X.Org EVoC To Be Tightened Up, Limited To Existing Contributors

[phoronix]

Phoronix: X.Org EVoC To Be Tightened Up, Limited To Existing Contributors

To complement Google's Summer of Code, the X.Org Foundation has long held the Endless Vacation of Code (EVoC) as a year-round, passive event where the foundation would fund students to get involved with X.Org/Wayland/Mesa projects. Sadly, however, it's not been panning out and the rules are being tightened up even more due to past failures...

http://www.phoronix.com/scan.php?page=news_item&px=X.Org-EVoC-Tightening-Up

[schmidtbag]

I understand their situation but I don't necessarily agree with it. If what they're saying is true then it's pretty irresponsible of them to think they could have handled something like this in the first place.

I don't know how they do things over there. If they're basically just saying "hey if you're a student looking for some easy cash, find something to improve X/Wayland/Mesa" and basically just give money to anyone who puts in an effort then yeah, this isn't going to work. The student should either be given a todo list that he/she can take a stab at, or, should come up with a new idea and discuss it with the X.Org Foundation. After they come to an agreement, the students should only get paid after completing certain milestones. That way, no money is wasted, the students work on their own schedule, there's an incentive to get something done, and if the project is left incomplete then someone can pick up where the student left off.

[robclark]

I understand their situation but I don't necessarily agree with it. If what they're saying is true then it's pretty irresponsible of them to think they could have handled something like this in the first place.

no, what would be irresponsible is not looking back and evaluating how well things have worked, and making adjustments accordingly.. which is exactly what happened.

I don't know how they do things over there. If they're basically just saying "hey if you're a student looking for some easy cash, find something to improve X/Wayland/Mesa" and basically just give money to anyone who puts in an effort then yeah, this isn't going to work. The student should either be given a todo list that he/she can take a stab at, or, should come up with a new idea and discuss it with the X.Org Foundation. After they come to an agreement, the students should only get paid after completing certain milestones. That way, no money is wasted, the students work on their own schedule, there's an incentive to get something done, and if the project is left incomplete then someone can pick up where the student left off.

Except that, as with GSoC, there is an initial payment. GSoC has a big enough purse that they can afford to waste a few initial payments here/there when some project doesn't pan out. We don't. Also, our pool of potential mentors tends to be stretched pretty thin already. If this weren't the case, do you think it would have taken this long to get gl4?

[schmidtbag]

no, what would be irresponsible is not looking back and evaluating how well things have worked, and making adjustments accordingly.. which is exactly what happened.

Considering the circumstances, I agree. But my point is with a little more thought put into it, the results X.org got could have been predicted and therefore prevented if a different approach was taken. I'm not going to pretend I know how everything was done though, so I could be wrong.

Except that, as with GSoC, there is an initial payment. GSoC has a big enough purse that they can afford to waste a few initial payments here/there when some project doesn't pan out. We don't.

That's what I mean though - why was an initial payment required? Like I said, pay the students when they achieve milestones and it's a win-win - you don't risk throwing away money and if you do have to pay them, it was worth it. If the students ask for payment in advance, then I'm sure some agreement could be made.

Also, our pool of potential mentors tends to be stretched pretty thin already. If this weren't the case, do you think it would have taken this long to get gl4?

I'm sure the organization always knew about the lack of mentors. So unless this wasn't always a problem, this is something that should have been realized before beginning EVoC. Anyway, I never had high expectations of students completing GL4 code. Intel has more [experienced] developers, disposable income, disposable hardware, and better documentation and they're still taking their sweet time implementing stuff that has already been added to mesa, so I can't imagine this is a simple task.

I'm not saying that X.org should continue doing EVoC, because I understand the limitations. I just think this situation could have been avoided.

[robclark]

That's what I mean though - why was an initial payment required? Like I said, pay the students when they achieve milestones and it's a win-win - you don't risk throwing away money and if you do have to pay them, it was worth it. If the students ask for payment in advance, then I'm sure some agreement could be made.

well, that is more or less amounts to the same thing as what we are doing.. no one said the candidate needed to be a long-time contributer to whichever project. Consider the initial contributions as being the same thing as the pre-first-payment ramp-up period, just without a specific timeline.

I'm sure the organization always knew about the lack of mentors. So unless this wasn't always a problem, this is something that should have been realized before beginning EVoC. Anyway, I never had high expectations of students completing GL4 code. Intel has more [experienced] developers, disposable income, disposable hardware, and better documentation and they're still taking their sweet time implementing stuff that has already been added to mesa, so I can't imagine this is a simple task.

keep in mind that the intel team is merely *less* outnumbered than their closed source counterparts ;-)