Posted by lambic on February 8, 2010 at 7:24pm
I'm looking for opinions on offering bounties:
- Is it considered appropriate to offer a bounty to fix a bug sitting in an issue queue? If so:
- Should the offer be made to the module maintainer directly or to anyone who wants to fix it?
- What is an appropriate amount to offer (obviously this depends on the size of the fix, but I'm looking for general ideas)
- Is there a way to manage the bounty agreement so both parties are protected or is it simply a matter of trust?
Comments
Not sure there are hard and fast rules...
I would say do both mention in the issue queue generally, and contact the maintainer. If the bug's been there awhile, it's usually because the maintainer is too busy. But I'd always give them first shot, out of courtesy.
No idea about pricing. :( In general, I think a little goes a long way... I know back when I was a module developer just starting out, someone offered to pay me $100 to fix a bug and I just about broke my keyboard fixing it so fast. ;) But if the maintainer is more on the professional side than the hobbyist side, they may have an hourly rate kind of deal, or need to handle it through their employer.
Not sure about bounty agreement thing, either. You could perhaps check with the folks in http://groups.drupal.org/legal. I was always naive and trusted people would pay me if I did a good job.
Comment Bounty Module
A request for a Comment Bounty Module just came up: http://drupal.org/node/713618
Patch crowd funding platform
@lambic I have a good proposal for you and for everyone interested : patch crowd funding platform. I recently started it and am ready for constructive criticism.
Answering your questions:
1) In this case all developers have equal access to all bounties assigned.
Of course, maintainer has a better chance of winning the jackpot - because he knows the code of his module better than someone else.
@webchick So, a maintainer is always prioritized.
@lambic 2) This problem does not exist at all because it is crowd funding, aka co-financing. The amount will accumulate for as long as there is someone who is willing to fix a bug for that amount - the optimal system.
@webchick We all understand that the client is willing to pay for the result while developer needs every day something to put on the dinner table.
In this case the problem does not exists at all because we are different in terms of scale: we have a patch rather than projects. Therefore, it is possible to achieve the desired result (write the patch) and get the reward long before the developer gets hungry.
3) Service pays - and therefore no problems with the agreements. This is something like "adjourned" or conditional donations (or feature donation driven development): the money from interested in a patch first come to the service. When an issue gets “closed” status service rewards the patch submitter.
Here is a link to the project: http://www.patchranger.com .
Thank you for asking, Mark.
Great start! Personally I
Great start! Personally I would like to see a project that adds a priority category to issues. I'm not personally into this for the money, but adding a bounty to an issue is a definate way to add a "real" priority.
My current strategy is the following :
I am relatively new to Drupal and finding a niche within the mountain of projects/issues is quite a task. Maybe your project could helpful to people like me.
Prioritization of issues is highly prioritized task
@nlambert I think I caught your idea : you would like to have a way to prioritize issues to work with them easier, right?
I am also concerned about this question because I am tired of the fact that in the process of working on a paid job I have to catch bugs in contributed modules. This work is not paid at all – but it is quite a big part of job. I had to port Patch Manager module to D7. I guess it is not my own problem only because I see how fast number of Patch Manager usages is growing. Patch crowd funding platform – is a way to solve the problem. Everything is important :
Thank you for your feedback, Neil.
my 2¢, this is going to brake the 'Drupal spirit'
My 2¢.
Making a system of this (i.e. bounties) is going to brake the Drupal spirit, break the community in Come for the software, stay for the community. Monetizing the issue queue, turning it into a systematized market place will have negative consequences on what makes drupal Drupal.
In my view, you're all suggesting we should open Pandora's box. I don't mean this in a bad way. A little extreme perhaps, but you get the picture (i.e. special effects).
There's nothing wrong per se with bounties. Nothing wrong with making a decent living either. But bounties will attract bounty hunters - professional bug killers. In the end, they will get most of the gold. Genuine Drupal developers (including the hungry ones), will not make a career out of bounties. Once the bounty hunter are in, the Drupal spirit will start eroding, guaranteed.
You need an issue fixed? You're willing to offer a bounty for it? Don't ask for permission. Don't waste your time creating a system to facilitate the process. Folks are already doing it on an informal basis - e.g. in Marketplace/paid-services. Just do it.
...
I would suggest that maybe, just maybe, there are other more interesting, community-building, in-the-spirit-of-drupal solutions than bounties...
I must be a 5¢ by now. So I'll stop. :-)
Being. Knowing. Sharing. Ip Man
Drupal spirit won't be hurt
@bisonbleu Your doubts and concerns justified.
Open source community is always on the lookout for anything that is related to money, because it really threatens their freedom.
But let's look at things from the other side. Money - is power. It can raise, and can make dependent. Exclusively by enthusiasm you can not survive, sooner or later (usually sooner) there are those who are willing to buy open source project, lock, stock, and barrel. Is not it better to find a way of self-financing than become some department of a corporation? Let's think together, once the real world is so closely connected with money, unless the proposed method of monetization is not the best of all possible ones?
Bounty hunters? I'm for. Why? Because maintainers of modules are the first ones to become bounty hunters, because who knows code better? This is simply an organized way of donating. The difference is in that it is not in general manner, but pointwise, targeted. These donations are investments in the future - how it will be drawn by the community. Do you want to see Views in core? - Vote by your wallet. Voluntarily, as before. But now your donations are targeted (target - the issue) and conditional (first result - then the reward).
I need an issue fixed. But I do not have enough dollars to provide a worthy reward. So does everyone, I guess. That is the problem. And together - we've got. That's the whole point.
_
you've already proven the very point those who've expressed concern feared.
I just spent the better part of 15 minutes cleaning up your spam in the drupal.org forums and issue queues. That's 15 minutes I could have spend assisting users.
Personally I think a bounty system on drupal.org (a little more advanced than the paid services forum) is a worthwhile idea.
I don't think it should be off site and I don't think spamming links on long dead posts in a transparent effort to siphon off drupal.org users to an external service is a good way to go about lobbying to create one.
Prairie Initiative discussion
To everyone who is interested in the subject of this post I suggest to refer to the discussion within the Prairie Initiative about how to fund the Drupal development : http://groups.drupal.org/node/142779#comment-800273. I propose to continue further discussion there.