
say that they'll do it for the money.
2. More than one developer can register an interest, and people can give
their money to whoever they think will do it best.

Option 1 seems a bit unfair to people who feel they can do it faster and
better than the person who got there first, and Option 2 might shift the
focus from the task to the person doing the task (which may or may not
be a good thing).

Also, what if the task is implemented independently? Would the money be
refunded, or would the developer be paid a portion of the funds,
depending on how far s/he'd come along with the patch?

It sounds like an interesting idea :-)

Jon

On Sat, 2005-03-12 at 11:29 +0100, Maurizio Colucci wrote:
>
>1.  With the street performer, usually you pay *after* listening a song that 
>you like. You pay out of your *generosity*. If you didn't, nothing would 
>change for you. You have already got your service. OTOH, in my proposal, you 
>pay *before* the service. And for purposes that are completely *selfish*.
>
>This may sound cynical, but I believe a serious model cannot rely on people's 
>generosity (this is utopistic).
>
>
>2. But let us assume you pay the street performer in advance. In this case 
>too, there is a difference. With the street performer, after you pay, if he 
>decides not to play, you don't get the money back. And the street performer 
>has no interest in playing well (unless others are watching). OTOH in my 
>proposal, if the threshold is not reached, you get the money back. And if the 
>coder does not finish in the given time, he gets a bad reputation in the best 
>case (he gets sued in the worst case).
>
>
>Maurizio