[RFC] Mister open source bounties anyone?

[flag26838]

Hi there,
i've been wondering for a while if we shouldn't have a "Mister bounties program".

If you are not familiar with bounties, check the RetroArch page:

https://docs.libretro.com/development/bounties/

they even link to a website designed to handle open source bounties:

https://www.bountysource.com/

For specific components (e.g. implement a cycle accurate ARM7 core or the GBA
custom chipset), or for the initial "heavy lifting" of a system (e.g. put
together a PcEngineCD or MegaCD system, etc), i think it makes a lot of sense.

The above page says that bounties are attached to Github issues, so people can
even fund fixes for a specific problem (e.g. the MD core shows 6 pixel of garbage
with setup XYZ) and i was thinking that setting up a per-feature funding channel
might encourage third parties / hardware companies to fund FPGA work in the
open.

I know there's Patreon, but i feel like adding bounties might accelerate some
work and better reward people that, already, do a lot of the development.
What do you think?

[Newsdee]

I think it is a terrible idea. This gig economy mindset is ill suited for the long term intellectual work involved.

It turns development into a popularity contest, and attract people who wouldn't otherwise care about a system, motivated to meet the bounty but nothing else. A lot of the final fine tuning of a core is contributed by many people and this model discourages that.

And as you pointed out, there are already existing ways to tip developers.

[Sorgelig]

Those who are motivated by money usually go commercial way and doing something like Vampire or Analogue products. Definitely not open source. You can't cover that profit by bounty.

[aristogeiton]

I think this is an excellent idea! Very keen to participate.

[zomgugoff]

MiSTer is an evolving project, which involves core maintenance. A bounty may get the core made, but it wouldn't necessarily keep it working. Something like this could pave the way for a minefield of unsupported/dead cores lying around. Or even redundant cores *cough*retorarch*cough*.

[djmartins]

MiSTer is an evolving project, which involves core maintenance. A bounty may get the core made, but it wouldn't necessarily keep it working. Something like this could pave the way for a minefield of unsupported/dead cores lying around. Or even redundant cores *cough*retorarch*cough*.

Don't some devs here use Patreon and take donations already and talk with supporters and get input
on what to work on?
Seems to me that this project has an amazing amount of cores already and most of them work really well
all things considered.
If one really wanted features in a core isn't there two ways of achieving it?
Do it yourself (easier said than done) or talking with those who can do it.

[zomgugoff]

Don't some devs here use Patreon and take donations already and talk with supporters and get input
on what to work on?
Seems to me that this project has an amazing amount of cores already and most of them work really well
all things considered.
If one really wanted features in a core isn't there two ways of achieving it?
Do it yourself (easier said than done) or talking with those who can do it.

As long as you can get whoever is charging admittance to provide a "complete" product, payment isn't a bad idea. But, whatever is produced has to stay open-source, which can be a conflict of interest.

[kitrinx]

I prefer the work stay as a passion project. Bounty systems lead to all the points above and situations where you end up with irresponsible devs just adding bloat features for one guy with a couple of hundred bucks to throw at it.

[djmartins]

Founder of this project has been very generous and clear that this is an open source project.
Look how it is working?
I used to want a MiST but a few issues with it held me back from getting one.
Glad I waited for MiSer!