On Mon, Nov 21, 2016 at 06:52:06PM -0500, andreas wrote:

Hi,

I was thinking if a reward system for contributions would add motivation to those who want to contribute.

Or, it could destroy the motivation of those who want to contribute for the love of the language and a desire to give back to the community, and encourage those who don't care about Python or the community but are motivated by the selfish desire to collect points and rewards. Giving people rewards is very risky. You can just as easily encourage behaviour you didn't intend to. Before coming to your suggested solution ("Rewards! Upvotes! Downvotes! Reputation Points! Badges!") perhaps you can explain what problem you hope to solve.

Something in the lines of a reputation system.

If Python had a reputation system from the beginning, I expect that Guido's reputation score would be about -10000 by now because he won't add braces to the language.

This has been done before with success.

Oh? Where? Can you give us more detail please? -- Steve

andreas writes:

I was thinking if a reward system for contributions would add motivation to those who want to contribute.

Something in the lines of a reputation system.

I think if you look at the number of patches sitting on the tracker waiting to be reviewed, you'll see that we have plenty of patch submissions from less experienced contributors. Probably the best motivation for those contributors would be halving the time to review by committers. But new committers don't grow on trees, and they tend to be very busy people. They're generally not motivated by badges. It's true that there are jobs that less experienced contributors *can* do that are un-fun, like confirming defect reports and even triaging issues. These might be motivated by a reputation system, but they're also easy to do sloppily if done too quickly. So we might need reviewers to check them. As Steven d'A points out, it may be more straightforward to have them done by people who are motivated to put in careful hard work in the first place. Documentation is another area where new contributors can often make contributions, but again these contributions need review or they can do more harm than good (and that's true for experienced contributors as well). One practice that worked pretty well as motivation a few years ago was "review exchange" where a committer would offer one patch review in return for confirmation and triage of five new issues. But I haven't seen that offer made explicitly for years. While I agree with Steven that we need to be cautious about changing the reward structure in principle, and there will also be practical issues of both workflow changes and coding new features since our mailing lists and issue tracker don't support reputations and badges, if you want to push this forward, I'd recommend being more explicit about the "successful examples" you mention, and describing best practices in this area. Steve

On Mon, Nov 21, 2016 at 06:52:06PM -0500, andreas wrote:

Hi,

I was thinking if a reward system for contributions would add motivation to those who want to contribute.

Or, it could destroy the motivation of those who want to contribute for the love of the language and a desire to give back to the community, and encourage those who don't care about Python or the community but are motivated by the selfish desire to collect points and rewards. Giving people rewards is very risky. You can just as easily encourage behaviour you didn't intend to. Before coming to your suggested solution ("Rewards! Upvotes! Downvotes! Reputation Points! Badges!") perhaps you can explain what problem you hope to solve.

Something in the lines of a reputation system.

If Python had a reputation system from the beginning, I expect that Guido's reputation score would be about -10000 by now because he won't add braces to the language.

This has been done before with success.

Oh? Where? Can you give us more detail please? -- Steve

andreas writes:

I was thinking if a reward system for contributions would add motivation to those who want to contribute.

Something in the lines of a reputation system.

I think if you look at the number of patches sitting on the tracker waiting to be reviewed, you'll see that we have plenty of patch submissions from less experienced contributors. Probably the best motivation for those contributors would be halving the time to review by committers. But new committers don't grow on trees, and they tend to be very busy people. They're generally not motivated by badges. It's true that there are jobs that less experienced contributors *can* do that are un-fun, like confirming defect reports and even triaging issues. These might be motivated by a reputation system, but they're also easy to do sloppily if done too quickly. So we might need reviewers to check them. As Steven d'A points out, it may be more straightforward to have them done by people who are motivated to put in careful hard work in the first place. Documentation is another area where new contributors can often make contributions, but again these contributions need review or they can do more harm than good (and that's true for experienced contributors as well). One practice that worked pretty well as motivation a few years ago was "review exchange" where a committer would offer one patch review in return for confirmation and triage of five new issues. But I haven't seen that offer made explicitly for years. While I agree with Steven that we need to be cautious about changing the reward structure in principle, and there will also be practical issues of both workflow changes and coding new features since our mailing lists and issue tracker don't support reputations and badges, if you want to push this forward, I'd recommend being more explicit about the "successful examples" you mention, and describing best practices in this area. Steve