Want to read Slashdot from your mobile device? Point it at m.slashdot.org and keep reading!

 



Forgot your password?
typodupeerror
×
Linux Software

Metcalfe claims Linux Can't Beat Win2000 744

Bruce Inglish writes "InfoWorld Pundit (and inventor of Ethernet) Bob Metcalfe just posted his 99/6/19 column entitled: "Linux's '60s technology, open-sores ideology won't beat W2K, but what will?" in which he predicts that "Linux will fizzle against Windows" and compares the Open Source community to communism and the Back-to-the-Earth Movement. "
This discussion has been archived. No new comments can be posted.

Metcalfe claims Linux Can't Beat Win2000

Comments Filter:
  • by Anonymous Coward
    I don't want to beat a dead horse, but the concept of the GNU GPL is very close to communism. The central theme of communism is community ownership of property/the means of production, and the aim of the GPL is to force community ownership of software and the means to create it (the source code).

    Microsoft-style software distribution is a perfect example of capitalism, in that the software sources are owned by the capitalists (the owners of the software companies), while binaries are distributed for a fee to consumers (users), and the workers (developers, testers, managers, etc.) are paid wages/salaries for producing the software.
  • by Anonymous Coward
    He did invent Ethernet and he has always been one of those one-size-fits-all (dare I say) fanatics about it.

    Problems with bandwidth to home? Ethernet to the home.

    Problems with networking at work? Ethernet, not that nasty Token Ring!

    Problems with that Ethernet at work not performing as well as Token Ring? Faster Ethernet!

    Problems with that faster Ethernet really not doing much better at all than the 16Mb Token that you replaced? Switched Ethernet!

    Cell phone connectivity not what you want? What we need is cellular Ethernet!

    Switched, 100Mb Ethernet not cutting it (and you are about to be fired for replacing all of that Token)? Gigabit Ethernet!

    ATM? No, Ethernet!

    155Mb Token Ring? No, Ethernet!

    Etc., etc.

    I, for one, am a little tired of hearing people like him spout off about stuff they have not been current enough with to have had a reasonable opinion on in twenty years. This reminds me way to much of my time at GM, when I was dealing with legions, I mean thousands upon thousands of old, fat, white guys who thought that the coolest thing (and the thing that the kids would just flip over) would be a 1992 equivalent of a '66 Eldo. These were people who were delighted to get a Park Avenue as a bonus and thought that it was "classy." These were Yanni-loving, spreadsheet-worshipping, soul-dead parasites who though that Glen Miller was a little too "hip" for them. They used words like "wacky" in sentences, and saw no irony. Bob Metcalfe reminds me of sitting in design meetings with people who looked like country club republicans discussing how to get the "inner city rap sensibilities" into design, giving the design tweaks even more leeway to present things that could not be manufactured properly, leading to stuff getting manufactured poorly to keep the style. And the managers got bonuses, and the smart people left, and the shareholders continued their slow burn.

    Wow, wow, wow. He is really of a type that American industry has been better off getting rid of, and I wish that he would just shut up. He hasn't had anything useful to say in years, and too many like-minded managers read his stuff, see the credentials, and bring it up in the next meeting. With me. Forcing me to take time from my busy day to "un-bob" them.

    I can't wait to speak to marketing later this week ...

    And, no, for the culturally impared, I am not slamming people who are a)old, b)fat, c)white, or d)guys. I am a bitter old redneck. I am talking about the product of a coddling corporate culture, second rate state schools in areas outside of the "Old South," the effects of the aggressive middle-class homogenization of culture and mores that makes people afraid to have feelings or (even worse) express any feelings in public, and the entrenched lack of any perceptiveness to style that lies like an asbestos fire blanket over almost all of this country except the South (and the weirdos on the left coast)(and the Texans, but they make Islamic terrorists look stable, so I am not sure if we want to hold them up as an example of anything other than the world's largest open-air loonie bin).

    This is why I went back to Georgia, basically -- I wanted irrational and sometimes downright strange managers that were occasionally hung over, not corporate weasels like Bob. And I can eat grits at my desk here (in fact, the data center has instant grits by the coffee).
  • by Anonymous Coward
    Actually, that is one of the biggest problems in classical marxist theory. You'd want to encourage people to explore new ideas, while at the same time, you giving monetary incentives would screw up the system (after all, economic differences are what you are trying to get rid off).

    The solutions depend on who you ask. Some marxist factions want the design by comittee approach (basically elected officials choosing which projects to go with). Some want to nurture non-monetary incentives, such as using awards, giving extra vacation time or shorter work hours (but do you really want your highest producing people to take more time off?), giving persons who innovate high ranking positions etc. etc. to make people want to innovate even though it doesn't pay off in hard cash.

    You'd still have to convince someone it's worth it later on, to get capital to realize the idea. Under market based marxism, this could very well be in the form of getting companies to invest in your idea, just as in a capitalist setting.

    In my opinion, this would be the most workable solution. It also requires the least change of mindset for most people, and utilize a proven method for maximizing profit (the free market), just under a different political system, and with different ownership of the companies.

    It would also be concievable to have "venture capital funds" just like today, run by people with an economic background, where the goal is to maximize "profit" (allthough profit for a community can come indirectly through higher employment rates or benefits from the product delievered, not only by direct earnings from the company, as under capitalism).

    But actually this problem was one of the reasons that Marx himself warned that development under communism would likely be slower than under capitalism. This is also one of the reasons he warned that communism according to him can't succeed in any country until capitalism is well developed in that country, because introducing communism in an underdeveloped country would only make everone poor rather than give everyone a reasonable degree of wealth.

    The Soviet Union and most of the other "socialist" states demonstrated that very well: many of them started out with good intentions, but after it became clear that they were too poor to be able to provide wealth for everyone, abuse of power became frequent.

    In this regard, the open source movement has actually in many ways been able to do community based development a lot better than Marx thought possible. But unfortunately the two aren't directly comparable, since Marx dealt mostly with "scarcity economics", that is, the economics of goods and resources that are limited, and that will under certain circumstances be scarce.

    While the access to enough developers can compare to acess to a large enough workforce, there is no direct equivalent to scarcity of goods.

    But I do believe that researching the open source movement would be very worthwhile for marxist econimists and researchers, though, since it demonstrate many mechanisms that has been largely "untested" on a large scale in a real world setting until now (especially development of ideas and products not driven by traditional capitalisms "holy quest" for more money).

  • by Anonymous Coward
    Metcalfe's article makes no sense; he claims Linux is 60's technology--and then admits that Windows is 60's technology as well (VMS in particular). Yet he provides no argument as to why the 60's technology Windows is based on should be any better than the 60's technology Linux is based on.

    In fact Linux incorporates significant improvements over the original 60's Unix. Linux is in fact based on System-V, as is Solaris for one thing. It also incorporates the BSD innovations for another. Let's remember that the last time Unix squared off against VMS, VMS lost.

    I'd be more impressed with Metcalfe if he had put some of his intelligence into providing reasons for his opinions. As it stands, there's little substance in his article--only a series of sarcastic insults.

    Too bad.
  • Folks should treat writers like Dr Metcalf like
    they do other folks who post/write flame-bait -
    ignore him.

    For myself, I cancelled my Info World subscription
    a year ago. And I've quit speaking to their
    reporters.

    If enough folks were to do this, they would
    collapse. Or maybe start putting their tabloid
    reporting on supermarket checkout shelves, where
    this form of "journalism" belongs! :-)

    - John
  • by Anonymous Coward
    Scott McNealy is such an expert.

    As a software engineer, as many of the readers of this site are, I laugh at such a proposterous comparison. Every software project has a theoretical "highly optimized", likely error-free state that can be achieved if you greatly limit functionality (i.e. a microwave, a car (yes a car is simple. The job at hand is incredibly straightforward unlike the PR that most people eat up), or a VCR). Ironically as these simple devices start becoming more like a real computer the problems are beginning, and simply blaming Microsoft won't solve the inherent problems in the current state of software engineering. Case in point is my new DVD player : I can't play the "behind the scenes" on (uh...the girlfriends....) "You Have Mail"...it locks on it. And would ya believe that Microsoft didn't even help program it?
  • Mr Metcalfe would probably be the first to admit the original ethernet technology he developed is past it. The problem I have with his view is the notion that new technology is inherently superior to `evolved' technology (meaning old technology which has been regularly updated). This second category would include 10Mb, 100Mb and 1Gb ethernet.

    Oddly, this mindset actually ought to favour Linux over Windows or UNIX. Linux is, after all, a relatively new OS (even if it acts like UNIX), where as Windows and UNIX (BSD or System V) represent the evolution of systems which date back 20 and 30 years, respectively (though only one of those original systems was, shall we say, sound).
  • by Anonymous Coward
    In terms of administration, UNIX (or Linux) has strengths and weaknesses, as does Windows.

    UNIX's greatest strengths are its fairly-standard filesystem layout, and the fact that configuration information is almost always managed via text files which are easily processed with its superb console tools. These make the system very easy for seasoned professionals to manage efficiently, but do nothing for the average user (who simply finds it arcane and unfriendly).

    Windows (NT or DOS-based) has fairly easy-to-use GUI tools for managing the system, but the lack of a solid console environment (even on a local machine) makes complex administration tasks tedious, difficult or impossible.

