Can GNU Ever Be Unix?

An anonymous reader writes "The question isn't whether Linux can be certified as Unix. At least some distributions no doubt can. But who would pay for it? And is it worth the trouble? Jem Matzan asks these questions on NewsForge, and reminds us that the Open Group, not SCO, owns the Unix trademark,"

Who cares?

[Fished]

Who cares if Linux is Unix at this point? We are rapidly approaching the point at which UNIX is a Linux-like operating system rather than Linux being a UNIX-like operating system. I'm more or less convinced that proprietary UNIX is dead as a major force in the market. Moving forward, Linux will be setting the agenda and proprietary UNIX will be playing catch up.

This is particularly evident when you notice that the major improvements in some recent version of Solaris (8 & 9, but not 10 apparently) is to add more open source software and stability improvements.

Boy -- talk about your pointless questions...

[YankeeInExile]

I think it is almost certain that some distro of Linux could easily pass OG's test suite. It is also almost a certainty that FSF/GNU would never opt for it on religious grounds.

The rest of the thread is now available for stupid /. jokes.

In Soviet Russia, The Open Group petitions GNU for certification.

All your Linux Standard Base...

[tepples]

GNU/Linux seems to be evolving as its own standard

And this standard is called LSB [linuxbase.org].

Why?

[toupsie]

What's the point? To be facetious, Unix is old and busted, linux is the new hotness. Instead of being focused on the past, look to the future. Being stamped "Unix" doesn't have the same meaning today as it did ten years ago. Bean counters today aren't asking about a Unix solution but a Linux one. Its the tech buzz word of the last 5 years. To the general public Linux has better name recognition than Unix. In fact, I commonly hear non-tech people referring to real Unix systems as "Linux".

Re:Boy -- talk about your pointless questions...

[imkonen]

Is there are particular reason that FSF/GNU would object to certification? I haven't read the 55 pages of requirements, but I didn't get the impression that either "closed source", "not free to modify" or anything else incompatible with the GPL are part of the requirements. If you take a certified version and modify it, you undoubtedly can't call it certified anymore, but you're still free to redistribute your modified, "no-longer certified as Unix" version.

But I think the more significant point is that it's not FSF/GNU who would have the most incentive to get a distro certified as Unix. As the article pointed out, it's probably the hardware companies like IBM and Sun who would find it worthwhile.

Open Group certified Linux

[foobsr]

If you ask google [google.de] , it does not seem a good idea. SCO comes up first.

CC.

Paying for unix?

[Cow007]

Free BSD as the name suggests IS free. Unix is not a valad trademark beacause beacause of its many forks and variations it has become a generic term. IMHO I think the people at SCO and Sony (with its joke of a digital music player) fell off the stupid tree and hit every branch on the way down. Protecting your copyrights has become difficult these days. In a world of convergance, reverse engineering and "hey that was my idea" tactics A review of copyright laws and procedures needs to take place. If we come together and decide on open and fair standards that make things work. Things are going to change, biusness models that were viable before will no longer work. But its time to evolve.

Re:Boy -- talk about your pointless questions...

[Halfbaked Plan]

it's probably the hardware companies like IBM and Sun who would find it worthwhile.

IBM have specifically taken a hands-off approach to Linux. They have team members contributing to the 'greater' Linux source trees, but there's no sanctioned 'IBM' brand of Linux. They don't want nor need an 'official' IBM version of Linux.

The reasons for this are complex.

Re:Who cares?

[Zeinfeld]

Who cares if Linux is Unix at this point? We are rapidly approaching the point at which UNIX is a Linux-like operating system rather than Linux being a UNIX-like operating system.

I have been saying that for several years now. UNIX is all but dead. The only commercial UNIX likely to still be arround in ten years time as an ongoing product is OS/X. Solaris will have long since joined IRIX, Digital UNIX and VMS as O/S you can still buy and occasionaly see a minor upgrade for it.

There is a basic set of core functions that O/S do and this has not changed in principle for over a decade. Log based file systems, threads that work etc are now standard, but none of this was new ten years ago.

The interesting stuff all takes place either above or below the O/S layer. .NET, J2EE etc are where interesting stuff is happening.

At the driver level I think that both Unix and Windows have the model hopelessly wrong. We have at last got past the point where we have to recompile the kernel for each new driver. But drivers are still mostly executable code while the differences between devices of the same genre are with very few exceptions the type of thing that can be described by code tables.

