Open Source Developers Mostly Pros, Not Weenies

SpinyNorman writes: "Survey shows open source developers mostly veteran pros, not slashdot weenies. Slashdot weenie Hemos should have submitted this himself already seeing as he was involved in it as LinuxWorld! Open source a needed outlet for programming pros." Like any survey, it's bound to miss some avenues of exploration, but this is the best look at a large group of open source developers I've seen yet. The survey itself (a joint project of the Boston Consulting Group and Slashdot-parent OSDN) lives at www.osdn.com/bcg, or you can jump straight to it in either PDF or html.

Comments

Interesting, but not THAT interesting

[RareHeintz]

This is cool to hear about, but I can't say it comes as that much of a surprise. Most of the pros I know really love hacking, and most of them have programming- and tinkering-related projects they pursue on their own time. That these people should make up a significant part of the open source community just seems to make sense.

Just my US$2e-02. OK,
- B

So this is where that spam came from.

[plover]

According to this link, the study was conducted by spamming the Linux Kernel mailing list and random SourceForge developers.

Is it significant that 34% of SourceForge developers responded but only 2.4% of Linux Kernel list subscribers? Does this survey prove anything more than "SourceForge developers are more likely to be successfully trolled than Linux Kernel list subscribers?"

Re:So this is where that spam came from.

[prockcore]

I actually recieved this email (I run a project on sourceforge). I didn't fill it out, but I had thought about it.

They did some good trolling.. they refered to the project by name, said that the project was related to the survey they were doing, etc etc.

As far as the 34% of SF developers versus 2.4% of the LK list... that's easy. Not everyone who reads the Linux Kernel mailing list is a developer.. but nearly everyone who runs a sourceforge project is.

Broken Links

[3ryon]

As the PDF link is broken, and the HTML link is annoying, here are some short-cuts: Motivation for programmers, Job functions, Time Spent, Feelings about OS software, Project desires

Re:Very timely study

[ergo98]

most open source zealots are damn good coders, and yes, many of them even have jobs and a house/apartment!

Everything was great until I came across this: Most open source zealots haven't written a line of code in their life. Most open source authors are pragmatic and appreciate the benefits in particular areas, and their approach is anything but zealotry. There is a vast chasm of difference between the attitudes of a Slashdot Warrior advocating the true way of enlightenment, and the people who are behind the usable open source.

Not surprising, really...

[mystery_bowler]

A friend of mine who, like me, is a professional software developer once likened programmers to guys who love their cars, they just love to show you what's under the hood.

I mean, really, it doesn't take much poking and prodding around the Internet to find dozens of programmer resource sites, most of which have tons and tons or free code. Programmers, for the most part, feel a sort of comradery (sp?) with each other and as such aren't very hesitant to share their code with other programmers. There have been many occassions where I've gotten a third-party product developer to open up the code base, despite licensing or legal issues, so I could help him/her customize the product to suite my employer's or client's needs.

Also, I believe there comes a time when a professional developer is either experienced or learned enough to understand the "how" of just about any piece of programming, even if they don't have the experise to jump in and do it. For example, I understand how 3d engines work, despite my never having coded one from the ground up (so to speak). I think opening up the code base of your neat-o project is a little easier to take when you realize that other competent developers are going to basically understand how you accomplished what you've done.

As for "weenies" not showing off their code...perhaps they don't have code of their own to show?

Re:Very timely study

[gmack]

That's true actually.. most of the flaming I see tends to be from people who don't contribute. And it's not restricted to Linux either I've seen the same in the FreeBSD camp.

Real programmers shut up and let their work speak for them.

Inherent in the process?

[PM4RK5]

I don't know about anybody else, but I have only ~8 months of programming
experience, and I feel that I'm relatively proficient. (I know there
are always people better than me, and things to learn - this isn't a flame/troll).

But my point is, I learned most of this stuff myself, and I can't help but
wonder if part of the reason I grasped some of these concepts faster is that
I was able to actually examine Open Source code multiple times to learn how
things were done. So in that way, I was learning from professional programmers,
which in turn, could possibly (most likely) generate more professional
programmers, which due to their background of learning from OSS will probably
also contribute to the OSS community, and some other new programmer will start
the same way.

(Self-fulfilling prophecy?)

Re:Inherent in the process?

[TheAJofOZ]

So in that way, I was learning from professional programmers, which in turn, could possibly (most likely) generate more professional programmers,

Not quite - you are learning from the code of professional programmers, not the programmers themselves. The big difference doesn't show up in your code but in your documentation, design and process skills. Initial coding is only a very minor part of the battle of creating great software, the rest is in the design and maintenance and looking at code will teach you very little about documentation requirements.

You can certainly pick up some aspects of design from code directly, but don't fool yourself into thinking you are competent at design if you have only learnt from code - it really is an artform. Probably the biggest failing of open source software is in it's documentation (ie: there is extremely little and documentation is almost always behind). I am definitely not one to support producing copious amounts of wasted paper but I am well aware of how much difference a solid design, fully planned before any code is written, can benefit productivity in the long run.

So by all means participate in open source development and learn from the code, you will learn vast amounts about code that way but don't stop there. Go out and get a degree in software engineering (or something else that focusses on design and maintenance since you already know how to code well), read as many books and white papers on software design as you can or better yet, do both (and whatever else you can find).

I know there are always people better than me, and things to learn

That's the spirit! There is some really cool stuff coming out in white papers these days both relating to code and design - keep an eye out for Genetic Software Engineering from the Software Quality Institute they're doing some really cool stuff.

Maybe if we all go out and study up on design and management (yes, yes, but it's important even in opensource) the next survey will show that open source developers are brilliant at code, design and make the best managers.... Or maybe that's pushing it.....

this is because the /. doesn't do anything

[Frothy Walrus]

You have to admit, the current generation of computer users, raised on the Web and AIM, isn't of too much use for a lot of things. Years of passive entertainment has dulled their senses, and they would rather the computers played with them than they play with the computers.

