Slashdot Mirror
Unix Isn't Dead
windows bios world writes: "Compaq, Sun, SGI, and IBM are releasing new machines running Unix. From cnet.com: 'Compaq has begun shipping test versions of a new line of AlphaServer Unix servers using the EV7 "Marvel" version of the company's Alpha processor. ... As expected, IBM released on Monday its p670, a 16-processor machine that's essentially a smaller version of Big Blue's top-end 32-processor p690 "Regatta" server introduced in late 2001.' Also, Sun teamed up with Sony to release video-on-demand servers." And of course, there's OS X.
Isn't Unix a 64-bit OS? (or could it be?) If so, then it most definitely has life for the forseable future.
I have a feeling that MS wants everybody to run a 64-bit Windows Variant in the server world when Itanium architecture becomes wide-spread. That's why they're pulling the whole "Unix is too expensive" bit. They may sort of have a case. If MS provides software that makes setting up a 64-bit server real easy, then they could claim you don't have to custom program it. I think the true cost is a wholly different story, however. If you're running Unix on a 64-bit server, isn't it likely that you're going to want a custom program anyway?
I dunno. I think the only think Unix needs to fear is MS using marketing against it. Anybody who pays attention to what Unix is and what it does is going to have a hard time being convinced that Unix is dead.
"Derp de derp."
"Tru64" Unix is what DEC I mean Compaq puts out on Alpha-based computers. It's based on Mach 2.5 I believe.
Apple's OSX is based on Mach 1.0 I believe so there's a sort of kinship there.
And now for some stuff I'm less sure of:
1. MSFT Windows NT used to run on Alpha CPUs albeit not using the full 64-bits of addressing those CPUs can do. Rumor has it that DEC got a real sweetheart deal on NT licensing because the NT source code was (illegally!) based on "Micah" the operating system that Dave Cutler was working on at DEC before he moved to MSFT in 1988. Comments in the NT source code in the mid-90s confirmed this allowing DEC to get a bit of leverage when dealing with MSFT.
2. Sort of in contrast the first edition of "Inside Windows NT" described an operating system that just could have been Mach 1.0. A lot of the original NT was very reminiscent of Mach 1.0 except less rigorously done. I don't imagine there was any real similarity between the OS described in Helen Custer's book and the real NT though. Mach and Unix were scrupulously ignored in the bibliography and index of "Inside Windows NT" 1st edition. At the time MSFT clearly wanted to emphasize the "N" in NT as "new" even though it wasn't.
The unix mindset has become too pervasive in the midrange computers. Nobody is implementing new ideas because everything has to be `posix compliant'.
Better operating systems are not getting a chance, e.g. plan9, hurd (I am not sure about this myself).
You could argue that unix can assimilate things, but that can only go so far. Some time we have to break out of the mindset.
Linux is nice but has not advanced the state of the art.
Even though Unix is not profit-making like windows it has the same power as microsoft in stiffling innovation (to some degree).
People here spend so much time staring at Microsoft that without noticing they start believing the Microsoft Marketing Department holds keys to the future in its hand. So eventually every phrase said in Slashdot is formed as an answer to reality as marketed by Microsoft, even when no question was asked.
Funnier still, since the [non-]linked article never states Unix was dead or dying.
Unlike many Windows servers, my experience with UNIX servers has been that its longevity is one of its endearing qualities. Services running on UNIX servers tend to have a very long usable lifespan, IMHO due to the fact that the underlying system runs well enough that the application tends to be updated before the system needs to be.
But there is a caveat with using UNIX. The people who can successfully design, architect, administer, and maintain UNIX servers are a tight knit bunch, and as a result of its longevity, they don't tend to move around very often because a given server may be alive far longer than the average Windows server. Additionally, it's been my experience that the longer an individual concentrates on a given subject, such as a single UNIX server, that the more in-depth knowledge they begin to amass about that OS and therefore, they become even more valuable/pigeon-holed into a given organization's IT plans.
This combination of longevity and expertise results in a decreased pool of available personnel available for UNIX projects to organizations at any given time, compared to what I perceive as a larger pool of available Windows talent at any given time. Does this necessarily lead to new projects being run on Windows because the only available talent is Windows? Perhaps...
My vision of UNIX's biggest fear, is that it won't necessarily die, but be bred out of existence because new projects tend to be addressed by whatever resources are available at that time, and if there aren't any available UNIX experts, then nature abhors a vacuum and the projects will be filled with whomever is available at that time.
Rule #1 -- Politics always trumps technology.
Comment removed based on user account deletion