    There are other issues which have to do with the roots of the operating systems in question. UNIX originated as a time-sharing system, and remains excellent for multi-user systems. For single-user systems, its necessary complexities can make it very difficult for novices to use effectively. Some people even use the root account for day-to-day work! Such use goes against the design of the system, hence it is very easy for a novice to destroy data or make the system unusable.

    Windows, on the other hand, has always been a single-user system, designed in such a way that the single user has complete control, but with safety valves and an easy-to-use help system. As a result, it's easier for novices to handle simple administrative tasks (installing a new application, for example). In a multi-user setting beyond simple file-sharing, however, Windows is essentially useless.

    At the end of the day, I think UNIX is a much more competent server platform, and a superior development platform (especially the FREENIXes), but Windows (NT) is a very good client for things like web-browsing and general office work. These are the things the average user uses a PC for, and Windows is probably a better choice for most of them. For technical users/developers and server administrators (a much smaller group of users), the situation is reversed.
  • by Anonymous Coward on Monday June 21, 1999 @02:46AM (#1841498)
    Actually, leninist "communism" centered around extreme central planning during the socialist stage (transistion to communism). Many other variations of communism tend to be more moderate in that regard. And none support much central planning once communism is reached, and the state has been abolished.

    The main failure of the self proclaimed "communist" and "socialist" states the world has seen so far has been that rather than getting rid of class divisions etc., they have increased them, and enforced an extreme top-down command chain, instead of bottom-up, that is the foundation of marxist political theory.

    But I agree with your conclusion: He doesn't understand what went wrong in the states he see as communist (probably because he mixes them up with "proper" marxist theory on communism, and compare the ideals of that, with the result of the flawed implementations of socialism).

    Sure, there are elements of the open source movement that are close to the marxist ideals of communism, but the movement has hardly anything in common with the self proclaimed "socialist" states.

    If anything, commercial, closed source, software development is what is closest to stalinist economy:

    • You are told everything works great.
    • You aren't allowed to see what really goes on.
    • You can submit suggestions, but they're ignored, unless it is of benefit for the ruling class or the software company to fix it.
    • The official view is that there are no real problems, whenever something is "fixed", it means they have made yet another glorious improvement to something that was already perfect, not fixed a problem.
    • You're the one who gets to pay for their mistakes, either with more work, or by paying for upgrades...

    So, open source may seem like marxism, but then closed source seems like stalinism... If those are the ones I get to choose from, I'd prefer marxism any day...

  • to beat the next one.
  • Ethernet stinks anyway. The only reason it is so popular is that it is cheap and plentiful.
  • I'm getting sick and tired of all these so-called industry pundits sitting on their little soapbox spouting the biggest amount of bullshit ever to be found in the universe. This article was one of the same breed, the same age-old cliche's wrapped up together in a poorly written article. "Unix is old so it must be crap", yeah sure, computers are old too, lets ditch them, they're worthless anyway. "Stallman is a communist". Oh fuck off. The same old cliche repeated over and over again. Can't they come up with something new. The moron talks about Linux as if the only useful application is emacs. What rock has he been under since he invented Ethernet? Must be a whole pile with his "windows will rule the earth" nonsense. All these articles have one thing in common, and that's that they are not written to be analytic pieces discussing something, but piles of crap written solely to get more hits. The crappier the article, the more people will become outraged and run from slashdot to the site, thus gaining a nice amount of hits for the publisher. I don't understand why anyone would want their name above such poorly written articles. It only makes them look like idiots. So the guy founded 3com. The only thing that teaches me is that the founder of 3com is a brainless idiot that's out of touch with reality. I don't understand whats the fun of producing crappy articles to get people pissed. Reminds me of throwing garbage of the top of an apartment building and laughing at the crowd below. Something for little kids, but not reasonably intelligent people. *sigh*
  • by strredwolf ( 532 ) on Monday June 21, 1999 @02:50AM (#1841502) Homepage Journal
    Let me go through this article and do a few items here, eh?

    Why do I think Linux won't kill Windows? Two reasons. The Open Source Movement's ideology is utopian balderdash. You mean that you like your dependence on upgrades which don't fix bugs? Are you sure that Microsoft's products are secure? Sheesh, we already have holes in Win95, Win98, and NT 4! You have to pull patches to lock 'em down and be secure on the net, and yet still not be able to be fast enough to keep up with the continous hits on the Internet. Server software it isn't. It wasn't designed to be server software. I doubt Win2K will be server software. Besides, Linux and the Open Source Initiative is not just theory, it is fact. Alot of security patches go out for Linux and it's utility programs. Linux had a fix for the Pentium FOOF bug within a week, and that's not with Intel's help. Microsoft brushed it off, saying that it's products are for single users only!

    And Linux is 30-year-old technology. Relibable 30 year old technology that's secure, and recoded, reapplied, and rewritten for current hardware. Also, many components such as GTK, GIMP, and IMlib are new software and toolkits, not 30 year old technology.

    A Soviet Linux lies ahead, with successive five-year plans every three. Nope, we haven't had any shootings yet...

    OK, communism is too harsh on Linux. Lenin too harsh on Torvalds. It's actually Socialism, not communism (which the two get so mixed up it's suprizing you even mention it). Socialism, by Karl Marx, dictates that everyone, not the government(that's Communism), owns everything. Marx also states that it will be a slow progression into Socialism. Communism got screwed up by Lenin et al trying to speed things up. Guess what Russia got into now?

    If North America actually went back to the earth, close to 250 million people would die of starvation before you could say agribusiness. When they bring organic fruit to market, you pay extra for small apples with open sores -- the Open Sores Movement. Negative. My local Giant supermarket chain carries organic, "Back to earth" style apples. They're the same size and with no open sores, no defects, no bugs as the ones treated with chemicals. They taste the same too. Try one. They're in the next same-size pile over. Brought in from local farmers. And do some better research next time.

    Stallman's EMACS was brilliant in the 1970s, but today we demand more, specifically Microsoft Word, which can't be written over a weekend, no matter how much Coke you drink. Multinational corporations are themselves technology invented to get big things done, things that sustain us in the complicated modern world. I find alot of users tripping over Word myself here in college, and I suggest to them more control in the form of Corel WordPerfect. Don't you know? WordPerfect been ported all over the place. Macs, Win3.1, Win9x/NT, Solaris, Linux, insert your favorite operating system here.... There's also alot of other good programs out there similar to Word.

    The Open Sores Movement asks us to ignore three decades of innovation. Three decades of security enhancements, TCP/IP implementations (remember, Unix had the first implementation!), web servers (NCSA httpd), ftp servers, and more. Even e-mail was origionally on Unix. And before the Internet, before the first router, there was UUCP. These are innovations.

    NT, now approaching 23x6 availability, is already overpowering Linux. Linux is 24/7. Virtually no maintance needed once properly set up. I have computer labs here that require me to reboot every few days just because NT's about to die. Every few weeks we have to put down a new image on a few NT computers. 23 hours/day, 6 days per week? No, we can't have that in a 24/7, we never close computer lab. It's on all the time or it's dead.

    Let's hope there's something coming soon that's better than both Linux and W2K. What would that be? Java or what? Let's be looking. I'm looking, but all I see is a penguin giving me source code and saying "BBBBWWWWAAAAPPPPP!!!!"

    Bob Metcalfe, you forget your computing history. Back to the university library with you.



    ---
    Spammed? Click here [sputum.com] for free slack on how to fight it!

