Slashdot Mirror

← Back to Stories (view on slashdot.org)

IDC: NT usage is mostly hype

Posted by ryuzaki0 on from the over-emphasized dept.

zealot writes "This CNN story reports that International Data Corp. has done research on OS use in businesses and has determined that the amount of NT usage is mostly hype and marketing. It is typically only used for departmental infrastructure, and hardly ever for mission critical stuff. UNIX is still alive and kicking. "

2 of 116 comments (clear)

  1. Does anyone remember this? by kzinti · · Score: 4

    I'm reminded of a joke I heard back before NT 3:

    Q: What machine runs NT best?
    A: A slide projecter.

    Yeah, I know, you've heard it before.

    A coworker who is very experienced with NT says he can configure a department mail/file/web server so that it doesn't need rebooting more than every 3 months. I believe the guy knows what he's talking about. He knows what services to turn off that make NT slow/unstable; he also goes into the registry to tweak things.

    After talking with this fellow, I believe that NT 4.0 has at its core a stable, reasonably good operating system well-suited for small to medium-sized department-level servers. But you have to be an expert to get that -- it doesn't seem to come "out of the box". So the results of this survey don't surprise me much: if you take NT beyond its capabilities, or aren't an expert at tuning it, you're going to struggle.

    Let's face it, though: NT is popular. That is, it sells well. If many of these customers are finding that NT isn't all it's supposed to be, well, you live and learn.

    The lesson from Microsoft, again, would be that marketing excellence is better than technical excellence.

    (BTW, my NT coworker is also a longtime linux/unix user.)

    --JT

  2. This analyst has mixed views by The+Dodger · · Score: 4

    A lot of media hype surrounding Linux doesn't make sense because nobody really understands it. Linux was almost completely unknown this time last year, even though companies like Oracle were already working on porting their software to it. Suddenly, it's become this huge bandwagon, which noone really understands, because there's been nothing like it before, and the media end up listening to so-called experts like Matrin Butler, who really know damn-all about Linux and it's capabilities.

    There's been so much hype thrown about by the media, presenting it as the Microsoft-killer and so on. For example, Linux is now being touted as a viable alternative to Windows on the desktop.

    Bollocks, I say. Linux is nowhere near the point where it can compete with Windows on the desktop. It can compete with Windows NT in the server market, yes - As someone else pointed out above, Linux+Samba kicks NT's ass into a sling. In fact, Linux+netatalk also kicks Apple appleshare servers into touch. But, bring Linux to the desktop in the same way as Microsoft managed to do with Windows, will be a long, long haul.

    The fact is, that, on the ground, people are going ahead, implementing Linux as a server for a variety of purposes, and ignoring all the hype. I've installed Linux machines as file, print, mail, web, and database servers. They require a fraction of the administration required by NT and are more stable and more powerful.

    However, would I reccomend a Linux installation for the desktop or for a high-performance mission-critical server? No. I'd reccomend Windows NT workstation (or MacOS) on the desktop and a Sun Enterprise server for the mission-critical stuff.

    In the future, this might change. Linux definitely has a future. At the moment, I'm doing some R&D into Linux/Beowulf/Clustering/High-Availability/distrib uted-database stuff, and it's all very cool. The reduced reliability of Linux on PC hardware is offset by it's lower cost.

    Anyway, the point is that there's no point in trying to predict what's going to happen with Linux. To forecast where something's going to go, you must know from whence it came, and Linux came out of nowhere, so the statisticians don't have any historical data. Add to that the fact that it's a completely new phenomenon - a free operating system hasn't never attained this position in the past, so the statisticians don't even have anything similar that they can use to make a model.

    Everybody's getting all worked up, but it doesn't matter what anyone says - Linux is going where it's going and noone can really influence what happens to it, because noone controls it.

    In fact, it's all rather cool. :)


    Jack