How to deal with a partial PR?

[kelpisland]

From @sintaxi on January 22, 2016 0:0

In my experience dealing with PRs on my open source projects is that the majority of PRs will be partially implemented/working. I'd be curious how this could be dealt with in respect to the bounty. Ideally, we find a way for people to work together on solutions rather than duplicating efforts. Thoughts?

Copied from original issue: bcgov/BC-Policy-Framework-For-GitHub#27

[kelpisland]

From @howardroark on January 24, 2016 18:39

Yeah it seems like the $1000 bounty concept depends on completeness. Ostensibly, completeness is sort of an ideal that may not be truly achievable. What would nice is if the $1000 could be "allocated" somehow and paid out as completeness is worked towards.

Perhaps you could have a top level pool of funds that has a two way relationship with pull requests as they grow and shrink in potential... vs tying up funds in endless semantic debates. If this is a government thing, the politics of perception carry a lot of weight. If you can decentralize the allocation process you would distance yourself from politics.

(Crypto currencies may favour an evolving strategy much more, and be more welcoming towards developers from the developing world. You can loose a lot of money through international remittence. Also they are less "real" in the eyes of political outsiders and may be more suitable for experimentation.)

[kelpisland]

This is a great discussion! I want say that the intention of the $1000 pay for pull is to build community and reward people for their effort. I like the notion of a "working towards completeness" model to encourage collaboration, what do you think would a reasonable way to allocate/split a bounty between collaborators?

I'd like to move this issue to our bcdevexchange/rapid-adoption repo so it is linked to place where we are building the pay for pull wiki. Any objections?

[bmann]

@kelpisland "build community and reward people for their effort" <-- so, minus the reward part, you just described open source. $1000 doesn't build community -- that part is separate. What we are doing here -- jointly discussing issues -- is in part community building.

If I had spare cycles and was a better WP / JavaScript dev, this is the kind of thing I would do "for no pay". Especially when it's pretty mechanical -- you've told me exactly what you want built, which will take some puzzling out and knowledge of REST / AJAX calls, and of course probably infinite browser bug testing :)