The biggest mystery in the history of Linux

The biggest mystery in the history of Linux

Summary: I think it's safe to predict that future historians looking at the history ofLinux will spend a lot of time muttering about the events preceeding, surrounding,and flowing from, Novell's purchase of SuSe in late 2003.

SHARE:
TOPICS: IBM
67
I think it's safe to predict that future historians looking at the history of Linux will spend a lot of time muttering about the events preceeding, surrounding, and flowing from, Novell's purchase of SuSe in late 2003. Personally I'd very much like to understand what happened, but I don't have the information needed to reach any conclusion -and if you do, I'd appreciate a comment here or an email.

There's some stuff we do know. Just prior to the sale, SuSe was the leading distribution, arguably in terms of technology and certainly in terms of positive press both in Europe and North America. IBM had its cities initiative in full court press right across Europe with Munich as its PR star (mainly because of Ballmer's disasterous attempt to read the riot act to its city council) and significant successes either nailed down or within reach in Austria, Germany, France, and possibly Denmark.

At the time, however, the biggest plum at hand seemed to be Daimler-Chrysler with some serious people predicting a worldwide desktop win for the IBM/SuSe combination.

Those rumors died within weeks of the announcement that IBM had put $50 million into Novell without apparent reason and Novell had bought SuSe at a firesale price.

Today SuSe's key people have left or are leaving, the Daimler-Chrysler deal is off the front burner, IBM is facing layoffs across Europe, and the cities initiative, although continuing, appears to have lost momentum.

So what happened? Does anyone know? Are these things related or is the timing coincidental?

 

Topic: IBM

Kick off your day with ZDNet's daily email newsletter. It's the freshest tech news and opinion, served hot. Get it.

Talkback

67 comments
Log in or register to join the discussion
  • Somehting happened in DP that wasn't

    Bill Gates' fault. C'mon, you made that up.
    BXLE
    • Microsoft has always had good luck...

      ... in its opponents.

      Whatever the philosophies and attitudes involved, commercial competition comes down to real companies making real money. Or not.

      And mistakes by competitors have often come just in time to save Microsoft from difficult situations.
      Anton Philidor
      • Maybe not luck

        M$ keeps the pressure on its competitors AT ALL TIMES - and that pressure can lead to mistakes (there are many real-world examples of this). The problem here is that M$ competitors CAN'T AFFORD any mistakes! One mistake and you're dead! Hmm, and people don't believe in Natural Selection . . .
        Roger Ramjet
  • What I know`

    A few years ago, I was in charge of developing a Linux server load for Company "F". The company had chosen IBM as its primary IT vendor, and thus the x-series e-servers we to be used for this load. We (of course) chose DeadRat Linux for the server load, and I merrily went on my way to develop the process. We purchased 6 licenses for RHEL and I was trying the load on different servers (some older Dell servers, etc.). More and more departments wanted to alpha test (gee NO problem finding volunteers!) the load, and some of them we allowed to. As the months wore on, DeadRat became impatient with Company "F" - as the big orders had not yet materialized. First they kinda FORCED us to pay for some VERY high-priced consultants (they were very good though), THEN they started demanding that we audit our systems and pay for every license that we found.

    Now most EVERYONE knows that when you work with a top 10 corporation, the process moves SLOWLY. You need to look at the future, and spoon feed the company until they make the big decision to go with your product. DeadRat was apparently UNAWARE of this, and was treating Company "F" as you would a small company (where threats MIGHT have some meaning).

    IBM was partnering with us all the way on this project. Their reps confided in us that DeadRat was like this all the time, and that they REALLY preferred to work with SuSE (of course this was off the record stuff). So after 8 months of research and many dollars to DeadRat, the plebians (SysAdmins) pleaded management to look at SuSE. The AMAZING part of this is that THEY DID! Anyone out there KNOWS how tough it is to get management to change their direction - especially after they have invested money, but I was stunned at how easy it was. SuSE sent their TOP man out to help us with the load (the AUTHOR of Y@st). Where with DeadRat everything was "sure we can do that, pay up!", SuSE just DID IT! Needless to say, DeadRat was shown the door!