  • No, that's what some installers do on MacOS, and I hate hate hate it, too. It is intensely annoying because I tend to have my menus set up in a particular way, for a reason, and don't wish to have them disrupted.
    Linux is going to have to deal with people like me, because _I_ am the sort who actually goes and uses linux, gets it working, plays with shell scripts etc. Taking my control of my linux installation out of my hands leaves me with no reason to even bother with it anymore...
  • This is what really ticks me off: that people can decide to target the whole essential concept that I and other techs I know are _desperate_ for. I'm sorry- open source is not negotiable, that's why I release occasional code under GPL and not, say, a BSD license or public domain.
    We are _constantly_ harassed by manifestations of proprietary software, of closedness. _This_ CD-Rom simply thrown away because you can't get it to connect to anything but a Sony MB, _that_ monitor gathering dust because it requires a funky cable to even work, _this_ new compression software upgrade featuring a wonderful new format that causes the previous version to download the whole file and silently delete it on failure to expand it, _that_ web browser rewriting global image file mappings every time it runs. I could go on and on, and I could start to make stuff up, and I'll bet you proprietary software will equal or surpass the most horrible things I can even imagine. After all, it's becoming legal for them to bear no responsibility for anything, and it might even become illegal for you to even _try_ to find out what this stuff is surreptitiously doing- so _what_ is stopping closedware from becoming even more insanely manipulative, destructive and inadequate? Not 'quality': guess what? When people don't have a choice, it's more profitable to turn the screws on them. A free market only exists when people can _legitimately_ jump to something else, when nothing can hold them hostage. The very nature of data makes it a damned good hostage, and all the better when it's proprietary and you're not allowed to deconstruct your _own_ data, only to edit and view it- and the next logical step after reverse engineering is made illegal is to make the reverse engineering of data formats illegal.
    Well, there's an element of truth in a lot of the whining against Linux. It is still awkward, maybe even more awkward than Windows is, there is little chance of making a lot of profit in it, and its mama dresses it funny ;) However, what a lot of people are overlooking is this: WE OWN IT!
    We _OWN_ it. I don't care if it seems clunky, or if it doesn't seem like a classic money-spinner business opportunity. Some of us consider computers and the worldwide networking and connectivity now available to people as _important_ things: things of tangible value in and of themselves regardless of whether the interface is slick or clunky or takes a few days or weeks to master. The problem is this: the power is being centralized in very much the wrong hands- in some ways proprietary software, particularly Microsoft, is similar to the Soviet Union at its most repressive, because the power is all centralized with people you can't vote out or protest to. You're a total peon, a helpless peasant who has to endure whatever the ruling power feels like decreeing, because you have no say in what goes on, and in a very large number of cases you have no recourse, no really feasible alternative, and sometimes you can't even opt out. People are forced to use Windows for pragmatic reasons every day. It's normal to be forced to use Windows for some purpose for which only Windows software is available. Microsoft has put a great deal of effort into, and waged dedicated war towards, ensuring that these non-choices are everywhere- that you can either be a Luddite with a ballpoint pen and legal pad, or you can run Windows: your choice, such as it is.
    This all leverages off proprietary software.
    By contrast, the less ornate Linux may have functional superiority over Windows in many ways, but the really serious difference is the property issue- you _own_ Linux. If you run software, typically you _own_ that as well, which extends to your being free to alter or change it in any way. It doesn't matter if _you_ personally can't code to save your life- if someone makes the new compression program work so that stuff you compress cannot be opened without the recipient purchasing new software themselves, like an 'upgrade virus', then you are free to encourage somebody who _can_ code to revise the program, fix the 'bug' and then fork development to route the course of software development around the developer who is trying to hold people's workflow and data hostage with an uncooperative format. This would happen so naturally, that it is a profound disincentive for open source developers to prey on their customers- and none of it requires that customers all be programmers- it only requires that there be one other programmer out there somewhere who believes people should get to own their own data and not be treated like peasants.
    This runs a hell of a lot deeper than ease of use, my friends- it frankly doesn't matter if linux is ten thousand times harder to use than Windows, _it_ still is the best place to stand if you care about rights to own data and software- if you give a damn about owning any of those bits on the HD, if you bridle at the notion that you're only borrowing it on sufferance from the higher authority which is so good as to permit you to rent bits of cyberspace, arrangements of electrons to store ideas or records or accounts- oop! but mustn't forget, they're not actually _your_ electrons, that arrangement of bits _belongs_ to Microsoft bucko, and you're only borrowing it. If your idea has to be TAKEN AWAY from you by error or as punishment that's just too bad- shoulda written it with a ballpoint if you're so anal as to think that you OWN any of it. Right?
    Linux is _the_ ideal high ground to take a stand against this- and what's really at stake is new meanings for the word 'property'. The central irony is that Linux is incessantly portrayed as communist and freedom-eroding, when in fact it is proprietary software itself that functions like known examples of historical Communism. It's proprietary software when you purchase a program and it only works a certain way and parts of it are broken and you can't fix them. It's proprietary software when you try to upgrade or change software and your data does not communicate properly and you're not allowed to pry into the inner workings of your own data to fix the problem- that's not your place to do so. It's proprietary software that behaves like Communism- but customers, users are the poor bastard peasantry, and the programmers end up as the Politburo, able to twist the knive any way they please, entirely insulated from any concern for the consequences of their actions- with regard to big corporations, there is an additional level of similarity, in that expansionism comes before everything, specifically before any concern with the peasantry, and this is never questioned for a moment. When was the last time Microsoft said, 'Gee, maybe the world might be a better place if we only controlled 3/4 of the users and other vendors got to go around serving the people whom for whatever reason don't like our flavor of computing- ya think?'.
    When people are backed against the wall, they can get noisy and desperate. To a large extent this is precisely what is happening with Linux. If you really care about ownership of your 'virtual property', data, software, OS, then you have no ground to give- you can do things like run MacOS but the _same_ issues constantly arise, until you feel that there's no hiding place anywhere, no way to coexist with this incredible movement taking power out of your hands and turning every computer-hosted property of yours over to central authorities of one sort or another, your rights withering and ebbing away, new laws springing up to punish you if you're fool enough to try and take that power back by reverse-engineering some of those cyber-toys that are Not Yours.
    Then you go with Linux, and hang onto it as a life raft, not so much because it's that phenomenally more refined or more effective than the commercial sector (though with a bit of enthusiastic work, often it becomes a personalized Linux that pleases you), but because you OWN it, and it cannot be taken away from you. It seems a reasonable tradeoff that programmers can't earn so much at it- they can't hold you hostage either, seems fair. You have a hard time understanding what 'right' the programmer classes have to jerk people around for a boost in the ol' profitability. You've done the equivalent of homesteading a cabin off in the boonies, with your own farm and flintlock musket, and you know that when you were given the choice of 'live free or die' you picked 'live free' even if it meant a bit of hard times. Doesn't seem so bad, since the peasantry have kinds of trouble that you don't ever tolerate anymore. It feels a little bit like the ol' Revolutionary War, where those proprietary software Redcoats were busting into your 'house' any time they wanted, charging taxes on your household stuff, taking no responsibility for damages, stealing information out of your house without asking you: and now you don't endure any of that, you just won't stand it, you feel about ready to write some kind of Declaration of Independence...
    And then, in a horrible trainwreck of mingling metaphors, the proprietary software Politburo looks down on you and your Colonial flintlock muskets and "Don't Tread On Me" patriotism, from their wealthy dachas and unchallenged power and Soviet-style expansionism and complete unconcern for the peasantry they're bleeding dry...
    And calls _you_ a Communist.
  • 1) He's writing about Win2000, which is not based on DOS, not win95, which is.

    2) Linux did not write the entire thing from scratch. He adapted Minix, which was itself a mini-copy of AT&T's UNIX.
  • RMS is not the problem. RMS is on of the major creators of the current Free Software movement. Without him, all your IPOs and ESRs would not exist.
  • You say that RMS is a communist. Do you have evidence for this? I don't recall him stating this himself. If he were indeed communist, the GPL would not allow businesses to sell GPL'd code. It does.
  • Every single thing he said in his rebutal was factually correct. Even the Fuckwit part. The definition fits you.

    www.fuckwit.tm [fuckwit.tm] has detailed information on what exactly a fuckwit is and even includes some examples.

  • This is the same guy who has said, for each of the last three years, that the Internet will collapse by the end of the year and that the only solution is to charge per-minute fees for Internet access.

    So far he's batting .000 here. I suspect his annual "the fall of Linux" speech will start to acquire the same patina as his annual "the collapse of the Internet" speech, for the same reason -- Bob simply does not understand how an anarchic network of developers can keep something like the Internet or Linux going. Bob is the classic Cathedral guy in Eric S. Raymond's "Cathedral and Bazarre". The operations of a bazarre frighten him. Ethernet has a classic simplicity reflective of Bob's highly ordered personality. When he sees the Internet or Linux, he sees chaos, and cannot see why it does not collapse under its own weight.

    -E
  • It's not just Linux that Bob 'doesn't get'. It's the Internet as a whole. Linux is just the latest Internet phenomenon to be the target of Bob's bile.

    Bob doesn't like the Internet because it is anarchic, has no central point of control, and seemingly develops every which way without any rhyme or reason. This offends his sensibilities. Ethernet had a classic sensibility which reflects Bob's psyche. The Internet doesn't. Every year he predicts that the Internet will collapse under its own weight by the end of the year. Every year he is proven wrong. But that does not stop him from predicting, every year, that only per-minute charges will save the Internet from collapse.
  • Posted by cmcclure:

    And clueless he is. I wonder how much M$ paid him to write that one? Linux - W2K roadkill? I don't think so. W2K is road kill.
  • Posted by Ungrounded Lightning Rod:

    > Correct me if I'm mistaken here, but it seems to me that most of the innovations in the computer industry in the last 30 years
    > took place on UNIX machines. The Internet, the WWW, the GUI, the mouse, the workstation, the list goes on and on. I don't
    > think Microsoft could continue to "innovate" unless there was a UNIX (or even an Apple) to steal from

    Some of that got done on Unix, but a lot of it was done on other OSes and then ported to Unix. One of the big advantages of Unix was its ability to absorb nearly any valuable innovation and weld it together with all the stuff it already had, creating a more powerful system than whichever from-scratch newbie had the useful feature that was being cloned.

    And the thing that made Unix able to do this was its nearly open-source model. Though the core code WAS quasi-proprietary, Bell Labs' had leaked it big-time - shipping source to universities around the world. Up to about when the System V rewrite (IMHO mainly a move to thrash the code base so the ownership would no longer be vulnerable to legal attack) the common model for porting was:
    - Port your bootleg source to the new platform.
    - Call up Bell Labs and buy a license. (Terms were reasonable and they held no grudge.)
    - Sell your product, with Unix on it.

    This same relatively free availability of source made it easy to add or port anything you needed into the kernel - and created a large community of authors and porters.

