bounties, bounties, bounties - oh so chocolatey and coconutty

[Mark McLoughlin <markmc redhat com>]

Hey,
	So, I was surprised and disappointed to see the announcement[1] of
another round of the bounty hunt. Since I never sent out a mail
detailing my experience with the last round of bounties, it was
suggested that now would be a good time.

	As panel maintainer I knew I was going to be affected by the
clock/evolution integration bounty and had misgivings about the whole
thing from the very start, but really wanted to at least give it a
chance.

	Now that its over and done with, there are quite a number of things
that has left me feeling unhappy about the whole deal:

  - Throughout the whole process I felt under a huge amount of pressure 
    to make sure the feature got in and the bounty got awarded. What 
    with the fact that money was involved and there was a huge amount of
    hype, I really felt like I wasn't in the position to say "sorry, no 
    time right now". I felt like I had to make time, and ended up 
    spending a substantial part of my Christmas vacation working on it.

  - Presented with two competing patches, neither of which we 
    acceptable, I really didn't know how to proceed fairly. Tell both 
    that they weren't acceptable and they shouldn't get any money for 
    their work? Proceed with both, going through multiple revisions of 
    both patches until they one was acceptable and let the other 
    contributor get nothing for all his work? Just pick one, ignore the 
    fact that it wasn't perfect and let the other contributor walk away 
    with nothing? No, in the end I started from scratch again using 
    ideas from both patches and awarded the bounty to both submitters - 
    fair, but a huge waste of my time and the contributors didn't get a 
    chance to learn through the patch review.

  - The bounty was announced without any consultation with me (as the 
    maintainer) whatsoever. There was an assumption that this was a
    feature which belonged in the panel and that it should be 
    implemented a certain way. I'm still not entirely sure I agree that
    it does belong there but either way, it felt like I didn't have a
    whole lot of choice in the matter.

  - The bounty immediately went very high up my priority list. Why? Not 
    because I felt the feature was any more important that anything else
    on my priority list, but because people who were expecting to get 
    some money wouldn't get it unless I prioritised it. I resented my 
    priorities getting set like that.

  - Overall I just felt that it was wrong for the awarding of money to 
    be so closely tied up in the our development process. "Who decides
    what features are important? The one with money! How do we get
    hackers to do something? Offer them money!".


	Anyway, there were some positives out of this - the feature got
implemented and the hackers involved are still hacking on GNOME (I'm
curious would Jon, Christian and Martin not be hacking on GNOME in any
case - I think they would). However, if this time around there was a
bounty for a module which I was a maintainer of, then I'd be asking to
have it removed.

	I wish I had some good suggestions on how this could be done much
better, because I know there are good intentions here. But to be honest,
its left me with such a bad taste for the whole idea I don't want even
want to get into a discussion about it.

	(Also, I'd echo the concerns Mikael just mailed about - the Firefox
thing, in particular, is very worrying)

Cheers,
Mark.

[1] - http://mail.gnome.org/archives/gnome-announce-list/2004-August/msg00025.html