Ransom Love's Secret Master Plan for Linux and UNIX
Before I start reminiscing about Linux's poor-but-honest days, here's Ransom's advice for ISVs developing for future SCO products. It's simple. "Develop to Linux now." If he's sure about nothing else, Ransom is sure that Linux is the UNIX-on-Intel standard. As for the ISVs, "None of them are going to be able to say that they don't want anything to do with Linux." Whether the ISVs ultimately decide to support Linux or not, they will be learning it.
But this doesn't mean SCO's current offerings are off the menu. They're just getting a coat of Linux paint. "Partners and OEMs need high-end today, and UnixWare can fill that role today." The Linux kernel personality on UnixWare is being demonstrated here at the show, and is now undergoing performance tuning and testing. It will be a released product within a few months. So there's your write-once solution for those two OS flavors, anyway.
The three big server OS contenders, Ransom says while holding up three fingers, are Solaris, Linux and Windows. Dismissing the Windows finger with a flick, since Linux has it beat handily, he moves on to the big challenger, Solaris, which scales all the way up to Sun's not-quite-big-iron Enterprise servers. Since Linux doesn't go up that far, UnixWare "continues to deliver a value proposition for a period of time." Caldera is offering a family of Linux-compatible OSes that scales across a wide range of Intel boxes, just as Solaris scales across a wide range of SPARC-based boxes. Naturally, the Intel-based boxes are cheaper, so Ransom has a potent competitor for Solaris. Do I have that right?
Beyond that easy-to-explain story, Caldera's future holds an intricate OS breeding process. Applications will play happily in a we're-all-Linux-here user space, under guidelines enforced by the benevolent, long-awaited LSB, but anyone interested in hacking the OS itself will need a scorecard.
"Linux is being pulled across the spectrum of IT solutions, and a single kernel won't scale," Ransom says. The considerations that are important to embedded Linux, to the small server market Linux rules now, and to the midrange and high-end server markets are different. As Linux "forks" - hopefully through a proliferation of compile-time options, not a real fork, Ransom hastens to add - the high-end parts will end up participating in some sort of technology-sharing arrangement with UnixWare.
So, Linux, UnixWare, Openserver, Monterey (or whatever they're calling it now) - what is the secret master plan? I draw a chart - OSes down the left, years across the top, fill in "Linux 2.4" in 2001 with a question mark, and ask Ransom to fill in the rest. Arrows sprout from Linux and spread like fungus tendrils into the "UnixWare" and "Monterey" areas - that's the compatibility thing - and a big arrow moves forward into the future along the UnixWare/Linux dividing line. This represents the spawn of Linux and UnixWare, an über-OS with a yet-to-be-determined licensing policy. Ransom says you'll be able to see the source code, but parts will be open source, and parts will be "viewable source" - you'll be able to read it, but not modify and redistribute it.
A dangerous course indeed, in this market. Caldera can't afford to keep UnixWare alive as a proprietary product if the mainstream Linux kernel blows past it in performance and features. "Our model is to innovate, give back, and innovate again," Ransom says. So, if current industry trends are any guide, Caldera will be making a lot more of UnixWare a lot more open than they're expecting.
So, for all you UNIX goddesses and gurus out there, the big question has got to be: What chunks of historic UNIX will Caldera release under an open-source license? Love says that code is coming, but, "the specifics we don't know yet." Let's help him out. Eric Raymond wants some document-processing tools, including a program that determines the "reading level" of text you feed it. I'm sure the rest of you have some ideas as well. So, please write and tell me which UNIX tools or other code you want, and I'll visit Ransom in Orem, Utah and present him with a list of the top requests. Since (1) Ransom is a nice guy and (2) he won't make any money from this stuff anyway, I predict he'll GPL a good-sized basket of software.
One more question for Ransom is whether he got any software patents in the acquisition. He doesn't know, and adds, "That wasn't our intent." If it turns out Caldera now has some software patents, I asked if we could put them under a free license, and Ransom doesn't see why not. So, I'll get in touch with Caldera's corporate counsel to talk about that. And, Ransom, let's print up some "UNG (UNIX is not GNU)" shirts for when you put that UNIX code under the GPL.
Webinar: 8 Signs You’re Beyond Cron
11am CDT, April 29th
Join Linux Journal and Pat Cameron, Director of Automation Technology at HelpSystems, as they discuss the eight primary advantages of moving beyond cron job scheduling. In this webinar, you’ll learn about integrating cron with an enterprise scheduler.Join us!
|Android Candy: Intercoms||Apr 23, 2015|
|"No Reboot" Kernel Patching - And Why You Should Care||Apr 22, 2015|
|Return of the Mac||Apr 20, 2015|
|DevOps: Better Than the Sum of Its Parts||Apr 20, 2015|
|Play for Me, Jarvis||Apr 16, 2015|
|Drupageddon: SQL Injection, Database Abstraction and Hundreds of Thousands of Web Sites||Apr 15, 2015|
- "No Reboot" Kernel Patching - And Why You Should Care
- Android Candy: Intercoms
- Return of the Mac
- DevOps: Better Than the Sum of Its Parts
- April 2015 Issue of Linux Journal: High-Performance Computing
- diff -u: What's New in Kernel Development
- Designing Foils with XFLR5
- Drupageddon: SQL Injection, Database Abstraction and Hundreds of Thousands of Web Sites
- April 2015 Video Preview
- Consent That Goes Both Ways