Why can't Unix and Linux code just get along?
Nothing in life or the computer industry is a sure thing, but based on the sheer number of installs and the overwhelming industry traction, I think I can say with near certainty that Linux has become the clear successor to Unix and the future looks tremendous for Open Source software. Put the nails in the coffin of proprietary Unix systems. The big “U” is done. The Penguin shall rule the earth.
Well, maybe not.
Unless we have a major disruptive shift in the thinking of the major industry Unix vendors (IBM, Sun, Hewlett Packard, and Apple), it’s going to be a long time before we see big, monolithic, proprietary Unix boxes go away completely. Because even with Linux’s promise and capabilities, there are still things that proprietary Unix does better, especially in the scalability and reliability department, even when deployed on equivalent x86 hardware.
Sure, if you follow high-performance computing’s approach, where you can cluster the hell out of a few hundred Linux boxes, you can make some pretty nice supercomputers. But when compared to all the workhorse HP 9000, SPARC, and POWER computers stuffed into umpteen corporate data centers in the wild, running mission critical databases and legacy apps, those big crazy supercomputing clusters we all love to hear about are rare successes.
You could take the common view that it’s simply a matter of time before those Unix boxes are replaced. After all, hardware is becoming further commoditized and Linux is beng embraced by key players, such as Oracle and SAP. Perhaps, but there are a number of forces that are sure to keep Unix relevant for the foreseeable future.
For starters, there’s the prevailing notion among many CIO’s and CTO’s that “If it ain’t broke, don’t fix it.” If an application runs perfectly on Solaris, HP-UX, or AIX, why port it to Linux? Certainly, new application development has a better chance of ending up on Linux, especially when you start getting into Linux, Apache, MySQL, PHP/Perl/Python (LAMP) and JBOSS stuff, but your back-end database boxes and truly mission critical systems?
Yeah, I know Oracle says Linux is the future and claims you can run your company on it, and, yes, the company has migrated its own internal infrastructure to Linux, but are Fortune 500 companies really buying into the hyoe whole hog? I have to go with my gut and say no.
This problem would go away if Linux and Unix were really the same operating system– the impossible dream and the holy grail of the open source and open systems religion. Yes, you can run the same open source code on Linux and Unix, but there are enough differences between the two to make application migration difficult enough to stop a lot of companies from doing it entirely. If Linux, Solaris, HP-UX, AIX, and even Apple’s Darwin were to be unified into a single kernel tree, and if there was a multi-platform equivalent to the Linux Standard Base, there’d be a helluva lot of rejoicing, and I believe we’d see a great deal more open source activity than we see now. Think expansion on a geometric scale.
True, much of what I’m talking about here is pie in the sky. But it’s not like the problem is technical. It’s political, requiring the untangling of a lot of history and divergent thinking.
If we can eliminate a lot of the superfluous open source licenses, that alone would be a big step.
If Linus and his crew and the folks over at Sun could come to an understanding about the GNU General Public License Version 3 and code sharing between Linux and Solaris, it would be a sea-changing event.
When SCO finally dies its deservedly horrible death and Unix SVR4/SVR5 can be released by IBM or HP (or Novell, who supposedly has the real rights to the thing) as GPL’d code without further litigious distraction, that will also be a great day.
And if Apple were to truly “Think Different” and to re-license Darwin as GPL code, that would also go a long way to bringing about the “Unixfication”.
The realization of any of the things I’ve stated above would make the world a much better place for all of us in the world of open source.
A dream? Yes.
Impossible? No.

Jason Perlow has a dream. And then he wakes up. You can reach Jason at class="emailaddress">jperlow@linux-mag.com.

Fatal error: Call to undefined function aa_author_bios() in /opt/apache/dms/b2b/linux-mag.com/site/www/htdocs/wp-content/themes/linuxmag/single.php on line 62