There are so many users who are *lost* when the network is down... it's like they don't know what to do with a computer. The past generation knew how to tinker because it was the entertainment... now the spirit's being lossed. It's a good thing the old-timers are so tenacious.

Here's the correct link

[jon_c]

http://www.osdn.com/bcg/BCGHACKERSURVEY.pdf

-Jon

There are lies, damned lies...

[dave-fu]

...and statistics coming from a not-even-remotely-detached source.
Don't feel like calling into question the verity of the methodology employed in a survey of open source developers by a open source company; it's obviously as trustworthy as a survey of databases as performed/sponsored by Oracle.
I'm not even sure what a sweeping generalization like this proves other than some of us really like to program.

Professional, in there on opinions

[_wintermute]

Wow, 9/10 Open Source developers are personally convinced that they are 'professionals' who know more than everyone else.

This is news why?

Isn't this obvious

[cs668]

How many times do you get to work on something really creative at your day job?

Most work is either database, accounting, or doing web apps with some really broken methodology/tool.

The most fun I have had is working on non-work realated things just for the sake of writing some fun code.

Report may be true

[ahde]

but, if you go back 5 years or more, you will find a completely different demographic.

Most open source developers started in college or before. The maturity level in open source represents the *maturing* of open source. No one has replaced Linus, but there are newcomers like Marcello. Many other projects are still spearheaded by the same people that started them in their basements when they were teenagers. As far as I know, there hasn't been a massive UV radiation die-off, or kidnapping/cyborg replacement program of key developers

a measly 2%

[asparagirl]

I was one of the measly 2% of the survey respondents who was female. (Yes, guys, geek girls do exist- now wipe the slobber off your screen and move along please, nothing to see here.) But where oh where are all the other female open source folks? I mean, I know female CS types are rare, but are they rarer still in the open source world? If so, why?

I'd also be curious to know which *types* of open source projects seem to attract larger-than-usual numbers of women to the contributing ranks. From personal experience, I can say that Post-Nuke, an open source fork of PHP-Nuke, has a fairly high percentage of female participants/developers. But why? Maybe because it's modular and women can go off and work on a module on their own and then go back and submit it later, and thus feel less pressure to need to know all of the ins and outs of every little system. Or maybe it's because it's tempting to play around with modifying the themes/skins (not to get all stereotypical about women being attracted to the GUI end of things while guys do the coding of the guts of the program, but some stereotypes have a good basis in reality). Or maybe because it's a content management system, and women have long had a foothold with online personal publishing or personal communication systems (blogs, online diaries, IM'ing, plain ole personal home pages having long had a sizeable female early adopter contingent- geez, maybe we womenfolk will only pick up a technology if we can use it to *talk*?).

[sigh]

Re:a measly 2%

[mccalli]

I have yet to meet a woman who I would consider a hardcore hacker.

It's been my luck to know a couple. However, one of the funniest things I remember was a rather patronising social experiment, done in a psychology course for the Open University. I caught this programme on television - I wasn't part of the course. It's all quite a few years ago now as well - maybe 90/91? Don't know for sure.

The experiment gave an internet connection, via modem, to three women - one in her early twenties and a member of the women's darts team, one a working professional single mother in her mid-forties, and the final one looked like everybody's favourite grandmother.

The woman in her twenties discovered internet chat rooms (yes, plenty were there then. Anyone remember Cheeseplant's House?). The woman in her forties spent time with her child doing educational things. Next came the grandmother.

Of course, everyone expected her to have used the machine as a tea-cosy or something, so it came as rather a shock to find she had been participating in various freeware projects, running technical simulators and tweaking her connection parameters to get better throughput. You could feel the researcher slipping into shock...

Completely without knowing, the team had accidently picked one of the original Colossus team members, and she was putting her sudden luck to good use...

Cheers,
Ian

Well

[MisterBlister]

That doesn't surprise me much.

It's the pros who do the actual work. It's the weenies who sit around and bash Microsoft while pontificating and arguing the subtleties of the GPL vs. whatever or Linux vs. BSD on Slashdot all day.

Most interesting number

[Salamander]

I found the most interesting numbers to be on Page 37. There, 19% of respondents admitted that they were stealing time from their employers to write open-source software. Would anyone like to bet on what fraction of the 46% who answered "do not participate at work" were telling the truth, the whole truth, and nothing but the truth? What fraction of the "part of core job" respondents were in the "not known by supervisor" category at some point in their careers?

There are a lot of people who work on open-source software in their spare time. There are quite a few who get paid to do it. Bless them all. However, these numbers seem to indicate that at least 19% and probably much more than a third are regularly working on open-source projects while they're being paid to do something else. Maybe it's time to question whether the equation "open source == moral high ground" has any validity.

Re:Most interesting number

[lkaos]

I've been told by many managers that it is good to expect 33% true productivity from employees in a cube farm environment. People make personal calls, answer email, talk by the water cooler, etc.

Now, a good manager recongizes that this is ok. It's good for moral. If the biggest compliant of an FS programmer is that he spends some time at the office hacking away at a bug or scribbling some code on a notepad while the other goobers are talking about what they saw on Survivor, then that's a damn good thing because atleast the FS programmer is becoming a better programmer and will likely become even more productive.

Besides, since most FS programmers have a bit of experience (according to the article), it is likely to assume that they have enough experience to know how much they can get away with doing at work without work suffering.

Think of it this way, if your a manager would you rather see your C++ programmers spending an hour talking about the Super Bowl or spending an hour figuring out a non-work related bug in a C++ program that is FS and could therefore be used by the company in some way in the future?