Sponsor An Open Source Driver Dev For a $1 a Day???

[tarceri]

Hi guys, for anyone still interested i this topic I have started a small scale campaign to do some Mesa development. In an effort to better understand if crowd sourcing Mesa development would work I have setup a small project on Indiegogo with the Goal of implementing the GL_KHR_debug extension (OpenGL 4.3). As a stretch goal I aiming to create a bit of a guide to get started with the basics of Mesa development based on my experience. Please feel free to ask me any questions, comment on my idea or in good old Phoronix fashion flame me for whatever your reasons my be.

To view my campaign just go to indiegogo and search for Mesa.

[archibald]

If we pay them on a hourly/daily/weekly/monthly basis, they have the tendency to get lazy.
Think of it this way:
"Sorry guys, I'm not really in the mood to work hard today, but who cares, I'll get paid either way."
If it is milestone based, then they will be motivated to work harder.

I understand the idea you are trying to convey, but I think it is oversimplified. It also seems to suggest that everybody should be paid on commission in case they become lazy. The AMD and Intel open source teams seem to do quite well despite being paid a regular salary - I doubt many would stick around if they didn't have a regular income: I imagine inconsistency would be quite hard when mortgages and food bills come in a regular cycle.

[bug!]

If it becomes milestone-based then count me out. I wouldn't want my employer to pay me only when software was released or bugs fixed, so I don't think it's fair to ask somebody else to be paid like that. Expenses are regular (rent, food, electricity etc), so I think the developer's remuneration should be too. Unless the idea is that this is for people who already have jobs but want some extra cash.

Aside from that, wouldn't a milestone based system discourage somebody from refactoring code as necessary, incentivising them to rush something out that meets the milestone but isn't of such high quality? "Sorry it's ugly guys, I wanted to pay my rent".

If we pay them on a hourly/daily/weekly/mothly basis, they have the tendency to get lazy.
Think of it this way:
"Sorry guys, I'm not really in the mood to work hard today, but who cares, I'll get paid either way."

If it is milestone based, then they will be motivated to work harder.

Remember, they are the ones who got paid to write it, therefore they are the ones who have to make sure their work is up to standards.

As I said, someone knowledgable of OpenGL/Mesa/Gallium3D should lead this project, as they know how long it would take to implement feature X, and how much it is "worth", including all the structural changes/bug fixes/etc.

I'll support this project either way, but milestone based is, in my opinion, better suited.

[archibald]

If it becomes milestone-based then count me out. I wouldn't want my employer to pay me only when software was released or bugs fixed, so I don't think it's fair to ask somebody else to be paid like that. Expenses are regular (rent, food, electricity etc), so I think the developer's remuneration should be too. Unless the idea is that this is for people who already have jobs but want some extra cash.

Aside from that, wouldn't a milestone based system discourage somebody from refactoring code as necessary, incentivising them to rush something out that meets the milestone but isn't of such high quality? "Sorry it's ugly guys, I wanted to pay my rent".

[bug!]

I have some spare cash on my Paypal account that wants to be put to good use .
( I hope I can pay with Paypal... )

However, I do agree with others here. We need a goal/milestone based system, because time-based systems suck.
I want to pay for the job the developer completes, not for how long he works on it.

I don't care if it takes them 1 day, or 4 weeks to complete something, but they should keep in mind that the sooner they finish, the sooner they'll get paid.

We will require someone to who understands a lot about OpenGL/Mesa/Gallium3D to lead this, as they know best where the money is currently needed.

Count me in!

[crazycheese]

Someone has to start this project.
It won't take off otherwise.

An iterative approach would be fine IMHO - it hasn't to be perfect right from the start.
But as I said, I think it is difficult to find someone who steps up and volunteers for the job.

I think this position was already taken, lookup OP...
If not, we will find someone, sure.

BTW, I'm completely ok if someone from this projects takes small $ percentage to offset his work. It just has to be corresponding and project should really roll (as in benefiting).

[entropy]

You should start collecting money to sponsor someone... waaait...

(sorry, couldn't resist ;-))

Mach' dich nur lustig!

[droste]

I think it is difficult to find someone who steps up and volunteers for the job.

You should start collecting money to sponsor someone... waaait...

(sorry, couldn't resist ;-))

[entropy]

Someone has to start this project.
It won't take off otherwise.

An iterative approach would be fine IMHO - it hasn't to be perfect right from the start.
But as I said, I think it is difficult to find someone who steps up and volunteers for the job.

[NomadDemon]

i agree with you. thts what we need to. GOAL system, not "$ per day"