APLawrence.com -  Resources for Unix and Linux Systems, Bloggers and the self-employed

How to make money with Open Source

© September 2004 Tony Lawrence
Sun Sep 5 22:49:04 2004 How to make Referencing: https://groups.google.com/groups?hl=en&lr=&ie=UTF-8&threadm=k5ghj0lg8hlnrfko562vbkgfvoj4aaohef%404ax.com

In the thread referenced above, Jeff Liebermann says:

I kept asking the same question: "How am I suppose to make money with Linux"? and getting no intelligible answers.

He points this out as an indication of the mistakes and bad decisions that brought SCO to the point where (Jeff again)

I'm not sure it could have been avoided. I firmly believe that SCO would have done a Chapter 11 last year were it not for the stock price jump, investments, and licensing.

Jeff may well be right. SCO made a lot of bad choices, and until this very year, did nothing to try to recover from their foolishness in trying to kill their own best product. As Jeff said

Meanwhile, OpenDesktop was selling and supporting the company. Enormous efforts were expended in arm twisting users to switch from ODT to Open Unix (Unixware). ODT continued to outsell Open Unix. So, they switch to shoving them in the direction of Caldera Linux. That didn't work for a long list of reasons that I don't wanna get diverted discussing. Eventually, Caldera got the clue, changed the name back to SCO, redirected additional development efforts toward what was bringing in the revenue, and hoped for the best. It was the right move, but far too late. The attempts to kill off ODT had finally succeeded, more by neglect than intent, and projections looked dismal.

Yet other people say you can make money with Open Source. SCO absolutely could have taken existing Open Source projects, ported them to OSR5, bundled them into useful applications and services, and thereby given the world a reason to buy its OS. What they never saw that they were in a far sweeter position than RedHat or anyone else trying to push a strictly Linux based solution.

Apps made SCO successful once. Open Source apps, where most of the work has already been done by someone else, are a wonderful opportunity. There has been very little integration of Open Source programs into "application suites" - packaged mail servers, web servers, whatever. What integration that has been done has been on Linux (and perhaps BSD) as the base OS. But what stops you from doing it on your own OS? Well, most of us don't have our own OS, do we? Sun does, HP does, and so does SCO, but it's a pretty small crowd.

Build a great integrated app on OSR5 and you have bait for the OS you want to sell. The app can be 100% Open Source, and the beautiful part is that you can tie its performance strongly to your own proprietary OS, thereby making it more difficult for anyone to take your work and put it back on Linux. The quirkiness of your OS now becomes an asset, not a liability. Just as it is difficult now to compile many Linux apps on SCO, it would be difficult for Linux people to back-port the SCO modifications! Naturally, you'd write your code with just that in mind: maximize for the strengths you have, obfuscate wherever you can, and build a package that works better on your OS than it does elsewhere. Yes, someone else CAN perhaps take your work and bring it back to Linux, but remember that you have profit coming from the OS sales, and they don't. The best they can hope for is to sell support, and you can sell that in addition to the price you charge for the OS! You should have more money for development, so you should be able to stay far ahead of the people using Linux.

What it takes, of course, is imagination and effort. SCO has a fledging Mail Server now. That's the right idea, but it should have been built with Open Source components and everything developed in-house donated right back - of course stuffed with calls to their proprietary Unix kernel. Nothing in the GPL says that you have to make it easy for your code to compile on another OS - you don't even have to provide a Make file for any other OS - as anyone who has ported Linux apps to SCO well knows.

There is the necessity here that you do it better than your competitors. As https://www.sys-con.com/java/46131.cfm says, you need an "editorial style" that makes folks want your version. If you can't do that, at least well enough to attract whatever base you need, then you lose. But if you can do it, your proprietary OS suddenly becomes an asset again. The bundled apps are the razor, the OS is the blades.

It's probably too late now. And the nay-sayers will pout that existing SCO installations would take advantage of the free apps without paying anything. Of course they would, and that would be good for business, because it would jump start your customer base and also keep upgrades coming down the line. If the product is good, other customers would come, and for the first time in many years SCO might see entirely new sales sources.

If I had been in charge of SCO, this is the path I would have taken. I might still have failed - Jeff could be right - but I would have gone out with honor and pride.

This is the direction Sun and HP need to take. Strangely, it's also something Microsoft could do, with even more success than anyone else. That's an awful thought, isn't it?

Got something to add? Send me email.

(OLDER)    <- More Stuff -> (NEWER)    (NEWEST)   

Printer Friendly Version

-> How to make money with Open Source

1 comment

Inexpensive and informative Apple related e-books:

Photos for Mac: A Take Control Crash Course

iOS 8: A Take Control Crash Course

Take Control of Parallels Desktop 12

Take Control of Automating Your Mac

Take Control of Numbers

More Articles by © Tony Lawrence

---September 6, 2004

Well. Sun "gets it" were SCO doesn't. And Solaris definately has capabilities and trust that Linux may never acheive.

First they donated StarOffice code to create OpenOffice.org to spur the developement and acceptance of a potential MS Office "killer".

Next do you think that they are realy going to go thru with OpenSolaris by openning up Solaris 10 code?

And if they make it GPL-compatable and people start integrating honest to goodness System-V based code legally into Linux, were does it leave SCO and SCO's lawsuit?

To bring up a old joke I said once:
Will GNU change from GNU is Not Unix to GNU is now Unix? Will RMS finally have victory over the closed source community...

Oh and Windows already has open source software in it's OS. To be specific "free software" as in BSD code. Ever notice the pecular directory name of the directory that contains the hosts file in NT-based OSes? This is because of the use of the BSD TCP/IP stack in NT. Also they used plenty of BSD code in their Window's SFU software.

Of course MS likes BSD because they can take and doesn't have to give code back. Or anything back, or even acknowledge that they took BSD code, which was a liscence violation back in the day (not anymore). With the GPL he would have to give back code. So to Bill Gates BSD is good, while the GPL will ruin the economy.


---September 6, 2004

My point is that they don't have to opensource the os - just use open source apps. If you tie it tightly to your own kernel, you don't have to worry about "giving back" - you are giving back something near useless while meeting the demands of the GPL.



Printer Friendly Version

Have you tried Searching this site?

This is a Unix/Linux resource website. It contains technical articles about Unix, Linux and general computing related subjects, opinion, news, help files, how-to's, tutorials and more.

Contact us

Printer Friendly Version

A Perl script is "correct" if it gets the job done before your boss fires you. (Larry Wall)

Linux posts

Troubleshooting posts

This post tagged:






Unix/Linux Consultants

Skills Tests

Unix/Linux Book Reviews

My Unix/Linux Troubleshooting Book

This site runs on Linode