I would like to see device manufaturers get out of the device driver writing business, have a genuinely generic driver in the O/S and discover the repetoire of a particular device by reading a configuration file - preferably one that can be read from the device. From a pragmatic point of view XML would probably be a good match for the task since you would inevitably need structured data and a way to extend the basic data structures.

Unix once had this with the printcap and termcap files. Unfortunately people just seem to be unable to resist turing complete code.

Re:Why?

[Anonymous Coward]

Unix based?

This must be an Amiga from a parallel universe you're talking about.

I still remember the joy of the developers when the GCC compiler was ported to it, but it was not until 1990 if I remember correctly.

You were not talking about the 2500UX, were you? That's another thing completely.

Re:No. Unless Linus or Posix makes a change.

[Anonymous Coward]

Sure he does. But he understands that he's working for the distro companies. If RedHat/Suse/IBM/etc thought it was a priority to get UNIX-Certified, Linus would merge their patches.

Note that Linus did a 180 on "Linux Threads" versus "POSIX Threads" because Linux vendors wanted portability.

Re:Why?

[Brandybuck]

Best reason: The Unix monicker would stop all the GNU fanatics from frothing at the mouth every time they heard the word "Unix".

Case in point. I'm at a conference. Someone asks me what I use. I reply "Unix". Suddenly their eyes bug out, their ears turn red, and they scream, "no you're not, GNU/Linux is not Unix!!!" Then I explain that I'm not running Linux and they're heads finally explode.

Okay, that's a bit of hyperbole, but in my experience most GNU advocates (a distinct breed from Linux advocates) take great pains to enforce the Open Group's trademark.

UNIX matters

[mnmn]

well, it mattered to me.

Back in the days, around 1995, my friends and I were looking for any UNIX to put on our machines to learn. We tried an old copy of SCO Unix which didnt work, and were busy snooping till we found Linux just as it was getting popular online. We got into Linux because we were out looking for UNIX.

Nowadays I've got AIX and Solaris on ultrasparc to play with, so I can finally brag about knowing 'unix', but would be real nice if Linux is called UNIX. Even though SCO has spilled cold water on the brand name, it still carries enough weight, and maturity of two decades, to get attention. Linux is still new to the scene, and UNIX has carried the full weight of the Internet since its birth... that means something.

Linux means alot more now, so can UNIX be Linux, or at least its former self? Thats possible, if Linux is branded UNIX, and UNIX can once again claim to be a popular flexible modern OS. Cant do that with SCO Unixware.

Close enough

[Anonymous Coward]

For the amiga's design purposes, it was pretty close to unix. It was intended to be a single-user system, so obviously, it didn't have all of unix's multi-user/network stuff. But it had true multitasking including it's own form of task prioritisation, pipes, similar commands, dynamic libraries, a powerful scriptable shell, etc.

Sure, it wasn't unix technically speaking. However, if you allow a certain 'artistic license' like we do for OS X, and add in a 'handicap' since this was a 16-bit 7Mhz machine usually with 1MB RAM or less, we can start to see just what AmigaOS was. IMHO, it was closer to a personal version of GUI-enabled Unix than OS X has managed yet. Technology has moved on, but I still miss the productivity I had on AmigaOS.

New Acronym

[NitsujTPU]

I don't know if this is a new idea, but why not just use Linux as an acronym.

LINUX:
Linix Is Not UniX

Similar to PINE:
Pine Is Not Elm

Re:UNIX matters

[boaworm]

I know that this is not entirely correct, but this is the way it "feels" like for me, personally.

UNIX is the actual operating system (which Linux has made a very powerful and capable clone of). It could be OS X, Solaris, AIX, *BSD or whatever. Fine, now I have my UNIX station, what am I going to do with it ?

Of course, I'm going to run GNU software on it. That's the whole point of running UNIX, the GNU software. Killer apps like the X server(s), Emacs, ftp/web/dns servers and virtually any other software you could ever imagine. I'm running UNIX (or clones) to run the GNU software.

I'm curious, does anyone else share this view ?

Re:It's GNU/Linux!

[ultranova]

Hurd is definitely a good idea, but so far it is only that: an idea.

I have been hearing about Hurd at least since 1992 or so, ever since Linus started his project. This is 12 years now, and nothing concrete has come up yet, that can be adopted by the masses.

