So, I've posted on here about looking for open source projects to work with before, and I've worked with Mozilla in the past and Chromium currently. I've almost always been embraced with open arms.
Open Source projects are normally just pumped to get the help. Even Chromium, where the vast majority of developers are Google employees, was really helpful in getting us online and moving. They love outside engagement.
And I've learned more about writing real software in my brief efforts with bother Chromium and Mozilla than I ever did at my Cisco internship. I think it comes down to this: when they're paying you to be there, there's no burden on them to somehow educate you. You're there to work. On the flip side, open source projects see helping you as an investment in their future.
He's not saying "Open source doesn't like the help". He's saying "more people end up working in lead in activities to software development in commercial settings than in open source settings".
He's not saying "OSS is bad". He's saying "OSS for whatever reason is 99% white and asian males, so if that becomes a large place for introducing people to programming, then we've got troubling diversity issues upcoming"
One of Dr. Guzdial's big pushes has always been computing "inclusiveness", that being, getting everyone interested in computing. Here is a comment buried on one of the blogs in question by Guzdial:
>Kirby, I'm not claiming that open source doesn't desire LPP. It simply isn't happening for people not White or Asian or male. Whatever is happening to draw in a more diverse crowd, to get a more diverse group involved in the periphery -- it isn't working. There are lots of incentives for women and minorities to get involved in commercial software development, but that 24% number is pretty paltry. That's why efforts like BPC and NCWIT exist, to try to figure out how to go beyond those incentives to get the diverse workforce that will lead to better and more innovative designs.
The ACM (one of Mark Guzdial's associations) doesn't even believe in non-paywall academic publication. So it isn't too surprising he doesn't like seeing software published either.
The man for years has been building up the open source smalltalk (Squeak) that lots of things are being built on these days.
He's saying when you look at who's participating, company based software development is empirically including more people in more "lead in" activities to software development than open source is.
That's all.
And the fact you expect a tenured professor to not belong to the ACM is laughable. MG is the director of undergraduate studies at GT, and the ACM does plenty of work with and for the school.
A good point- I wasn't at all fair. Now that I have thought about it a bit, I probably am in closer agreement. I very much like having open source software (and the source to such software around)- but I don't think I could in good conscience encourage students to subject themselves to the procedures and politics often inside such projects.
I don't expect him to not be ACM- but I would like more push from chairs to get ACMs policies to a more open position (such as making more material available on the web).
Mark Guzdial responds in the comments, and it's worth reading. He says that his comments were related to the demographics of open-source development, and in that context I don't really disagree--although I'm by no means convinced that the problem is open-source. I think it's more a symptom than anything.
(Also, I'm a student at Wesleyan University and I've been very involved with the Humanitarian FOSS Project since the summer and will be at the HFOSS Symposium in March. If anyone has any related questions, I'm happy to answer them.)
In companies, developers are forced to work together -- in good and in bad. This gets stuff done but it also pushes for questionable code quality as personal conflicts remain unresolved and that takes time from doing good work.
In open source world, meaningful contributions only happen when the contributor and the project maintainer connect on a certain level. Nobody wants to work with people they consider jerks if they don't have to. And this is not a fixed set either: differences in opinions can arise and get settled later and a contributor--maintainer connection is respectively broken or (re)established.
Open Source projects are normally just pumped to get the help. Even Chromium, where the vast majority of developers are Google employees, was really helpful in getting us online and moving. They love outside engagement.
And I've learned more about writing real software in my brief efforts with bother Chromium and Mozilla than I ever did at my Cisco internship. I think it comes down to this: when they're paying you to be there, there's no burden on them to somehow educate you. You're there to work. On the flip side, open source projects see helping you as an investment in their future.