    Which makes it funny to see Linux folk calling Microsoft "The Borg". Oughta call 'em "Borg, the new generation." B-)

    (Does that make Linux "Borg, the Gnu Generation?")
  • Comment removed based on user account deletion
  • up to now, the word "troll" was used for usenet articles and comment systems like this one. infoworld, however, should be awarded a medal for making the word apply to articles in (online) magazines.

    it's really getting obvious by now. write some crap about Linux, get a) praise and more ad money from M$ and b) the link posted on /. which yields you a 5-digit number of hits quickly, which again pleases your ad people.

    rob, I request you don't link to those troll articles anymore. doing so just creates more of them.
  • Some of the best ideas are timeless. Another post cited cars and guns. You find something that works and you stick with it until there is something better. The age of an idea has nothing to do with it's validity. All that matters is whether it works. Linux corrects the Windows mistake. When Windows 2000 comes out it'll take 2000 flushes before it goes around the bowl and down the hole.
  • ...check this guy's user info. He's an anti-Linux troll. Don't get too worked up, folks; save your energy for another day.

    --
    Get your fresh, hot kernels right here [kernel.org]!
  • But stop whining about all your Windows buddies suddenly using Linux.

    Boy, have you hit the nail on the head. I hadn't really thought about it much, but this makes perfect sense, as far as why I've seen an increase lately in bitter anti-Linux trolls.

    These guys feel threatened, because their friends are now using something they know nothing about.

    When everyone was using Windows, they felt safe; now, there's some uncertainty, so they lash out at The Penguin.

    --
    Get your fresh, hot kernels right here [kernel.org]!

  • Actually, Unix wasn't the centrepoint OS in the early days of the ARPANET. Most machines in those days were 'frames and PDP-10s and 11s running DEC's OSes. Unix and the ARPANET didn't really get crossed until 1982, when Berkeley introduced their TCP/IP stack code in 4.2BSD.

    -lee
  • I think the thing missed is rate at which Linux has improved. In 91' Microsoft has a foot hold in the computer reveloution. Linux had a couple thousand lines of code that no one wanted. Years would go by befour small amounts of people started to notice Linux. When Windows95 came out Linux was starting to shape up. ( this is when I came in ). Linux, even with its poor driver support still managed to be installable on most PC's. In the past 4 years Linux has made leaps and bounds. The user base has gone from thousands to millions. More programmers started to take notice of Linux, and liked it over windows. Hardware manufactures are now helping to develope drivers for Linux. The worlds best Game company has ported Doom, doom2, Quake, Quake2. Now we are getting Quake3 in a box with Linux stamped on the system requirements.
    I myself do not know of a time when the user base of Linux has gotten smaller. I think It is unfair for anyone to say that Linux is a fad. We are living in a exciting time of Linux history. There are things on the horizion to get very excited about right now. X-windows 4.0, Better support for highend hardware, Common desktop enviroments, and increasing stablity. I don't want to forget the attention of companys like Oracle, IBM, Compaq, and DELL.

    The developers that contribute to all the Linux projects should be given a round of applause. You have done great work and I thank you.

    The reason I don't think Linux will be dead anytime soon is: There will allways be people that want to be able to control every part of there computer ( or soon device ). For these people a MS GUI does give the power to control. Also the idea of clear text. Meaning that you can read and edit any config file. Or set up powerful scripts to configure and maintain the system. Things people in redmond don't understand why you would want to. But its the fact that they don't understand the user can do anything concept that will defeat them.

    As for the communist remark. It would seem to me that a communist operating system might do things like track your documents with id numbers that they would not want you to know about. Or enable features that put you at risk then be told that they are on by default for your own good. A communist Operating system might even tell you the lie that you need the newest version inorder to be productive. Or Keep secrets from their public to maintain the illusion that the OS is safe from your competitors or criminals.

    My opion is that this guy is wronge. Linux will not die. As long as anyone can download it for free it will never die.

    Personnaly..... Between you and me.... I think he's pissed they did not put him in the movie "Pirates of white bread vally."
  • I agree. I think the other guy was comparing Dictatorship and Democracy. These things are different from Communism and Capitalism.

    I do disagree that the means of production is the source code. The means of production is the computer and a compiler. The source code is the blueprint, the design. The difference between traditional production and software production is that many people have access to the means of production. In traditional production very few own/have access to the machinery and plants to produce items.

    In the industrial age there were limited means of production. The people that controled those means became powerful in both capitalistic and communistic countries. As the computer age continues almost everybody has access to the means of production. This is why the community idea is able to produce a viable product in the computer age.

    There are many differences between the community of open source software and communism, but some of the same basic principles are similar.

  • Voluntary Socialism. That sounds a little better. I was going to mention socialism in my earlier post, but it's been many years since I've studied this stuff and didn't want to say something that was wrong.

    Yet I still don't think that is right. OSS projects are accomplished through a community of people. And it is often volunteer work. But we aren't trying to run a country here or do many of the other things that governments have to do. We are just working together on a common goal. Much like a community working together to to build a shelter, or fixing a community building up, or sand bagging a broken levee. Sure they could have just raised the money and had Moe's construction do the work for them. But why, when they can do it themselves for less money and better quality? Yet these people aren't accused of being communists or socialists. Why? Because they aren't. Doing these things for the community have little to do with how the community is governed.

  • Look at you people! Someone writes an article stating some painfully obvious things about Linux (all you need to do is read the comments from many /. readers to confirm many of his comments) and many of you start running around like red ants attacking him any way you can.

  • Emacs is a text editor and lisp interpreter. When you compile Emacs you get a base set of lisp which helps you with day-to-day editing. If you wish, you can download additional lisp programs to browse the web, read mail, read news, write books, etc. But none of these is required to use Emacs, nor does Emacs perform worse without them.

    One can turn Microsoft Word into a (horribly slow and unstable) web browser or news reader via VBA. But you laugh at the thought.

    This is the difference between Microsoft's technologies and the kind that survive.
  • Which kernel developer is that? The guy I'm thinking of doesn't seem to have any code in the kernel tree... as far as grep can tell.

    /usr/src/linux/CREDITS:
    N: Eric S. Raymond
    E: esr@thyrsus.com
    W: http://www.tuxedo.org/~esr/
    D: terminfo master file maintainer
    D: Editor: Installation HOWTO, Distributions HOWTO, XFree86 HOWTO
    D: Author: fetchmail, Emacs VC mode, Emacs GUD mode
    S: 6 Karen Drive
    S: Malvern, Pennsylvania 19355
    S: USA
  • Cool! This sounds great! Where do I download the kernel code?
  • I think you missed the point. Emacs only does everything if you configure it that way. In fact, you can have all the lisp installed you want, but none of it is sitting in RAM unless you load it.

    I use vi all day for tasks like system administration or a quick e-mail, but I use XEmacs to write code--it's just a better environment. If I want to write a book, I can start LaTeX mode and start writing. I might even start GNUS to read news. I can then close these buffers and keep my code open. Emacs doesn't crash.

    If I want to write a quick letter to be printed and mailed I'll use a word processor, like AbiWord [abisource.com].

  • You sure do a lot of considering. Have you considered _using_ BeOS? BeOS isn't "internet aware?" You "assume" the web browser isn't "up to snuff" with Netscape's? You're not a BeOS expert like you probably don't study quantum mechanics on the toilet.
  • What a detailed and eloquent response. Unfortunately I still don't understand what you're problem with BeOS is. You claim BeOS lacks sufficient technology to be of any use to you, but you've never used it? And to defend the point you never had, you just shout "moron" and run away?

    I think you'll have to do better than that.
  • Okay, maybe I missed the memo, but when did linux become solely about world domination and success. I think it happened the minute the media started look at it. I think the concept of "We're writing this operating system because it is interesting and useful to us" doesn't make good press.

    So linux was around long before anyone had the thought "I'll install linux instead of DOS/Windows/OS2/BeOS" and its continued success has nothing to do with people thinking that.
  • It's kind of interesting that this post is marked up as "insightful". Fact is that Win95 uses DOS (IO.SYS) as pretty much a boot loader only. It's like judging Linux on LILO.
    Go read schulman's book where he debunked this bit of propaganda. NT was something new; 95 is still DOS in a clown suit.
  • I agree with you wholeheartedly. Also, someone should tell Mr. Metcalfe that Win9x and NT have slowly and surreptitiously been borrowing ideas from UNIX for many years. Why? Because the guys who developed UNIX put a lot of careful thought into it, especially to making it simple and powerful. Oh, and there's that TCP/IP thing too... :)
  • Gee, it's got a BASH shell like UNIX.. It's got a Posix API like UNIX. If it looks like UNIX, and smells like UNIX... well, it's at least pretty closely related. :)
  • Heh. Exactly. Just like VMS before it, from which NT draws a fair share of its internal architecture (by proxy - several VMS core team member, Dave Cutler for one, worked on NT) tried to slay Unix, and failed. There have been so many who claimed they were the David to Unix's Goliath. I think the day that Unix will be replaced is still a fair ways off.
  • Why are you screwing with Xi's X server? I tried it a ways back (v2.1) and it hung just about every time I switched away from a running X session. I've heard the same problem with more recent versions. Before you give up, try XFree86 and see how it works fot you. Oh, and have you tried the latest pcmcia-cs release? I think 3.0.12 is current as of now, so if you're not using the latest version, try it. If you get the same problem, try to trace it down as best you can, and report it as a bug. That's part of what Linux is all about - tracing down the problems so we can eliminate them.
  • I know it was used as an ASIC-type deal in a few Intel-made devices (similar to how Intel's i960 RISC chip has oft been used). Hoever, I don't know of any actual consumer computer systems that used it as the CPU.
  • I would consider sendmail to meet your criteria of "better job, support, clear instructions".

