| 1 |
On 3/20/06, Daniel Drake <dsd@g.o> wrote: |
| 2 |
> "more open"? I can't think of a decent way to phrase the subject line |
| 3 |
> which might make it sound it was coming from a native English |
| 4 |
> speaker..ahem..anyway: |
| 5 |
> |
| 6 |
> I read a complimentary comment from a Gentoo user recently (can't |
| 7 |
> remember exactly where, so this is from memory). It was something along |
| 8 |
> the lines of "Gentoo is great and will continue to be great for the |
| 9 |
> foreseeable future. You have built the required structure to keep up |
| 10 |
> with the rate of change in your environment (i.e. the increasingly rapid |
| 11 |
> rate of development of open-source sofware)." |
| 12 |
> (if anyone can point me to where I read that, I'd appreciated it). |
| 13 |
> |
| 14 |
> I think there's a lot of truth in that, especially the way that he/she |
| 15 |
> highlights the fact that simply keeping up with what goes on around us |
| 16 |
> is key to our "survival". |
| 17 |
> |
| 18 |
> I won't go as far to say that I *don't* think we can keep up with our |
| 19 |
> current "system", but I think there is plenty of room for improvement. |
| 20 |
> |
| 21 |
> One of the bigger problems is that we have a huge user community who are |
| 22 |
> keen on contributing, but we have such a high barrier for entry to the |
| 23 |
> developer community. Quite rightly so - we're dealing with a live tree, |
| 24 |
> so we can't give out commit access on the street. |
| 25 |
> |
| 26 |
> At the same time, I feel that we're missing out. Comparing Gentoo with |
| 27 |
> some other large open-source communities that I am personally involved |
| 28 |
> in, I feel that we're too closed. |
| 29 |
> |
| 30 |
> A developer recently compared Gentoo dev-ship to a marriage. In an ideal |
| 31 |
> world, sure, we'd love for every single person who makes any kind of |
| 32 |
> contribution to the project to become a full-time contributor who never |
| 33 |
> goes AWOL or sleeps with another project. But more realistically, I |
| 34 |
> think we need to become more open and flexible - as volunteers, people's |
| 35 |
> interests change, some people will stop contributing after they have |
| 36 |
> fixed whatever problem motivated them to contribute, etc. How can we |
| 37 |
> handle this better? |
| 38 |
> |
| 39 |
> We have a large expense on both sides when adding a developer to the |
| 40 |
> project. I personally have lost developer candidates, undoubtedly more |
| 41 |
> technically experienced than myself, who simply did not have the time to |
| 42 |
> go through a month-long recruitment process which involved studying |
| 43 |
> various documents not even relevant to the small area they would be |
| 44 |
> contributing to. On the other side, it's a fair expense to add a |
| 45 |
> developer to the project due to all of the |
| 46 |
> quizzing/assessing/account-creating/access-elevation/... |
| 47 |
> |
| 48 |
> Additionally, a significant percentage of developers who have joined |
| 49 |
> recently have gone AWOL after a few months. That hurts us, given the |
| 50 |
> expense we went through recruiting and adding them, and the time needed |
| 51 |
> to reverse that and retire them. |
| 52 |
> |
| 53 |
> I am not claiming this is an easy problem to solve - we do need to be |
| 54 |
> especially careful that any changes made do not decrease the quality of |
| 55 |
> commits to the live portage tree. This is why I am asking for help. |
| 56 |
> |
| 57 |
> I'm looking for ideas - preferably big, drastic, shiny ones. Ignore any |
| 58 |
> issues relating to migration away from our current system. What would be |
| 59 |
> the _ideal_ way for Gentoo to handle contributions from anyone? (note |
| 60 |
> that I'm dropping the user/developer community separation in that |
| 61 |
> question, as the boundary between those could change in these ideas) |
| 62 |
> How would an ideal recruitment process work, if there would be one at all? |
| 63 |
> |
| 64 |
> Please try and keep replies on-topic - I'm not trying to start a |
| 65 |
> discussion/flamewar on the current recruitment system or anything like that. |
| 66 |
> |
| 67 |
> To get you thinking, I suggest reading the section titled "Open |
| 68 |
> Development Team" at |
| 69 |
> http://www.samspublishing.com/articles/article.asp?p=23200&seqNum=3 |
| 70 |
> which is part of a (very good) larger article detailing why Linux kernel |
| 71 |
> development works so well. |
| 72 |
> |
| 73 |
> Any ideas? |
| 74 |
|
| 75 |
perhaps having some proxys of a sort that accept patchs and such |
| 76 |
from trusted users that would commit fixes to portage would help. |
| 77 |
similiar to the kernel format that way users can 'commit'/help out quickly |
| 78 |
without having to go thru the long process of becoming a dev |
| 79 |
|
| 80 |
|
| 81 |
> Daniel |
| 82 |
> -- |
| 83 |
> gentoo-dev@g.o mailing list |
| 84 |
> |
| 85 |
> |
| 86 |
|
| 87 |
-- |
| 88 |
gentoo-dev@g.o mailing list |
Archives