Of course, this is propably caused at least partially by Linux's success - Linux is drawing all the the good OS programmers.

The dev mailing list archives seemed depressingly empty when I checked :(.

Don't get me wrong, I like many of the ideas and design decisions they have. But my gripe is that their model does not allow hordes of programmers to join in and get things out faster, like the Linux model.

Ironically, the Hurd should (at least in theory) be a lot easier to modify and develop. I wonder if I should give it a go... Maybe someone could port the Mach microkernel so it would work under Linux (or put a "Mach emulation layer" into the kernel) ? This would allow the Hurd to use the Linux device drivers, and allow people to try it without needing to repartition (you could use files as block devices via the loopback device). It would be just the kind of kick this project needs...

Of course, you could make an user-mode Mach emulator, but that would likely be pretty inefficient. Or would it ? Hmm...

Great, I already have all kinds of projects underway, from a roguelike for Java to a Usenet picture grapper with a web interface, and now this tought just had to pop in ;).

Mach, with an emphasis on Mac...

[MsGeek]

Apple's second foray into xNIX-like operating systems after A/UX was MKLinux. Basically MKLinux was the bastard child of Linux and the Mach microkernel. It's no longer actively developed by Apple for obvious (MacOS X) reasons, but it's still got a bit of a community left at http://www.mklinux.org/ [mklinux.org].

I think it only runs on Power Macintosh, and it's the only flavor of Linux that will work on NuBus (first generation) Power Macs.

Re:UNIX matters

[AstroDrabb]

UNIX is the actual operating system (which Linux has made a very powerful and capable clone of). It could be OS X, Solaris, AIX, *BSD or whatever.

The thing is, is that Max OS X and *BSD are not UNIX [opengroup.org]. If you look at that list, according to the Open Group, the only recent OSes that are _true_ Unix and allowed to be called UNIX are Sun Solaris, IBM AIX and Compaq Tru64. So if you need a true Unix, these are your only choices. However, for me and probably many others, if you need close-to-Unix, then Linux, *BSD and even Mac OS X are very, very close and will do the job very well if not better then the current _true_ UNIX system out there. I don't think that the Unix name will be that important in a few years. Linux, Mac OS X and *BSD have a name already in the IT market. What would getting Linux, Mac OS X or *BSD Unix certified do for them?

I do agree with you though about the GNU software. That is what makes a good Linux/*BSD system.

What Happened to Apple's UNIX lawsuit?

[HighOrbit]

Last year the Open Group sued Apple because Apple was advertising OSX as a "UNIX". This was reported on slashdot here [slashdot.org]. Apparently Apple had originally licenced the trademark, but had stopped paying fees and the licence lasped. Apple contended that "UNIX" is now a generic term and that they shouldn't have to pay to licence it. The Open Group, of course, felt compelled to defend their cash-cow trademark, so they lauched a lawsuit.

So, where is this now? I did a search but even the mighty power of Google can't seem to find any reference to the outcome or status of the case. Does anybody know what the status of this case is? Was it settled, or just languishing on the court's docket?

Re:It's GNU/Linux!

[alangmead]

The difference between what the HURD has designed and the various efforts to implement them in Unix like operating systems is in user space. The HURD allows (or will allow, once its done) per-process overrides of any system call. LUFS simply allows a user space program to tell the kernel how to represent a device. With the HURD, there is no reason why another user will even see the results of your translators.

What I don't understand, is how the HURD is so late when it has these features. Creating or modifying a new system call should just involve installing a translator, testing it, and at worst logging out. No rebooting to test a new kernel feature.

Could Linux be an extension to UNIX?

[sdcmk]

I think a good point to remember is that the UNIX standard only lists minimum requirements for a UNIX. Linux could merely implement these requirements and still be able to innovate.

Not only would this increase the capabilities of Linux but it could also serve as a way to bring UNIX applications into the Linux world with little effort. Once this is done, Linux can then efficiently replace UNIX itself.

What I believe the industry needs, is a "modern UNIX" that can compete with Windows from not only a user point of view but also a technical one. We are starting to come to the point where not only is UNIX but Linux is starting to become "left behind" by Windows through .NET and the upcoming Longhorn. If Linux is going to be relevant after Longhorn on a technical level, it needs to start to innovate. The UNIX standard is a good place to start.