    I wouldn't wish sendmail.cf and the bat-book on my worst enemy.

  • Oh, dear. Wasn't Metcalfe the guy who predicted the collapse of the Internet? If he is, I wonder why he still has credibility with anyone. Oh, well.

    Yeah, that's him. So maybe this prediction is a really good thing. If it's half as accurate as his collapse of the internet prediction then maybe it means that Linux will be the dominant OS within a few years.

  • I've not read Metcalfe before( and won't ever
    again if this article is indicative of his
    writing ) but it seems pretty obvious that this
    is a man who feels threatened by Linux and a
    software model that he just doesn't understand.
    This article is completely barren of any
    organized argument and is simply a knee-jerk bile dump. Let
    it go folks. Don't fill his Inbox with flames or
    even mail intended to gently "inform" him about
    Linux. He is either beyond help or hoping to
    produce an email /. effect which will provide
    him with more troll material.
  • Below is a letter I just sent to Mr. Metcalfe. I didn't mention how I personally know his "ultimate boss", the CEO and president of IDG, Kelly Conlin. :)

    To: metcalfe@idg.net
    CC: letters@infoworld.com

    Mr. Metcalfe:

    I can only assume your latest article (http://www.infoworld.com/articles/op/xml/990621op metcalfe.xml) was written to try and get a reaction from readers. I am sure that you have done that. After the article being reported on Slashdot.org, I'm sure that you have gotten a _lot_ of reactions, too, ranging in length and "quality".

    Let's put aside any beliefs I may have on the future of Linux for the moment (for the record, I give it more credit than you, but less than some "fanatics"). You still have glaring errors in your article, which, sadly, are indicative of poor journalism.

    First of all, it is true that some members of the Linux community disagree on how to pronounce Linux, since it was a name given to the product by someone other than Linus Torvalds himself. However, make no mistake that Linus pronounces the first syllable of _his name_ with a strong e ("leenus"), but pronounces _Linux_ with a soft i ("lihnucks"). Perhaps a minor error in your article, but something that is blatantly wrong.

    Secondly, you misunderstand EMACS. I am not a user of EMACS myself, as I am not accustomed to it and it does not currently fit my needs. However, many programmers find that they are much more productive using EMACS than they could be using a tool such as Microsoft's Visual Studio. EMACS' real strength is as a highly customizable and programmable development environment. It is not nearly as suited to word processing as you seem to imply that it should be, although some users do choose at their own discretion to use it for those purposes. How often have you seen Microsoft Word used work on the development of an operating system or other large C/C++ project?

    Lastly, you article clearly is intended to be incilliatory, instead of journalism. You use unnecessary derogatory terms such as the "Open Sores Movement" and "a notch above Luddism". You imply that open source software, such as EMACS, is normally written in a weekend with a lot of caffeine; the truth couldn't be more different. This is not the type of journalism I would expect from a magazine like InfoWorld, even for as a personal commentary in the magazine.

    We appreciate the many contributions you have made over the years via 3COM and Ethernet. However, this latest article accomplishes nothing productive.

    --
    Bob Bell | Compaq Computer Corp.
    Software Engineer | 110 Spit Brook Rd - ZKO3-3U/22
    TruCluster Group | Nashua, NH 03062-2698
    (E-Mail omitted) | (Phone omitted)

    (The views expressed are my own, and do not necessary reflect those of my employer)
  • I'm having trouble believing that this article is real. It strikes me as odd that they would print an article so unprofessionally written.

    It is, however, partially right on one of its points: the bit about ignoring 30 years of innovation. Not in terms of under-the-hood stuff, mind you; Linux is reasonably up-to-date with that (and with some of the latest things, like XFS and the others, it will be up-to-date in even more).

    But it lags sorely behind in interface. The default command-line does ignore 30 years, and to be honest I have difficulty believing that any operating system still carries such an archaism. X ignores 15 years, and is marginally better (I'll admit that I'm impressed with what the Gnome and KDE projects are actually managing to do with it) but it falls into more than a few traps (which can be expected; even at the time of X's invention superior GUI's, such as NeWS, were available, but were scoffed at for some reason). More powerful interfaces have been around for years. I'll be among the first to check Berlin out (once it's in a usable state) for this very reason.

    But hey, I'm a Mac and Linux user; I'd been spoiled by usable interfaces for years before switching to Linux. And if I do anything more I'll head off into a flamebait rant (which I honestly don't intend to do), so I'll shut up about it here.

    Regardless, the article is for the most part just unprofessional FUD, and not something to be taken particularly seriously.
  • This can't be a real article. Why?

    1) It's unlike InfoWorld to print such an unprofessionally-written, poorly-researched piece.
    2) It doesn't seem right that the inventor of Ethernet, of all people, would tout "23/6 availability" as a good thing. This guy knows that computers can do better than that (and often do; even a properly-maintained MacOS machine can outdo NT in stability and security).
    3) When it comes to "ignoring 30 years of innovation" Metcalfe may have a point where interface is concerned (I'm impressed with the strides Gnome and KDE have made in this area, but one can only do so much with such comparatively ancient technologies as X and the command-line) but he doesn't seem to get the idea that Linux is quite up-to-date in most areas, and with the recent Open-Sourcing of such things as XFS it'll be catching up in even more stuff.
    4) Going back to the lack of professionalism in the article: do you really think he'd do that? One doesn't stay in his position for as long as he has by being the sort of person who'd write that.
    5) Win2K doesn't have a chance. Linux, BeOS, and OSX will all stomp it flat (especially since two of them are out already, and the third, OSX, will likely be at least a year before Win2K ever sees a store shelf, knowing Microsoft's punctuality or lack thereof). What happens after that? I don't know. I hope the three can coexist harmoniously (I know it's possible) but I doubt that will happen.
  • Bob Metcalfe sounds like one of those National Enquirer prophets: "I correctly predicted that internet stocks would collapse! That proves my track record, now I predict the demise of Linux". Substitute the collapse of internet stocks with the assasination of John Lennon and the demise of Linux with the second coming of Elvis and its virtually identical with supermarket tabloid predictions and backed up by just a valid an analysis.

    His first premise: The Open Source Movement's ideology is utopian balderdash. Alright, maybe Richard Stallman is a bit over the top with some of his talk, but he's not the entire Open Source movement. There are other advocates, such as Bruce Perens, who have done work in getting existing companies such as Netscape or Apple to at least test the Open Source waters. Then he goes on to take a Linux Torvalds' quote out of context as proof that the Open Source community is a band of raving pinko commies. Oh, scratch that, it was just hyperbole. However, if Open Source succeeds a quarter billion people will die or some other such drivel. Oh, and the only means of writing a document under Linux is with EMACS doncha know! I guess Applix, Word Perfect etc. are just communist propoganda and don't really exist.

    The second premise: Linux is 30 year old technology and as such is senile. It's an interesting sentiment, but not correct. Linux is built up from concepts that are 30 some years old but only because technology often builds up on the past. Pre-emptive multitasking and protected memory are just good ideas, no reason to throw them out. In the mean time the kernel guru's have added in multi threading, multi processor support as well as support for late nineties hardware. On top of that there are a number of decent GUI overlays that can make normal day to day stuff just as easy as it is under NT (as well as some of the more complext administrative tasks as well). NT is no different in that regards, its really built up from thirty year old technology and doesn't really offer any new features. Linux performs well now. If Microsoft gets its act together maybe NT will work well tomorrow. Maybe. If given the choice between shipping something that doesn't works but people will buy or shipping something that works as advertised a little bit late guess which one will win?

    In a way this is the biggest evidence that Linux (as well as other operating systems) is relevant. If Microsoft was truly indominatible they could delay their operating system release until it actually worked and not worry about releasing expensive (to the consumer) bug fix patches as Windows 98 or Windows 99 etc.

    Is Linux going to kill Windows? No, I don't think so. I do think that things will be a lot closer to how they were in the eighties though. Windows will have the majority, but more like a 60 or 70% majority overall. A bit more on the desktop and a bit less in the server space. Linux and other operating systems will have that 30 to 40% market share segment. It won't dominate but will ensure that applications are ported to capture that segment of the market.

    Old gurus never die, their opinions just (in some cases) become obsolete or bought and paid for.
  • by edgy ( 5399 ) on Monday June 21, 1999 @03:00AM (#1841612)

    Well, you have to remember just how fast Linux is progressing. It's a much different beast than it even was a few months ago. There is far more support from game companies and 3d card vendors, there is far more multimedia support than before.

    As a matter of fact, the pieces are coming together more and more for Linux on the desktop. Sure, it's not here today for a lot of uses, but who's to say that the momentum behind Linux isn't still building? I don't think we're anywhere near the point of inflection.

