Am 01.05.2010 01:18, schrieb Paul Louden:
>> Sorry about that, didn't read down that far on the etiquette section I
We do like our ettiquette :)
>> Anyhow, it appears this bounty thing is not going to work. Please
>> forget I
>> mentioned it.
Please don't feel discouraged due to individual opinions. We have a lot
of different people involved in the project, and I do think that your
plan is appealing at least some of them.
> It could work, but you need to approach it with a few specific thoughts:
> 1) Recognize that an awful lot of work has already been done, and
> they've contributed it to the project for free. While you don't need
> to acknowledge this financially, you gain a lot of goodwill by
> building a donation to the project into the total to recognize their
I have a strong doubt that this is going to pay off for the donator if
we wants people to work on a specific bit. I also think think someone
offering a bounty is not interested in some goodwill, but really rather
in some work on very specific area.
> 2) Recognize that people are always going to object to bounties (and
> ignore those).
Not all. I think boundy/paid work is a good way to get stuff actually
done in the FOSS world since otherwise work will only be done on a "yea,
in this moment I'm concerned about this, but maybe not anymore next
week" - if at all. People getting paid to work on FOSS is not too
uncommon so I would appreciate if such statements wouldn't be made on
behalf of a project.
> 3) Recognize that your financial structure delivers a message about
> what you consider most important, but others may not (an idea might be
> to allow people to donate toward the sub-projects as they see fit,
> rather than having one pool that's divided by pre set amounts).
It doesn't really matter what's most important to the bounty provider. I
think he made clear that either Linux or Rockbox doesn't make a
difference bounty-wise. And even then, if he finds people that rather
port Rockbox for less than they'd get for porting Linux, then it's his
own decision and he can be lucky if he finds one.
> 4) Try to avoid financial incentives *not* to work in certain areas if
> you're setting up a constant pool. For example if you set the bounty
> for Linux higher than Rockbox, you're also in some ways encouraging
> people to pick it over working on Rockbox if all they're doing it for
> is the money.
See above. It's totally up to the "costumer" since he's not just asking
for it but offering some actual value. Right, it might be wise to not
discourage Rockbox guys by being too focussed on linux, so that advice
is clearly useful. But in the end, it's still up to him, and if he'd
rather see Linux than Rockbox he's probably better of offering more for
a Linux than Rockbox port.
I like the idea of bounties, because it still works good together with
the open source idea. You've not caught into a contract which steals
your copyright or so, you're free to do with your code what you want.
But your work is still acknowledged by something actually useful. I'm
not saying community acknowledge isn't useful or not encouraging, but in
the very end you cannot buy anything with it.
Am 01.05.2010 00:38, schrieb Alex Parker:
> Whilst not wanting to rain on your parade, let me just state my
> personal view.
> I plain don't like bounties. They ignore all the work that has been
> already done, they can lead to bad blood, are hard to administer, and
> unless the money is huge are useless in terms of getting someone with
> the required skills to work on it if they didn't already want to. It
> is also incredibly hard to accurately gauge how much work is needed,
> especially for things like reverse engineering. What if (for example)
> it is found that some chips are the same as those in other devices
> that already have Rockbox drivers? Then someone gets all the cash for
> someone else's work.
> Still, other may disagree :)
Right, I disagree. None of the points you mentioned particularly speaks
against bounties. I especially find the idea stupid to disallow payments
to work on our project just because it ignores the lot of work that
already has been done. Our license permits that without any payment at
all, so why should paid work not work with that idea? The other points
don't speak against paying an individual developer no matter how you
look at it (IMO).
We shouldn't discourage the idea of bounties IMO. If offering a bounty
makes people bring our project forward, then let it be. That's what we
all are aiming for afterall. It's entirely up to anyone who'd like to
dontate or even pay money. Either we wants to see the entire project
leaping forward, then he makes a donating. If he wants to see a specific
area being worked on, then he goes the bounty route. I find both ways
Received on 2010-05-01