Ransom Love on United Linux, SCO Unix
tit4tat writes: "Caldera chief executive Ransom Love confessed to ZDNet UK that "[Caldera is] not moving Open Unix [i.e., the former SCO Unix] onto Intel's 64-bit platform...." I suspected that Caldera bought SCO just to kill SCO Unix, even though they denied it at the time. Now, the first Unix I ever knew is about to be no more. "
the PS2 Linux. Clearly, this is where Linux will be moving in the next decade. Set top boxes, watches, microwaves, etc.
We need to stop thinking nostalgically about computing. The old days are gone. Let's move forward and think of new ways to utilize computing. And by doing that, it may mean thinking about moving beyond Unix.
-- You see, there would be these conclusions that you could jump to
I have never used SCO Unix but I have a lot of experience with OpenServer. IMHO "scoadmin", the tool administring everything on a sco box, was a work of art.
The transparent (to the user) method it had for Kernel compiles is something I would love to see Linux do. Not that I haven't cut a few Linux kernels myself, but it was very neat.
Another great thing was the software installer, and driver support from major manufacturers. Download drivers from Compaq, go to scoadmin/software, add the new software and it would recompile the kernel if needed. Sweet!!
Why is this such a shock? Who needs another x86-based Unix at this point? The only thing that kept SCO alive was the system vendors who needed to be able to run on cheap hardware but didn't want to use a "free" OS.
My own experiences with SCO have all been awful. Having been forced to install it in order to qualify some products with it for a customer, it was a nightmare. The documentation is bad enough that it makes life more difficult that if it wasn't there in the first place. The people at SCO were universally unhelpful, even when we were contacting them to BUY their product. It was a disaster and I can't say I'm sorry to see it go.
"Suppose you were an idiot..... And suppose you were a member of Congress... But I repeate myself."
Remember trying to get standard Unix source code to compile under SCO? Remember dinking with the makefiles, the oddball compiler flags, and non-standard libraries? I sure do. I remember how much I hated SCO and longed for a real Unix. The SCO compiler was closer to Microsoft QuickC (it was an MS compiler!) than it was to ATT cc. It was a royal pain in the ass to use, and I do not hold out nostalgia for those days of pain.
In the article it says, that they will only release the source code, but that they will sell the binary distributuins. I guess that means no more downloading of ISO installation images.
I'm wondering, though, what would they do if someone just decided to download the source code (I guess SRPMS), compile them and the install program and bang it all on a install CD?
Apart from that I like the United Linux idea. The guy has a point about not competing in an area where there's no differentiator between the different distributions. I mean Apache will still be Apache, Squid will be Squid and Postfix..., you get the idea, no matter who packages it (I know that they sometimes apply extra patches, but on the whole, if it's important then all will have it).
So UnitedLinux will remain an open-source project?
Absolutely. The only difference is that the UnitedLinux binaries will not freely distributed. People will be able to download the source code and compile their own binaries, but they will not be able to use the UnitedLinux brand.
...and...
Caldera will provide the product through its reseller channel; one problem that resellers currently face is that Linux is free. This way we give them more of a profit motive to sell Linux, because by adopting UnitedLinux they can generate more revenue.
Huh? So anyone can download the source and compile, (can't call it UnitedLinux, no problem), but you have to buy the binaries (no problem). Doesn't their business model fall apart when people start burning copies of the binary CDs for their friends?
This is, of course, allowed by the GPL, which most of UnitedGNU/Linux will be licenced under, I assume.
There are too many other projects out there right now for Caldera to be messing around with SCO. I think its attention should be on this United Linux thing (tacky name btw). Get that running on IA64 with enterprise features and reliability. Give SCO to the OSS developers that may want to borrow a few ideas. In addition, what does caldera have to gain from continuing work on an OS which is being replaced by more competent, open sourced equivalents (not quite equivalent)? I think it would help them to support the platform (for those unfortunate souls still clung to it) financially, but its time to phase it out. As for 'United Linux', I think I will keep my box 'United' with a clean copy of slack.
It's a little known fact that Sun Microsystems considered using SCO Xenix on their systems before deciding the write their own. The reason they wrote their own was that their main competitor, Apollo, had a fully System 7 compliant UNIX implementation, and DEC was rumored to be releasing that as well on their VAX hardware. Sun decided thta Xenix wasn't UNIX, so they wrote their own.
True story.
If tits were wings it'd be flying around.
Now, the first Unix I ever knew is about to be no more.
Ironically, the first Unix I ever used was Microsoft Xenix on the 68000 Tandy.
You need to learn an OS to be able to use it and understand it well.
/opt /opt.old
/opt (which we tend to assume stands for optional for some strange reason like that it's defined as such by the SVR4 spec) completely destroy a running system? Well, let's just say that SCO's idea of what comprises "optional" system components includes things like "ld.so".
Indeed you do. For instance, want to know how to completely kill a SCO system (and I mean "restore from installation media" kill) with a single command?
mv
Why, might a sane person ask, would renaming
I've learned many things about SCO in my time. First among them is that it will not be missed.
News for Nerds. Stuff that Matters? Like hell.