    As far as fragmentation, I don't think users really see it as a problem. As long as something is branded Linux and will run all the neat and nifty software, users appreciate not being told what's best, and getting to decide for themselves. A lot of people coming from the Microsoft camp have exactly this problem with MS products.

    Linux may not be there on the desktop today, but the Linux of today is not the Linux of tomorrow.
  • Hmmm,

    Well for those who know, ethernet one out over token ring for one major reason. Price. Ethernet chips are cheap, like pennys while token ring costs 10-20 dollars a chip. Big difference. Linux is cheap and it is more scalable. People are getting tired of having to buy 5-10 NT servers at god knows how much a piece to do one simple project. It comes down to dollars. Once linux takes the server market it will work its way into the desktop market. The issue is will companies like redhat and caldera destroy the market by setting up contracts with everyone and becoming the defacto standard. I don't mind redhat. I like debian better but I feel the one getting the press to the masses is redhat. If anything hurts linux it will be redhat and caldera, not win2k.
    IMHO
  • by Mr Z ( 6791 )

    While RMS may claim to not be a communist and not be anti-profit, he certainly does a tremendous job convincing people otherwise.

    It's too bad that many in the world view Linux as the latest way for college kids to hold their ideological protests. Rather than marching around with "Make Love, Not War" signs, we're painted as marching around with "Boot Linux, Not Windows" signs.

    And let us not forget the Nixon-esq "silent majority" of happy Windows users.

    *sigh*

    It's all too creepy. Fortunately, I doubt the Linux movement's leaders are going to start keeling over from drug abuse. :-) . Also, given that Linux users AREN'T just your average hippies and freaks, but rather are extremely pragmatic people trying to get things done (how come people always forget that tidbit), I think Linux has alot more chance to survive.

    That bit about the Open Sores movement was fairly clever. I'm sure somebody's going to get alot of toasty flames that can be hand picked with asbestos oven mitts for next week's column: Linux: Where The Zealots Go Today.

    --Joe

    --
  • Something tells me that his article makes alot of use of the fine art of sarcasm (see the bit about 23x6 reliability and how NT overpowers Linux for this reason). But, for sarcasm to work it needs to have a kernel of truth (pun intended) at its core.

    Wake up. Smell the coffee. And everyone, let's start giving Linux the image it deserves: Not the OS for the techno-nerds, but rather the OS for people that actually need to do something, rather than watch the nifty 3-D Accelerated, alpha blended, animated hourglass cursor that comes with the latest "Plus!" pack for Windows.

    --Joe

    --
  • With 8,203,248,398,439,298,283 lines of code and counting (or whatever the actual number is). W2K is the Jabba the Hutt of Operating Systems. It will be crushed by its own weight.
  • I didn't say it was perfect. I certainly won't claim it's cutting edge. And as a desktop, Linux is definitely far from cutting edge.

    I maintain that as a server, it works. I have watched IIS have to be rebooted because one didn't have the magic tool locally installed on their desktop to control it. I have watched exchange fall over again and again and again, while sendmail boxes set up by people even LESS clueful about unix chugged along merrily. I watched the exchange client corrupt mail folders regularly, and took the calls, no sorry sir, there's nothing you can do, did you back them up, hello? hello?

    The fantabulous gizmonic flying contraptazoid of NT may sure have lots of knobs and dials, but when you shove it off a cliff, it drops like a stone. Good for show.

    I am very glad that Win2K works for you. NT's track record has not been so kind to me or the MCSE's who called me for support. And the story has been repeated time and again with others.

    I hate NT. I hate Unix, and by extension, Linux. I just haven't been driven to the point of quitting because I couldn't fix Unix problems.

    Really I should get out of the computer business, but it's the only thing I'm any good at.

  • It's kind of interesting that this post is marked up as "insightful". Fact is that Win95 uses DOS (IO.SYS) as pretty much a boot loader only. It's like judging Linux on LILO.
    --
  • Yes, Win2K will run on your machine.

    Maybe. It looks like most pre-Pentium II systems have been dropped from the Win2000 Hardware Compatiblity List, although the older drivers still seem to be on the beta cD. (Of course the Windows HCL has about as much value as the Linux Hardware Compatiblity How-To -- you really have to try it and see in either case.)

    That having been said, Win2Kb3 is chugging along on my P5-133, 112MB RAM, and is noticably quicker than NT4 on the same box. At least from a user/GUI standpoint, Linux+KDE is much slower than either NT4 or NT5 beta.

    --

  • Your post sounds like wishful thinking. I would expect Windows 2000 to ship in the fall of 1999, and if you've been seeing a little anti-Linux FUD here and there, the explosion is really going to hit when W2000 finally ships. So, stock up on the ammo, Linux advocates.

    However, if you are actually responsible for IT decisions (and not just chatting on /.), 2001 is a more realistic time frame. Migrating a user base from NT4/Novell/whatever to either Windows 2000 or Linux is not a simple job. The rational types are going to wait to see how things pan out, or use a little of both, depending on the job.


    --

  • Is he the guy that ate his words about the Internet collapsing. (I mean literally, someone who predicted that chewed up their article and swollowed it.)
    --

  • Not to get into the holy war too much, but Win2K has lots of 'scriptablity' built in. I've only seen the VBScript/JavaScript stuff, but I would expect Perl to follow.

    Also, do you really think Linux distributions are "permanently innoculated against viruses"? I would suspect that Win2K and Linux are about equally "inoculated". (=resistant to system viri, not as resistant to user-space viri like Melissa or worm.explore).
    --
  • A large group of idiots is smarter than a small group of idiots

    Note in this case, the large group of Idiots are Windows users. The latest horrible "virus", Worm.Explorer (which you might guess is actually a worm), comes in e-mail with a message that basically says "Hey shithead, run this executable." When the shitheads have Linux on their computers, vulnerablity will be pretty much the same.
    --

  • NT has always had a CLI that you could do some routine Admin stuff from (register users, file perms, etc). In fact, certain things can *only* be done from the command line.

    As for hardware/network configuration, that's probably always going to always be GUI-based. (Which sucks because the mouse port on one of these switch boxes just croaked.)
    --

  • I've read Shulman's book. I think his argument is that 95 is *Windows for Workgroups 3.11* in a clown suit. (This was when MS was pushing 95 as "all new".)

    In short, the system boots in real mode DOS, and potentially (but ususally not nowdays) loads some drivers. WIN.COM throws the system in protected mode, and virtualized DOS is only used for legacy driver support and one or two other minor things.
    --
  • It doesn't seem right that the inventor of Ethernet, of all people, would tout "23/6 availability" as a good thing

    I think you and about 100 other people on this thread missed the irony in this statement. He was saying that NT has alot of momentum despite *not* being all that reliable.
    --
  • Ever wonder why Microsoft run their website on NT 3.51?

    Netcraft.com says:

    www.microsoft.com is running Microsoft-IIS/5.0 on Windows NT5 beta

    !
    --

  • Good point, except you don't need all of that stuff to run a web server or DNS server or file server or firewall (which are probably Linux's primary production uses right now).

    If people stopped looking at this like some epic battle of the Gods, and thought rationally about what to use where and when, some of this would become more clear. Innovation is nice, but it usually comes at the price of speed or stablity.
    --
  • It seems this is another article that offers nothing to either side of the linux argument. Besides being vague to the point of using obtuse analogies, Metcalfe doesn't offer any reason why Linux really sucks. He is like a cow, eaitng grass, passing gas all day. (How's that for an analogy?)

    Those who do not want Linux to win, will be silent. They will not help us by pointing out Linux's flaws. Those who want attention will either tout or slander Linux. These people are useless and should be ignored by all. Those who are proponents of Linux will point out flaws and nitpick, showing us where Linux needs attention, these people need to be heeded.
  • by swb ( 14022 ) on Monday June 21, 1999 @03:46AM (#1841684)
    Don't confuse the method of distribution (cash money sales vs. free downloads) with the overall philosophy guiding the two software models.

    Open Source is much more like capitalism than marxism -- ideas that are strong gain support based upon their inherent strength, not because XYZ supports it. If enough people have an interest in some feature (and it doesn't seem to take too many), that feature seems to get added and supported. The guiding goal is increasing functionality, just as capitalism has the goal of increasing capital. And sometimes it's the the detriment of the overall usefulness of the product (eg. lots of features vs. poor documentation).

    Closed source is much more Fascist/Stalinist -- an idea is pushed by the marketing/PR propaganda mill (think of a 70-foot Stalin on the wall, and now think of an MS billboard). Attempts to question the technology/leadership/wisdom of closed source vendors and you are quickly banished/ignored. (How many times have you been kicked off of a vendor forum for complaining about bugs?). The guiding prinicipal is the glorification of the state/company. Software functionality is only added to the extent it furthers glorification of the software company, otherwise it is not added. Just as in a closed economy, capital expansion is only taken to the extent that it serves the needs of the state.

  • "To the Editor:

    Bob Metcalfe's "Linux's '60s technology, open-sores ideology won't beat W2K, but what will?" should be relegated to the same garbage heap as the opinions of those who hold that tobacco smoking isn't hazardous to your health or that the Holocaust never really occured. He uses the same circular reasoning and knee-jerk Fear, Uncertainty and Doubt propaganda disguised as logic as any hyperconservative reactionary. Metcalfe's two main points are never really backed up in any useful way.

