OPINION ARTICLE: Did Linux Kill Commercial Unix?

Forum Forums General Other Distros OPINION ARTICLE: Did Linux Kill Commercial Unix?

  • This topic has 9 replies, 3 voices, and was last updated Mar 9-3:16 pm by Brian Masinick.
Viewing 10 posts - 1 through 10 (of 10 total)
  • Author
    Posts
  • #54532
    Moderator
    Brian Masinick

    Here is an interesting historical article about UNIX and Linux. Much of the article is actually about the history, first concerning UNIX and later concerning Linux. Eventually, as the title states, the author claims that Linux has virtually killed ongoing UNIX development. Interestingly the author knows of only the IBM AIX implementation of UNIX still available (I’m not sure if that is correct or not), but the author also mentions that IBM has invested around $34 Billion into Linux product support.

    If you are interested, read the article at
    https://www.howtogeek.com/440147/did-linux-kill-commercial-unix/

    Brian Masinick

    #54534
    Forum Admin
    rokytnji
    Helpful
    Up
    0
    :D

    Grinning as I grok that user supported Gnu/Linux is making the paid Gnu/Linux versions nervous, like redhat.

    Sometimes I drive a crooked road to get my mind straight.
    Not all who Wander are Lost.
    I'm not outa place. I'm from outer space.

    Linux Registered User # 475019
    How to Search for AntiX solutions to your problems

    #54535
    Moderator
    Brian Masinick
    Helpful
    Up
    0
    :D

    I agree with you @rokytnji. It’s not only commercial UNIX that has faced issues.

    Apple’s MacOS is impacted, commercial Linux vendors, such as Red Hat Enterprise Linux and SUSE Linux Enterprise Server are all impacted by the ability of individuals and companies to download software, although there are license limitations for commercial use.

    I also read an article about Chromebook systems. For those with the ability to purchase new systems, for a couple hundred dollars you can get a perfectly usable low end Chromebook, for $300-$600 you can get a pretty decent computer and for between $650-$1000 you can get a system that can match any network-based system; that has to have both hardware and software companies concerned too.

    Distributions like our antiX are great for instilling life into old computers like the one I am using now, and it makes new computers run faster than most other systems. No matter what else I use, I’ll always keep a copy of antiX nearby and use it regularly.

    • This reply was modified 4 months ago by Brian Masinick.
    • This reply was modified 4 months ago by Brian Masinick.

    Brian Masinick

    #54539
    Forum Admin
    rokytnji
    Helpful
    Up
    0
    :D

    My touchscreen xmass present cost only 169 bucks new. My defunct chromebox cost under 40 bucks.
    Who is gonna pay for a OS when those prices are available?

    My phone cost the wife more. I at least got to say android before she picked the cheaper Iphone.
    That was the brand, iphone, of the 1st phone provided to me.

    Lot’s of countries are using their own special brand of Gnu/Linux mash ups as their main OS of choice lately also.
    Some are so good some dudes are posting how to get the default language to english.

    I guess dpkg reconfigure locales or rpm commands don’t work on some of them. That they are hard coded for their default language.
    That is my take on some of this.

    Sometimes I drive a crooked road to get my mind straight.
    Not all who Wander are Lost.
    I'm not outa place. I'm from outer space.

    Linux Registered User # 475019
    How to Search for AntiX solutions to your problems

    #54540
    Moderator
    Brian Masinick
    Helpful
    Up
    0
    :D

    I did purchase a brand new Chromebook back in November. I had been looking at the Google Pixelbook Go but at $649 for the least expensive model that was too much.
    I did put a few hundred into it because I wanted a backlit keyboard and decent performance. I just missed the best deal around Cyber Monday but saved at least $150 over what my model is going for now, so I did shop for a decent deal.

    With that exception I have not bought any other new hardware in several years. I have a five year old laptop and three other “cast-off” units I got free.
    I’ve been getting my Android phones using Swappa or similar service, looking for good units that are around a year old with at least 1-2 years of software updates available.

    Naturally the older the computer hardware happens to be the more likely I have antiX installed; it’s on most of my systems. No commercial software on these computers!

    Brian Masinick

    #55472
    Member
    PDP-8
    Helpful
    Up
    0
    :D

    I understand the brevity of the article, but sadly shows how some of the major programmers efforts *prior* to Linus’ kernel had the most impact and started the demise.

