3 ms·

I wanted to contribute to my first OS project. I am an hobbist programmer, but I do have some good skills in python/flask. As I said I was still learning and on

by kfk 13y ago

I wanted to contribute to my first OS project. I am an hobbist programmer, but I do have some good skills in python/flask. As I said I was still learning and once I pointed the project coordinator to my half-assed github, he did not answer me back anymore. Point is: I can work on documentation and bugs, but if you behave like an asshole, how can I even do that? People are not contributing much to OS because many programmers reject the idea of mentoring. But think about it: it gives you more 1 hr of mentoring and then 10 hours of work of an hobbist programmer or 1 hr more of programming?

keithpeter 13y ago
Could you find a project with some form of mentoring happening on a forum or mailing list? Some projects have sort of 'funnels' where people get small tasks to do then progress to larger chunks of stuff. E.G. the Ubuntu manual project at http://ubuntu-manual.org/ http://ubuntu-manual.org/ looks like that on their mailing list... So basically avoid the less responsive projects.
daleharvey 13y ago
Somewhat of a Hanlon's razor, a lot of open source projects arent maintained, they are thrown over the wall, similiarly a lot of project maintainers are very busy and unless they get a pull request then they are quite likely to ignore it, as projects gets popular it easily gets the point where maintainers have 'no' time to code, and spend their time guiding contributions. Ultimately if a maintainer isnt showing interest in taking your contributions then it just isnt a project you want to carry on attempting to, it takes some a little time and maybe experience to find a suitable project. Certainly dont make sweeping statements about nobody contributing to open source because programmers dont like helping each other, there may be exceptions but the people contributing to OS is growing faster than ever
mkramlich 13y ago
nobody should be an asshole agreed but sometimes the net cost to the senior/mentor person is actually negative, if the other party, the newb is sufficiently junior or incompetent. there are plenty of cases where somebody delivers one "fix" or "improvement" which then itself introduces other problems. save 2 hours, cause other people an additional >2 hours of work afterward, sometimes multiple of that. (time to add a bug? a few seconds. time to discover/home-in/fix/test that bug later? hours/days/weeks.) that's... not good. and experienced folks try to minimize it. the nice thing about the Internet is how we can all talk and learn and collaborate together more easily now than ever before. the bad thing about the Internet is folks who are still but wee infants, skill-wise, who, say 20 years ago would have had a long gestational period working in private and relative isolation, on their boxes before they emerged into the larger ecosystem (at a job or university, etc.) are instead often almost immediately jumping into forums, channels and repos and trying to "contribute" while they're still in that gestational stage. It's like there's now an army of cute fuzzy wittle lion cubbies who now try to tag along with the adult lions when the latter slink off to hunt for prey -- it impairs both parties and does not actually "help" either.
- kfk 13y ago
  Let's not get offensive, OK? I have been programming in University, learning on my own, both R and Python (and SQL, Javascript, HTML, CSS). I played in Python for 3 years. I am not a noob, I know what I am doing. My "problem" is a half assed github. And you know why is that? Not because of my bad programming, but because of lack of motivation. Really, we are not all ninjas here coming home from a 9-5 and then coding on side projects till 1am. Many people lack that kind of motivation and that is why we need people willing to mentor or at least being friendly. Finally, if a push from a total stranger can mess up your code base, I wonder if you know what you are doing. We do test driven development for a reason. P.s. I am giving you an up 1 so I make sure you read this.
  - mkramlich 13y ago
    I spoke in the general case based on decades of first-hand observations in RL. There are obviously always exceptions to any general pattern or effect. Also, not all projects have automated tests. Not all those with automated tests have 100% perfect coverage of everything that could possibly break. If anything I'd argue that the rise of unit test culture and TDD culture has actually increased the incidence of allowed sloppiness in some cases by folks having the mindset of, "I can deliver any diff I want. If no existing test failed, or the project has no tests, I did nothing wrong." This mindset too I've seen evidence of repeatedly. And there's nothing wrong with a person being new to programming. I can only reiterate the core point of my original comment to which you replied.

by kfk 13y ago

keithpeter 13y ago

Could you find a project with some form of mentoring happening on a forum or mailing list? Some projects have sort of 'funnels' where people get small tasks to do then progress to larger chunks of stuff. E.G. the Ubuntu manual project at http://ubuntu-manual.org/ http://ubuntu-manual.org/ looks like that on their mailing list... So basically avoid the less responsive projects.

daleharvey 13y ago

Somewhat of a Hanlon's razor, a lot of open source projects arent maintained, they are thrown over the wall, similiarly a lot of project maintainers are very busy and unless they get a pull request then they are quite likely to ignore it, as projects gets popular it easily gets the point where maintainers have 'no' time to code, and spend their time guiding contributions. Ultimately if a maintainer isnt showing interest in taking your contributions then it just isnt a project you want to carry on attempting to, it takes some a little time and maybe experience to find a suitable project. Certainly dont make sweeping statements about nobody contributing to open source because programmers dont like helping each other, there may be exceptions but the people contributing to OS is growing faster than ever

mkramlich 13y ago

nobody should be an asshole agreed but sometimes the net cost to the senior/mentor person is actually negative, if the other party, the newb is sufficiently junior or incompetent. there are plenty of cases where somebody delivers one "fix" or "improvement" which then itself introduces other problems. save 2 hours, cause other people an additional >2 hours of work afterward, sometimes multiple of that. (time to add a bug? a few seconds. time to discover/home-in/fix/test that bug later? hours/days/weeks.) that's... not good. and experienced folks try to minimize it. the nice thing about the Internet is how we can all talk and learn and collaborate together more easily now than ever before. the bad thing about the Internet is folks who are still but wee infants, skill-wise, who, say 20 years ago would have had a long gestational period working in private and relative isolation, on their boxes before they emerged into the larger ecosystem (at a job or university, etc.) are instead often almost immediately jumping into forums, channels and repos and trying to "contribute" while they're still in that gestational stage. It's like there's now an army of cute fuzzy wittle lion cubbies who now try to tag along with the adult lions when the latter slink off to hunt for prey -- it impairs both parties and does not actually "help" either.

kfk 13y ago
Let's not get offensive, OK? I have been programming in University, learning on my own, both R and Python (and SQL, Javascript, HTML, CSS). I played in Python for 3 years. I am not a noob, I know what I am doing. My "problem" is a half assed github. And you know why is that? Not because of my bad programming, but because of lack of motivation. Really, we are not all ninjas here coming home from a 9-5 and then coding on side projects till 1am. Many people lack that kind of motivation and that is why we need people willing to mentor or at least being friendly. Finally, if a push from a total stranger can mess up your code base, I wonder if you know what you are doing. We do test driven development for a reason. P.s. I am giving you an up 1 so I make sure you read this.
- mkramlich 13y ago
  I spoke in the general case based on decades of first-hand observations in RL. There are obviously always exceptions to any general pattern or effect. Also, not all projects have automated tests. Not all those with automated tests have 100% perfect coverage of everything that could possibly break. If anything I'd argue that the rise of unit test culture and TDD culture has actually increased the incidence of allowed sloppiness in some cases by folks having the mindset of, "I can deliver any diff I want. If no existing test failed, or the project has no tests, I did nothing wrong." This mindset too I've seen evidence of repeatedly. And there's nothing wrong with a person being new to programming. I can only reiterate the core point of my original comment to which you replied.