    His assertion that "the Open Source Movement's ideology is utopian balderdash" is ostensibly backed by two flimsy metaphors. He starts by comparing Open Source Software to Communism and then the Back-to-the-Earth Movement and then making statements that attempt at debunking those two philisophical ideologies, which is intended to sabotage OSS's credibility. But where's the link between OSS and Communism, for example? Because they're both shamelessly idealistic? Are we to believe that just because Communists also had an idea that was different for the time and had a good go but failed at making it work out, OSS should fail as well? Or is it that the Open source community isn't making any money, just as Communism attempts in a different way to abolish class distinctions? Does the quality of a product come from the fact that the authors profited from it? Obviously not; the proof is in the statistics. Just browse the Web and look for Linux server uptime stats vs. those of NT.

    Metcalf's other main assertion, that "Linux is 30-year-old technology," also implies fallacious reasoning. He even earlier counterdicts it himself, saying decent software "... can't be written in a weekend" How does Linux ask us to ignore "three decades of innovation?' It's not as if the Unix concept was implimented once and then abandoned, and now the Linux community is trying to emulate a product from thirty years ago. Besides, do we abandon our automobiles because they're based on 115 year-old technology? (As an aside: So many Microsoft backers use the term 'innovation' when they really should be saying "proprietary standards leveraged by an undue market share used to further eliminate competetion.") Then Metcalf goes on to demolish his own point by openly saying that "W2K is software also from the distant past -- VAX/VMS for Windows," Why the hell should we keep on reading at this point? Got me.

    I'm not saying that my arguments above imply that there will be a Linux firestorm tomorrow that will put Microsoft out of business. All I'm trying to say is that an op-ed piece has to do a little more than express an opinion and pad the rest of the article out with rhetorical bunk. If this sort of vapid tripe passes as publishable material at your mag, may I offer my own services as an author? Benefit to you: rational, well thought-out opinions, that, while not always true, at least advance the public debate. Benefit to me: the opportunity to be inebriated at work and still do a better job than the last guy."

    cygnus
  • I'm sorry to be so blatantly off-topic, but your closing remark reminded me of a snippet of dialogue from the TV show "Newsradio."

    Dave Nelson: Have you ever heard the expression "You can catch more flies with honey than with vinegar?"

    Bill McNeal: No. Have you ever heard the expression "Only a hillbilly sits around thinking up the best way to catch flies?"

    Well, I thought it was funny...
  • by evilpenguin ( 18720 ) on Monday June 21, 1999 @03:04AM (#1841713)
    Oh, dear. Wasn't Metcalfe the guy who predicted the collapse of the Internet? If he is, I wonder why he still has credibility with anyone. Oh, well.

    The best thing to do with something like this (vague, unsupported pronouncements) is probably to ignore it. Personally, I'm very tired of loudly justifying myself and Linux. I intend to just continue doing what I do now -- getting amazing work done more quickly than any of my cohorts who use another OS that shall remain nameless...
  • Metcalfe invented Ethernet. And then what? Maybe the guy is a genius or whatever you want to call him, but even geniuses can be wrong. After all, they're human beings just like you and me.

    This interview sounds like the guy is bitter at something or someone. I mean, ethernet is a PARC invention, isn't it? I seriously doubt he INVENTED it ALL BY HIMSELF. And not getting any royalties on it makes him jealous of Gates for being outsmarted on that one... I dunno.

    But what I see through this article is that he's against the fact that some talented people actually *do* something without a dollar-hunger in their minds. He's against people who *want* to improve things without charging for bugfixes.

    Comparing the Open source Community to Communism makes me smile. Maybe software is the only domain in which *communism* actually works, and works well. I don't see it as evil... Of course, it goes against the allmighty capitalism, but for what capitalism has done to the planet, I'd rather go with the Open Source Community summer camps. :)

    And, for being long today, I'd like to add that all those hippie ideas Metcalfe seems to be against are at the origin of concepts such as *ecology*. Then again, the allmighty American capitalists are the biggest polluters. I find it quite a shameful analogy to BAD software you have to pay a lot for.

    Bah, I'm gonna make plenty of ennemies anyway, so make it quick...



  • This quote from the article stuck in my mind: "The Open Sores Movement asks us to ignore three decades of innovation. "

    Correct me if I'm mistaken here, but it seems to me that most of the innovations in the computer industry in the last 30 years took place on UNIX machines. The Internet, the WWW, the GUI, the mouse, the workstation, the list goes on and on. I don't think Microsoft could continue to "innovate" unless there was a UNIX (or even an Apple) to steal from.

    Has Microsoft innovated anything? Creating proprietary versions of something that already exists doesn't count as an innovation.
  • Except that win95 is still dos-based, so it is _also_ 70's. BeOS would be 90's tech.
  • While this is a neat idea, it isn't too practical. Rob might instead just put a disclaimer, like:

    While Metcalfe has said some snarky flamebait bits, mindlessley counterflaming him is counterproductive. Be pursuasive, not a thug, if you're going to write to him directly.

    You catch more flies with honey than vinegar.
  • Hmm great t-shirt for copyleft, though accually I would suggest that on one the front it says make love not war (with a slashdot logo below that)
    then on the back it says make vmlinuz, not windows
    Hey is the t-shirt design contest still going on, I'm not artist so I couldn't really draw that, but still a great idea!
  • Sorry. forgot to include my explination, was going to say that I think we should have the make love not war on the front because most people wouldn't automatically make the connection. Even if they knew about linux, seeing just make vmlinux not windows might be interesting but wouldn't automatically force a connection.
  • Comment removed based on user account deletion
  • Comment removed based on user account deletion
  • Comment removed based on user account deletion
  • Why do I think Linux won't kill Windows? Two reasons. The Open Source Movement's ideology is utopian balderdash. And Linux is 30-year-old technology.

    And he's contrasting this to Windows 2000? Ummmm....waitaminit...W2K is just NT5.

    Windows NT was originally written by a bunch of guys from DEC who wrote VMS. Hence, NT can be said to have derived much from VMS, which is well over 20 years old (here's some history [montagar.com]). What does that mean for NT?

    Clearly, that age doesn't mean a damn thing. Newer does not necessarily mean better. I'd rather have an OS that has evolved from decades of trial and error than something just out of Redmond, and I'm sure many others would as well.

    I've been running Linux intermittently for several years (and exclusively for the last 6 months). I've also run windows 95/98, and NT4, and a beta of W2K when it was still NT5. NT5 is a pretty cool OS -- the relative (to win98) reliability and security of NT, and directx for games--yay.

    This seems to be where everything is headed currently -- what used to be solely server/workstation operating systems are now becoming gaming platforms. NT is adopting directx6; Unix is getting XFree86 4.0 with all sorts of cool additions. This is A Good Thing (tm), which you should acknowledge even if you hate NT.

    Curiously, the same reason that I used Windows is the reason I switched from Windows to Linux. Linux support for hardware used to suck. My TV card didn't work, there was no 3d acceleration (for my TNT), the games sucked. So, I used Windows and put up with the occasional reboot. Eventually, I got fed up with Windows. I hated having to run Exceed to be able to access some of the programs I needed for classes (Matlab, Maple, LaTeX, etc.). So I installed Linux and dealt with the lack of good hardware and game support.

    I can get my TV card working with a 2.2 series kernel (which I still haven't gotten to work without breaking AFS, which kind of defeats the whole purpose), and now with NVidia releasing open source drivers, I don't have to worry about 3D acceleration. I really don't have a single qualm about not running Windows...the only games I play are Q/Q2/Q3 anyhow.

    Simply put, I have no need for Windows. I haven't booted up Windows in months (I'm pretty sure mucking around with VMWare killed it anyhow), and if I do need it for anything (say, if I buy a digital camera and need to get the pictures off it), I can use VMWare (damn that's an impressive program).

    Windows is not going to disappear any time soon. Neither will Linux. Both OSes have built up way too much steam to just roll over and die. Deal with it. Use whichever suits you best. Believe it or not, Linux is not the best OS for some (gasp! blasphemy!) -- my mom still has problems copying and pasting -- I don't think she's ready to be configuring XFree86 (which, while RedHat 5.2 has made some significant changes to make configuring XFree easier, still requires some knowledge about your computer's hardware, which most people haven't a clue about [horizontal refresh frequency? dot clock?]). If you want to use Linux, use it. But don't unnecessarily evangelize an OS that is not ready to replace Windows yet.

    Better yet, use Linux conspicuously. Answer questions about Linux. Let them come to you -- don't force it down their throats. Then prove the esteemed Mr. Metcalfe wrong.

  • Perhaps articles should be moderated like comments. ;)

    If this article had been posted as a comment, it would have received a -1 for flamebait. I read a lot of flammable opinions without much data to back them up.
  • I have to disagree on this one. Linux can (and does) exist in a Windows-dominated world. Windows cannot exist as it currently does without absolutely dominant market share.