    Too often Keith Bostic, and Bill and Lynne Jolitz get swept under the rug of history and importance of their work.

    https://en.wikipedia.org/wiki/386BSD

    Of course, RMS had a major influence – let’s see, what compiler did these guys use, even those corporate releases from various vendors during the unix-wars long before the Linux kernel? GCC !

    Ask any active SunOS user who was saddled with bundled software what they did often? Use the GCC compiler and hang out in the comp.os.sources.* newsgroups.

    There are many legends well prior to Linus and RMS, it’s just a shame that many don’t know about Keith, Bill, and Lynne…

    #55492
    Moderator
    Brian Masinick
    Helpful
    Up
    0
    :D

    @PDP-8: I first SAW a UNIX system in 1976 on a lab computer; unfortunately it didn’t have much, if any, network access or major capability. The next time I actually got a hold of a UNIX system the “Unix” of the time (not necessarily the trademarked “UNIX” variety, was often a blend of UNIX System V and the Berkeley Standard Distribution (somewhere in between BSD 3 and 4). back in 1982. One model I got a hold of was an NCR “Tower” model; I later got a hold of a “PC” hardware version that was mostly UNIX System V and I also got a brief look at an early SunOS on a Motorola 68000 chassis; that was based on BSD; later when Sun Solaris came out, it was a much different OS with a UNIX System V base built collaboratively with AT&T and Sun Microsystems.

    All of these systems had at least SOME AT&T code, and were therefore proprietary licensed.

    When I joined Digital Equipment Corporation in 1985, I joined a Bell Accounts Systems team, and because of this, all of our customer base was either AT&T, a “Regional Bell Systems” operating company, or a telephony company a affiliated or competing with AT&T. Nearly all of them knew, and generally preferred, UNIX.

    In 1985, Digital actually had a few small UNIX operating systems groups for various equipment; none were the exact same base; each worked with different processors. Most of these UNIX variations were actually sold to either academia or small businesses. AT&T and their close partners had REAL UNIX System V licenses and wanted to run “REAL” UNIX. Digital had a group in New Jersey near various AT&T sites; one local office in particular would write device drivers for new hardware so the REAL AT&T UNIX would work with the AT&T code.

    TEN years later, in 1995, by that time, Digital had developed a hybrid UNIX called ULTRIX; this work finally evolved enough to meet official UNIX branding, and Digital UNIX was born; just before Digital Equipment Corporation was consumed by Hewlett-Packard, their Digital UNIX exceeded OpenVMS and their other proprietary systems in units sold and only “legacy systems” and classic applications remained on old systems – and thus one of the classic infrastructures faded into obscurity.

    What was also “interesting” was that by the time Digital actually got serious about UNIX, the beginnings of the Linux era were starting to appear; so once the 2000s came, UNIX, though not entirely gone, is now a sliver of what it was. I think IBM may still have their AIX available; but they do much more with Linux, as do HP, Dell, and other hardware manufacturers.

    The “smart” companies either rewrote their applications to work on other systems or purchased more turnkey solutions supported by software vendors. The ones who stick with their stuff may run mainframe systems or applications with hundreds of thousands or even millions of lines of code.

    By the time the 386BSD work was underway, as you know, Linux was beginning to evolve; the first Linux build was in 1991; by 1992 it may have been behind 386BSD, but by 1993 there were a couple of usable Linux distributions and the “race” was pretty much won, both for server and desktop configurations. Five years after that there were quite a few distributions, a couple of which were straightforward to install and use, both personally and professionally.

    • This reply was modified 3 months, 2 weeks ago by Brian Masinick.

    Brian Masinick

    #55496
    Moderator
    Brian Masinick
    Helpful
    Up
    0
    :D

    Bill Joy was one of the people at University of California at Berkeley who did a lot of the original work on their BSD UNIX;
    Keith Bostic, and Bill and Lynne Jolitz did a lot of good work on 386BSD, but it never really had much popularity; it was the other BSDs, foremost FreeBSD, and also NetBSD and OpenBSD that advanced the initial efforts of the 3 pioneers who first put the BSD code on commodity hardware. Because of the legal struggles between AT&T and the Regents of the University of California (predominantly at Berkeley), this may have been what enabled Linux to jump ahead of the BSD work because people feared legal action if they tried to build products based on a BSD variant. There may be other reasons in addition to this, but Linux took off while BSD is quietly getting by, mostly as an alternative Web server platform.