    This comparison between the 2 styles of Linux companies was seen by IBM in MANY different places. I'm not sure that Company "F" tipped the balance, but I'm SURE that IBM knew that it was counterproductive to continue depending on DeadRat. The purchase of SuSE by Novell was for IBM to get better control over the Linux distro process. But IBM didn't want to be the distro owner - THAT'S why they had Novell buy it (a company that is friendly to IBM and can be relied on).

    I have NO IDEA why ANYONE would use DeadRat Linux - especially if they had to deal with DeadRat salesdroids.
    Roger Ramjet
    • The moral of the story

      What I forgot to add was that Comapny "F" has over 900 Linux servers (MOST with multiple CPUs - i.e. licenses) today (and growing). ALL that money COULD have been DeadRat's, if they had just been more intelligent . . .
      Roger Ramjet
    • Good story, good comments.

      One observation: Novell bought SuSE for its own purposes, too. With Netware evaporating to Microsoft servers, they needed a package to sell that might slow the loss of their prior moneymaker, maybe even give them a replacement main product.

      That means Novell SuSE has differenmt priorities from SuSE as an independent company, which didn't make money.

      You wrote:
      The purchase of SuSE by Novell was for IBM to get better control over the Linux distro process. But IBM didn't want to be the distro owner - THAT'S why they had Novell buy it (a company that is friendly to IBM and can be relied on).

      You're right about IBM, of course, but Novell had its own motives.
      Anton Philidor
    • Thanks

      Thanks for sharing your experience, Roger. That was a very enlightening story.
      Real World
    • Some bad assumptions, maybe pay later

      If I were the guy at "F" company in charge of
      the decision that was described, I might want to look out 3-5 years and wonder if the OS I chose will still be supported, or if it will cost me more $ to deal with than the $ that the other Linux guy was charging (if it ends up being a dead end road where Novell can't figure out how to operate profitably with it...unless F agrees to buy $10 million of their other offerings that no one really wants, and starts using more Netware everywhere). At least the other Linux guy bashed in this post has proven that it can make a profitable business around Linux.

      One could also wonder that the Suse guys had to sell because they did business deals like those described with "F" company, and listened to the I company on many other similar type deals. If you add those all up, you get a bleeding company on one end, Suse, in a compromised position. Everyone else doing fine around them in this situation, and them with their face being scraped on the ground.

      Then again I grew up in Detroit and saw the auto companies say "The Japanese can't make cars" many times over in the 1970's, so the wisdom of F or G companies may not be good ones to follow.
      bassboy_z
  • Is it possible some people are just stupid?

    I'm serious. Sometimes screw-ups aren't diabolical. Novell didn't do its homework before it bought SUSE. It apparently appointed Messrs. Moe, Larry and Curly to the management team there. With predictable results.

    Cluelessless doesn't need an explanation.
    DanaBlankenhorn
    • Hear Hear

      NOW if you can JUST get Mr. Spicoli and his 911 conspiracy theorists to understand this! Did GW mastermind the biggest criminal conspiracy of all time - or did a bunch of bumbling bureaucrats fumble and stumble their way into it? I seem to remember that Moe, Larry and Curly did GREAT damage . . . ;)
      Roger Ramjet
    • SuSE

      SuSE was a distro with FANTASTIC engineers behind it. It was/is a very stable and strong server offering. Their marketing was predictably terrible - as were their management (too many tech heads). This is BEFORE Novell bought them.

      Great technology is NOT what made M$ the premier software vendor - MARKETING was! This cannot be overstated as MOST software vendors do NOT have this ability.
      Roger Ramjet
      • Microsoft's success...

        ... came from giving users and developers what they wanted: ease of use and striking features.

        By your definition, that's not great technology. It only produces $ billions in sales.

        Pleasing users is not marketing only. It's a strategic plan for software design.

        After all this time, people are still underestimating Microsoft. What kind of success does the company have to have for people who oppose them to take them seriously?
        Anton Philidor
        • You are right

          Yes I was being oversimplistic. M$ focus was on the user/developer and met and exceeded their expectations. This was SuSE's goal too (their audience was techies though), but their marketing was non-existant. This seems to be opposite DeadRat, as their marketing is good "We're number one!", but the customer experience is terrible.
          Roger Ramjet
  • Maybe Munich was an example for IBM?

