Re: Code Bounty

[Gregory Plummer]

Hi that is a bit dissapointing, I guess your right. Of course, I never intended to under value the work contributed thus far. I could only afford a small incentive but essentially agree with your post.

2008/9/13 <olafBuddenhagen@gmx.net>

Hi,

On Wed, Sep 10, 2008 at 08:10:56PM +1200, Gregory Plummer wrote:

> Hi all, I would like to offer a small cash incentive for tasks needing
> to be completed. It would be great if the Hurd could support file
> systems greater than 2 GB in size. How would I go about sponsoring
> this feature? Some input on the best way to do this would be
> appreciated.

First of all, please don't get me wrong: We greatly appreciate your
offer! :-)

However, I don't think that bounties for completed tasks are the best
approach.

The main problem with bounties is that they usually greatly undervalue
the necessary work: In this case, as Michael explained, it's not quite
clear, as the code is already there, and "only" needs to made ready for
upstream acceptance; however, if it was to be implemented from scratch,
the cost of a developer at a typical salary would easily amount to
$10000 or more...

Now one could argue of course that the bounties are not meant to attract
professional developers, but hobbyists etc. -- however, anyone who can
complete such a task -- even if he is a college kid still -- should have
no trouble finding a much better paid job...

Thus the money itself is hardly motivation enough -- it can only be
considered symbolic, only a fraction of the real value of the work... So
if we indeed have people willing to do it at this symbolic price, there
must already be enough other motivation to cover all the rest -- some
genuine interest and willingness to contribute to the project.

Isn't it a better investment then, to try to further these kinds of
motivation, so they fill also the missing bit; by making the project
more interesting, more visible? For example by sponsoring developer
conferences, or some small PR work perhaps?...

-antrik-