    Brian Masinick

    #55515
    Member
    PDP-8
    Helpful
    Up
    0
    :D

    We owe a LOT to DEC. First for the “raspberry-pi” like pricing of the time for “affordable” PDP-11 series of mini-computers for the first legends of ATT to work with – when what they really wanted were vastly expensive PDP-10 mainframes, but got shot down immediately by management after Multics. 🙂

    We also owe John “Maddog” Hall of DEC a lot of debt for getting DEC to give Linus a 64-bit Alpha so the kernel could be 64 bit, and more portable beyond the x86 architecture!

    I think 386BSD stalled not just for differences in project goals (academic research, centralized control et al), but because it was guilty-by-association when the ATT lawsuits started with BSDi’s “Its UNIX” campaign – even though 386BSD was *not* part of the lawsuit. I can’t speak for the Jolitz’ , but my *sense* was that they wanted 386BSD to be more of a research vehicle trying to drop a lot of old architecture baggage, no matter how much time that took from an academic perspective, as compared to the later Net/Free BSD’s wanted to get stuff to work and fix things later even if it wasn’t pretty at first – much like Linux was. Not so much a competition, but a difference in what research actually *means*. But those are my thoughts and guesses, and should not be construed as to what the participants during that history really thought. History is written by the winners right? 🙂

    Some OG friends of mine were just freaked out – even after the lawsuit was over – but didn’t want to take the chance of investing time or money in anything BSD waiting for the next lawsuit from some other entity to fall. Luckily Linux worked well enough, and with the GPL seemed defensible from the start.

    The funniest thing (sad really) is that I watched some jump-ship twice when the SCO debacle came around. When the first threat from ATT appeared, they dropped BSD to concentrate on Linux, and when Linux later got threatened jumped ship back to BSD or gave up to a proprietary solution. Didn’t want to take the legal chance of getting hammered without a big legal team behind them.

    Talk about “distro-hopping”. 🙂

    Interesting times which I think we take for granted looking back today. But back then, it could really keep you up at night.

    #55523
    Moderator
    Brian Masinick
    Helpful
    Up
    0
    :D

    @PDP-8: I see in your handle and in your knowledge of history that we have some common understandings.

    Name dropping warning: You mention Jon “maddog” Hall. It just so happens that when I first joined Digital Equipment Corporation in 1985 that one of the first UNIX luminaries @DEC that I encountered in the hall was indeed “The Hall”, a.k.a. “maddog”. Back when our office was in Merrimack, NH (now one of the Fidelity Investment offices), from 1985, for a couple of years UNIX development, and also RSX and RT-11, development, were in Merrimack.

    Since my job was involved with telephony companies, my physical location was just the other side of the break room. Jon and I would often grab a bag of popcorn in the afternoon and tell stories while waiting. Unlike many of his peers, Jon wanted to see REAL UNIX included in Digital’s version of UNIX, and later on, when there really WAS a Tru64 UNIX @DEC, Jon was a big advocate of Linux, and actually brought Linus Torvalds to a Digital facility, where he gave a lecture, which I was able to attend.

    In 1997 when I decided to leave Digital – and by then I was a member of the localization team of Digital UNIX, one of my friends who came to Newick’s, a lobster restaurant in Merrimack, was my friend Jon. A few months ago, I picked up a copy of Linux Pro Magazine, and found a familiar author of a monthly column, Jon “maddog” Hall! Interestingly, the article I read in the issue was about many of the things we talked about long ago in those hallowed hallways! Great memories!

    As far as “hoppers” go, I certainly qualify, but not out of fear, simply out of “geek” behavior; I’ve experimented with editors, browsers, different operating systems, etc. for as long as I can remember.

    One story I particularly enjoy is the time I got – fairly early in my career at General Motors – was to try out – and actually maintain – hourly personnel software. Most of it ran on mainframe computers. The oldest code was terrible and ugly, but it actually worked. Some of the “newer” personnel systems (that really didn’t enjoy a very long lifetime were on MULTICS, a VERY expensive, but wonderful operating system; we had it on a Honeywell mainframe; that tells much of the story right there. RCA, GE, and Honeywell were three of the short-lived hardware companies that at one time or another ran MULTICS, the full implementation of a great operating system that was before its time. What I find fascinating is that nearly all of the features in that great OS of the past are finally implemented in Linux, but only a few of them made it into the original UNIX, a.k.a. joked about as possibly being “Eunuchs”! Punsters, those Lab geeks from AT&T, Dennis Ritchie and pals.

    Lots more stories from the past… I’ve written about a few others in the archives here and elsewhere from time to time.

    Brian Masinick

Viewing 10 posts - 1 through 10 (of 10 total)
  • You must be logged in to reply to this topic.