Has there been any discussions anywhere about setting up a scheme for managing micropayments and/or bounties on CPAN and/or the RT infrastructure?
With the recession and everything, it seems as such a useful feature to have.
There has been a few attempts over the years to do something like this.
I remember one attempt that tried to assign bounties in terms of "beers". Fix this bug, we owe you three beers.
The problem here is not the payment part, it's the micro part.
Most of the people who write CPAN modules that other people depend on a lot about are professionals. If you assign payments to those packages, you need to be able to hit a number that is actually worthwhile.
For example, current my $work is paying the author of a module to improve it in several ways we care about (documentation, memory leaks, etc etc). But we're paying full industry rates for this. I'm quite sure that by the time it's done we will be looking at a bill in the high 4 figures at least.
At some point, the values become low enough that it can almost be seen as an insult.
Re:Not worth it
mpeters on 2009-06-04T20:13:38
My $work has also paid CPAN modules to add features we want or need. It's probably pretty common. But a system that allowed us to contribute a few hundred dollars combined with your few hundred, etc could easily reach a point where it's not an insult. Also if it's structured as a bounty then anyone can do the work. I think a system like this could work as long as it was done right.
The hard part would be making sure that the money got paid to right be people when the work was done in a way that's satisfactory to the parties involved. This would probably require some sort of 3rd party holding money in escrow and then having to deal with the legal (local and international) is probably enough to send most of us running.
Re:Not worth it
bart on 2009-06-05T16:51:47
A "couple of hundred dollars" is not micropayment. The latter is more like "a few dollars".Re:Not worth it
mpeters on 2009-06-05T17:04:07
Right, but what I was saying is that it adds up and it makes sense that the intended recipient of the work wouldn't do it until it added up to enough to make it worth it.
And in to those micro-payments the few hundred dollars that companies would be willing to throw in for specific bugs, etc and you'd have a pretty good amount. Such a system could even allow prospective developers to set a target bounty that would need to be reached for the work to be worth it.
Re:Not worth it
sjn on 2009-06-04T22:40:53
The problem here is not the payment part, it's the micro part.
Most of the people who write CPAN modules that other people depend on a lot about are professionals. If you assign payments to those packages, you need to be able to hit a number that is actually worthwhile.
I was thinking of a model similar to pledgebank.com - several people pledging a "micro" amount, and when the sum is high enough to motivate someone, then something gets done - with maybe some trustworthy middleman keeping track of each micropledge or actual micropayment until the task in question is completed.
IMO, it's not that important wether or not it's worthwhile to do for the money alone. We're mainly talking about open source development here, so motivation for fixing bugs or making features are already in place - the money would probably not function as more than a "sweetener" (at least to begin with.)
For example, current my $work is paying the author of a module to improve it in several ways we care about (documentation, memory leaks, etc etc). But we're paying full industry rates for this. I'm quite sure that by the time it's done we will be looking at a bill in the high 4 figures at least.
First, thinking this as a replacement for hiring someone fulltime is probably wrong. Furthermore, if a feature or bugfix is worth a four-figure amount to a company, there's nothing wrong in "bidding" that amount using any channel the company trusts - including a micropayment service like we're talking about.
Secondly, if the intention with micropayments is to give developers a good enough income to compete with "full industry rates", then yes - only people willing to accept low "wages" will take on a job. At least in the beginning.
I don't think that should be the goal though (even if it would make a difference for freelance programmers in India or eastern Europe or wherever.) The purpose, in my opinion should more like this:
- Give end-users and others without enough tuits or know-how a new way to influence or show appreciation for a project, without them having to hire a programmer. (The idea is to lower the barrier of entry for participation in open source projects so much that a couple magnitudes more people can get involved than possible today. More people => more money circulation.)
- Give developers another useful feedback channel from the public, by letting users put money behind their votes.
- Give another reason for increasing transparency and code understandability in open source projects. (Would _you_ pay someone to do a job behind a curtain, just to have them show you a black box when done, and say "It's done, really, now hand over the money"?) I think reducing the "black box" problem like this would give open source projects another very significant market edge.
:) - Open up for creating new revenue models for Open Source development - even if it takes a while to establish. If an individual one day gets enough revenue to make a comfortable living (even if it's in the countryside in Slovenia, like Jonathan Worthington has done) the GREAT. But it will certainly not happen if we say "Not worth it".
:-P At some point, the values become low enough that it can almost be seen as an insult.
Maybe. One person's insult is another's praise. Let the market deal with it.
;-)