    Expensive, long-delayed, politicized, required IBM front and center to resolve the problems.

    My favorite observation from the scene was that none of the city's available suppliers knew anything about Linux. Working ab ovo is problematic, even for IBM.

    Is it possible that the IBM cities initiative wasn't a potential gold mine for IBM? IBM is, after all, interested only in gold mines, and not philosophical triumphs.
    Anton Philidor
    • Expensive?

      The trouble with migrating to Linux is that its a BIG DEAL to NOT use M$ - so you INSTANTLY become newsworthy. Munich would have had less problems WITHOUT the Balmer visits and media circus.

      Linux on the desktop is like a new sports car - except there aren't any options. No fancy wheels, or leather interior, or cruise control - but its fast and reliable. How do you market THAT and how many people would be willing to buy (into) it?
      Roger Ramjet
      • Yes.

        The cost of migrating to Linux was greater than the cost of staying with Microsoft. Microsoft was the lower bidder, but Munich turned down the lower bid in order to be an example to the world.

        And that's before the unexpected costs started.

        There's a niche for cheap and incapable, but reliable. Takes some money to market, but IBM and all the tech periodicals frantic for non-Microsoft news and philosophically predisposed toward open source will boost Linux.

        Linux has its opening now. There may be a lot of problems with the Longhorn launch, but one good thing it will do: include a cheap and incapable, but (approximately) reliable version.

        These are actually Linux's best days.
        Anton Philidor
        • The short term cost

          was high yes, but the long term return would make up for it. People seem to be so short sighted these days.

          As far as I know the SuSE / Munich deal is still in the works. Havent' heard otherwise yet.

          As for Red Hat (Dead Rat suits them better...) they are like any other American company. All about $$$ and no quality to back it.
          Linux User 147560
          • How long is long term?

            Even if you were right and the annual costs for Linux are less than for Windows, the advantage would have to be significant and continue for many years before the initial costs were balanced. Present value of money...

            And then, Microsoft has been producing a lot of studies of TCO showing that even though the initial cost is higher, maintenance is cheaper, largely because of lower personnel costs, executives' favorite kind of cost to cut.

            I think the methodology of some of those studies that I've seen has been flawed, but if they were self-evidently bad the studies would persuade no one. And they'd ruin the companies that performed the studies.
            I suspect an objective look would find no glaring differences.

            So I expect that Munich's lost money will not return until long after the current politicians have gone wherever it is politicians go when no longer in office.
            Anton Philidor
          • Studies..

            most studies are biased towards whomever paid for them, no matter what the study. They don't ruin the companies that perform the study because the executives like these studies. They think they are actually objective.

            Could a truely objective study on TCO for Microsoft and Linux be performed - I doubt it. There are too many variables to accurately study with the small focus groups they use.

            Personally, I think the only difference between the TCOs will be license costs. Everything else should be about the same - including personnel costs. In that respect, Microsoft will always lose. These days licenses are the major cost of any upgrade. We were looking at new Exchange servers and the licenses were about 8 times the cost of the hardware.
            Patrick Jones
          • On personnel costs in studies.

            I think you're correct about the differences in TCO being minor. But that's advantage Microsoft. For many companies, especially large companies, the cost of the software is less than the cost of worrying about changing software.

            Look at the price of Office. Would any individual be willing to pay that much for basic software? No, Office pricing is in the context of how much money a corporation considers minor.

            So ties or near ties favor Microsoft.


            On personnel costs in TCO studies.

            In many cases, Linux has replaced Unix. The experienced, expensive Unix personnel are still there.
            I suspect that the personnel costs of Linux can be found higher in a study because of the existing staff, and not because of the software requirements.

            Of course, I shouldn't say this too loudly. I want the prior staff to continue, wouldn't mind a government mandate to assure retraining.
            But it is going to make the Linux costs higher sometimes.
            Anton Philidor