Slashdot Mirror
Open Source Limitations?
_aargh writes "This ZDNet article by John Carroll makes the claim that open source is flawed because there isn't a way for programmers to earn money by developing open source software. It annoyed me so much that I wrote this response to it on the O'Reilly Network."
And pretty much discounted it after I thought for 2 seconds.
Profit is not the only reason why people use Open Source (aka - me).
We use it/develop for it/fund development for it because we get more stable software. More secure software. Software that has less bloat, and is less likely to lock me into a vendor that might turn into an asshole later (like trying to get me to pay for my software every year...or else).
If you look in the narrow world of profit - then he's still only partly right. How many resumes get you into a great job saying "I developed the XYZ patch for Imagemagik/Linux kernal/Mozilla - and that proves my kick ass ability, so you should hire me if you want that same ability in your employees". Last time I checked, being able to prove your 1337 hax()r skills usually gets you a good job - and good jobs == money, and money == financial incentive.
Of course, I could be wrong.
52 Weeks, 52 Religions with John Hummel
Dear John: Regarding "The very real limitations of open source"
by Derek Vadala
Jun. 7, 2002
URL: http://zdnet.com.com/2100-1107-933913.html
Mr. Carroll makes a lot of mistakes in his article The Very Real Limitations of Open Source", posted earlier today over at ZDNet, but perhaps the most ignorant one is his statement that "open source must rely on the willingness of programmers to contribute code without financial compensation". This is most assuredly not the case.
First take the many corporations that help subsidize open source projects by paying programmers to work on open source software. Look at the many developers employed by companies like IBM, SGI, Quantum, and RedHat to name just a few. As open source becomes more pervasive, corporations like the ones I've mentioned will continue to hire open source developers because having them on staff allows these companies to better compete for very lucrative service and support contracts. After all, it's more likely that I'll hire IBM to help my organization migrate to JFS because Steve Best works there.
Or, look at the success of Marty Roesch (Snort) who landed $2 million worth of venture capital last March, in the middle of a declining economy. Marty's company Sourcefire builds on top of his open source project by offering consulting services and developing better management tools. While not every OSS developer can secure that kind of venture capital, that shouldn't stop anyone from following the same model by supplementing OSS projects with consulting and value-added services.
Carroll goes on to say "Proprietary software will always generate more revenue than free, open source software". The problem with this assertion is that it seems to only take revenue from software sales into account. It's been clear for quite some time that open source software uses revenue generated through support and supplemental services to compensate for the fact that the initial purchase is either minimal or at no cost. I'm sure Carroll and others would quickly point out that proprietary software can generate income through both software sales as well as support, and while this is true, that argument doesn't take into account the enormous (often bloated) development costs for proprietary systems and software.
I think the real problem with Carroll's argument is that he bases it not on the real world application and adoption of open source software development by both big business, governments and individuals, but instead on the philosophy of the Free Software Foundation. In essence he ignores the fact that many in the open source community are not in total agreement with the outlook of the FSF, he assumes that the motivations of all open source developers are equal, and he ignores the business models that have been built around open source software.
Derek Vadala is a consultant living in New York City, specializing in Linux and Network Security, and the author of O'Reilly's upcoming book, "Managing RAID on Linux."
oreillynet.com Copyright © 2000 O'Reilly & Associates, Inc.
My STB PCI tv tuner card. Windows 2k drivers? No. Linux drivers? Yes.
My Umax scsi flatbed scanner? Windows 2k drivers? No. Linux? Yes.
Generic external 36k external modem? Windows 2k drivers? Could only ever see it as 28.8k max, and rarely got that high. Linux? Maxed it out, and could care less what brand it was.
Remember, this is on a single computer of mine, not every half-assed example I could think up that I've heard about but never personally experienced. But for the finale, same computer...
Best Windows 2000 uptime I've ever seen? 30 some days, died on a random spontaneous BSOD.
Worst Linux uptime? 30some days... forgot to compile in a rarely used kernel option, and I like to play with that kind of stuff.
Oh, and stop lumping linux in with KDE/Gnome. Those are projects designed to imitate windows' instability... of course they crash. I use wmaker with a screen full of dockapps and rxvt's. Even on a computer this old, its just damn... *zippy*. Once I get rid of Netscape 4, and get around to compiling Mozilla, I will truly have the perfect desktop that I've always lusted after.