    People don't run Windows because of its technical superiority. People run Windows because the software they want to run is built for Windows. It is built for Windows because everybody runs Windows--it's a lot harder to make money selling to a smaller market than Windows. People run Office because everybody else runs Office, and thus Office is the interoffice standard.

    The biggest feature that Microsoft software gives you is community (or, in Microsoft lingo, market). The factors that make Microsoft software marketable are directly tied to the size of the user base.

    If another useful "community" exists, that is big enough (perhaps 20-25% market share?), then you lose the biggest reason to join the Microsoft community. Once that happens, the entire Microsoft business model collapses.

    Windows can't exist, in its current form, with a market share between (say) thirty and eighty percent. Microsoft simply cannot maintain a moderate market share; it can maintain either a high one or a low one. If it drops to seventy-five percent, it will either be pushed back up to eighty or freefall down to thirty.

    Linux, OTOH, can survive regardless of market share, from point five (and lower) up to complete market dominance. Linux needs money to thrive in a corporate environment, but not to simply exist. To paraphrase Vinod himself, Linux cannot be driven out of business by the simple fact that it isn't in business; it's just freely available code.

  • W2K incorporates everything that every Unix zealot has ever complained about, along with everything that every Microsoft admin has complained about. It's frankly just seriously cool.

    And not just because of the smoke and mirrors(i.e. Gnome and Enlightenment), but because of the technical details.

    Care to share some facts on these technical details?

    Is it stable? Is it lean? Can you run it remotely? Does it return all memory to the system when the program is finished? Has the BSOD been banished? Can I run it on under $1,000 in hardware? Is it permanently innoculated against viruses? Can security bugs be fixed in hours? Does it ship with a compiler? Can I monkey up some Perl scripts to make the machine administer itself?

    With Linux, I can truthfully answer "Yes" to all of the above. That's why I like Linux. Can you answer "Yes" to all of the above concerning W2K? If you can, I am duly impressed and want to see it. If not, I suggest you re-think your position.

  • If W2K is scriptable, kudos to it; it will be much more useful.

    Both W2K and Linux are innoculated against system-level viri. I still think that Linux is better innoculated against user-space viri. First, fewer Linux user applications can automagically run code. Secondly, those that do will only run them with the users' own permissions. That is, if I run a macro virus, all I can do is hose my own data and configurations. The virus won't be able to destroy applications because, as a rule, mere mortal end users lack the permissions to do so. They couldn't if they tried, so viruses running under their UID can't. A userspace virus in Windows (at least 95/98/NT) has more targets available to it.

    In all, thank you. A couple of facts beats a baseless rant any day.

  • Windoze sucks, yes its true, but unix (ALL unix) most seriously sucks as well.

    Unix is without a doubt the most hostile computing environment in use today. You simply cannot throw Joe citizen at a unix machine and expect any success. If you're ahppy with leaving unix as an enterprise serving OS, that's fine, then lets drop all this unix-as-a-windows-killer garbage.

    Linux belongs to millions of people. Hence, it will always serve a million masters - hence it will most likely never be much more than it is today - a great platform for serving open standard network services.
  • ARRGGGH!!!

    When are people going to realize that

    INSTALLING LINUX != ADMINSTERING LINUX

    Installing linux is simple. Administering it is a hassle.

    Linuxconf??? Hahaha.

    The underlying OS is generally not all that important to Joe User.

    It is if they can't figure out why the hell some obscure flag set in /etc keeps them from doing something.

    However, when something goes wrong, and they call their Brother/Father

    Yes, that's the general rule of adminstering unix - get someone to help you. That's the problem! Home users don't want to have to call in a child prodigy with a Jolt addiction to get things done.
  • Its so refreshing to hear someone on Slashdot admit the obvious - linux is an excellent server OS for people who know what they are doing. Otherwise, there are better tools.

    The typical "linux r0lz!" banter in here gets very tiring - mostly it demonstrates how inexperienced the posters are.
  • I was a newspaper editor in a prior life,
    and one of the oldest ways to spark interest
    in a publication's opinion section is to get
    a flame war going.

    Infoworld has a Linux advocate on their
    back page, so Metcalfe rolls a grenade
    into our tent from the bottom of the
    inside-of-the-back page.

    He is so hilariously off-base, and has been
    so spectacularly wrong in the past with his
    internet-gonna-crash-real-soon-now forecasts,
    that his rant can be safely ignored.

    Like many folks on Wall Street (my current
    career) he is a "fade" - someone whose
    trades you take the other side of, because
    they are so likely to be wrong!
  • by account_deleted ( 4530225 ) on Monday June 21, 1999 @02:37AM (#1841876)
    Comment removed based on user account deletion
  • Once upon a time, I wrote a piece called GU Is Considered Harmful [mit.edu]. I haven't updated it in a very long time, but the sentiment is truer now than ever. In reference to my old piece, a friend once sent me a copy of someone else's writings in a similar light.
    > [How *does* one copy *.c to *.c.old in Windows?

    One creates a new folder, calls it "Project Name 990909" (or whatever), copies the files to be archived, and pastes them into the new folder.

    That's how you make a copy of each file in a different folder, with the same extension, the thing that you do in unix with cp *.c "Project Name 990909" But that isn't what was asked for. What was asked for was to make a copy of the files in the *same* folder, but with a *different* extention. That's an entirely different operation, with entirely different results, especially on a system like Windoze where the result of clicking on a file depends on its extension. I suspect that there is no way to accomplish this task in Windoze, other than changing the names of the copied files one by one.

    This is typical of mouse-based UI's. The number of distinguishable mouse gestures is vastly less than the number of distinguishable things you can type on the command line. So a GUI can only do a tiny number of things compared to what a command line can do. But this tiny number is large compared to the even tinier number of things that can be done during a demo, so a GUI demos as being as powerful as a command line, and easier to use with zero training. Since most software gets sold on the basis of demos, this means that if you want to sell your app, it needs a GUI, even though this means that the set of things you can do is far poorer than the experienced user would like.

    Andy.Latto@pobox.com

  • [How *does* one copy *.c to *.c.old in Windows?

    As an advanced windows user (but planning on switching to something else soon) I can tell you it's pretty easy to do this: copy *.c *.c_old

    No, that's in DOS (the non-operating system, not the security issue). Not in Windows. You're using a command line, not a GUI, now. The point is that GUIs cannot approach CLIs in power.
  • But it lags sorely behind in interface. The default command-line does ignore 30 years, and to be honest I have difficulty believing that any operating system still carries such an archaism.
    Hold on there, guy. Calling a CLI archaic is just plain silly.

    Let's think about a piano. It's a rich and complex instrument that takes many years to master. Someone without experience on it is virtually useless. Compare this now with a wind-up music box. Obviously the music box is more convenient for beginners to use. So what? Just because there's a place for a music box doesn't mean that there isn't a place in this world for a piano. Which one is more powerful, more expressive?

    Not everything should be optimized for use by complete idiots. The CLI is infinitely more powerful than any GUI I've ever seen, because it tolerates and encourages communication between programs in ways never dreamt of by their authors. The Unix command line provides the developer with a broad set of powerful but flexible features that the creative mind can use to produce a custom design uniquely crafted to the particular demands of the situation.

    Go out and find someone who has been at this game for many years. Now just sit back and watch them at work. They have created out of base components their own IDE, possible with SUI or GUI components, but probably not. But I guarantee you that it will be one that suits their own tastes and aptitudes. Quietly observe them edit files, move them around, compile them, debug them, test them, etc. The entire development is integrated, like a top-of-the-line German sports car: functional, powerful, and elegant. You will be absolutely astonished at the speed and ease exhibited by the native speaker of Unix in his home territory.

    As on a piano, the art and skill of a CLI virtuoso can only be seen to be believed. That is the path to mastery -- all these cobbled little GUIs are expensive toys designed to sell a flashy demo using cheap tricks, and being optimized for immediate but shallow understanding rather than enduring use, are but a dim palimpsest of real tools.

    Those without creative minds and agile fingers are of course welcome to hurry up with my fries. And they'll probably use a GUI to take my order, too.

  • by sgm1013 ( 60138 ) on Monday June 21, 1999 @02:33AM (#1841899)
    Linux is not about technical superiority - it's about choice - freedom. Of course Linux isn't perfect - Linus et al keep working on it - knowing their history, they will always be working on it, improving it.
    No - what we have is an movement that can't be bought, stolen or steamrolled. We have a choice - we can try something, on our own, and exercise some free will in deciding our own course.
    And one more thing - technology that has been around for 30 years is there for a reason - it works. I have a radio sitting on my desk (early 1900's), I'm using a QWERTY keyboard (1920's), I'm looking a cathode ray tube monitor (1920's), and I'm setting under an electric light bulb (late 1800's). How old are you Bob?
  • Wow!
    what an extra-ordinary prediction. My understanding is communism failed, not because those guys had ideals, but because they believed centralized planning was the correct way to implement them. In fact, it turned out that the decentralized market, for all its faults, was the more robust and succesful solution to providing for our standard of living.

    Bob seems not to understand this at all, either he doesn't know what was wrong with communism, or he doesn't know that it is the open-source movement who are the champions of massively parallel, decentralized systems. Someone should tell him.

    philip

Almost anything derogatory you could say about today's software design would be accurate. -- K.E. Iverson

Working...