From - Sat Sep 5 20:47:08 1998 X-Mozilla-Status: 0001 Message-ID: <35EC3C11.F1E74BF9@ix.netcom.com> Date: Tue, 01 Sep 1998 18:25:21 +0000 From: "Karsten M. Self" Reply-To: kmself@ix.netcom.com Organization: Self Analysis X-Mailer: Mozilla 4.04 [en] (X11; U; Linux 2.0.31 i686) MIME-Version: 1.0 To: mike@elgan.com Subject: [Fwd: The Death of Windows] Content-Type: multipart/mixed; boundary="------------EA9A4A2BFB257D53008EA919" This is a multi-part message in MIME format. --------------EA9A4A2BFB257D53008EA919 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit 1.) Windows Has No Real Competition. When we first heard about the network computer (NC), it was supposed to kill Windows and transform computing. Now the NC barely registers on the "legitimate platform" meter. The NC was overhyped from the start. Remember "dickless workstations"? The fact is that we've now got an NC hybrid. Much of the advantages of the NC still apply, but falling hardware costs, and people's desire to control their hardware (and have local, fallible, storage) killed the cheap, dumb terminal concept. Look instead to OS/2 (running strong, silent, and deep in business), AS/400 (ugly motherfsckers, but good at what they do), and 16 processor, 10 GB RAM Sun E10Ks . Those systems will want desktops to talk to which *don't* assume the desktop is the center of the universe. Linux will also fail to unseat Windows. If you really believe this, you can sign on with a witness to the DOJ. Windows (or the current market leader) will continue to face newcomers in the market, or market shifts. The incumbent has an advantage, but it's not iron clad. The history of the computer industry is of leaders (IBM, DEC, Unix, C/PM, IBM PC, WordStar, WordPerfect, Visicalc) in various sectors being unseated. Replacing completely isn't the point. Linus himself says that a 20-30% penetration would be about right -- anything more is unweildy and carries too much responsibility. "Can't we all just get along?". Linux, a variant of UNIX, is the new OS/2. Hailed by devotees as the future of open computing, Linux is a seven-year-old PC operating system used by about 7 million people. We're all guessing, but the current estimate is 10-12. 7m was the midpoint of Red Hat's estimate in March. Doubling annually. Meanwhile, NT has been under development since 1986, and is about to have its 5th beta release (or are you going to tell me that NT 4.0 SP2 was a production-level product?). Like OS/2, Linux is considered by devoted supporters to be better than Windows because of a short list of features most people don't care about, such as superior multitasking. Do your own research, I'm paid by the hour. Try stability, no DLL hell, support, for starters. What are your windows gripes? Do they exist on Linux? Yes, many of the advantages are technical, and steer Linux (and the other free Unixes) toward server type roles, but they're also the basis of a rocking workstation. And, like OS/2 in the early 1990s, Linux is way too hard to install Bullshit. I'll mail you my 5.0 RedHat disk if you give me a postal. Insert, answer a couple of questions, and spin. No fair dual-booting (yes, many do, but it complicates install). Better yet, call Larry Augustin at VA Research (1-888-Linux4U) and get it pre-loaded. (Disclaimer: Larry grants webspace to a project of mine). When was the last time you installed Windows? and undersupported by application developers. Bullshit. http://www.linuxmall.com/ http://sunsite.unc.edu/LDP/products.html#software http://sunsite.unc.edu/LDP/links.html Oracle, Informix, CA/Ingress, Corel, Applix, StarDivision. Still, Linux is not completely like OS/2; it has some fatal flaws of its own. For example, like UNIX, Linux comes in multiple and incompatible flavors, Yes and no. The kernel is the same. The libraries (DLLs) can differ, which is about the same problem as MS's DLL hell, but on a standardized basis. You get to choose your GUI (or no GUI) -- this is bad? A number of standards efforts have been created, there was some noise last week when one of them got a bit overlordish. The upshot is that the two with the best chance of making it (Linux Standard Base and Linux Common Standard (I think)) have joined efforts, and the upstart is having a hard time breathing. Even KDE and GNOME announced a common standard for data interchange (these are the two primary contenders for the desktop). isn't backed by a blue-chip company and has Which is hardly the same as saying Linux is not supported. http://sunsite.unc.edu/LDP/products.html#consult http://www.redhat.com/commercial-support/ IBM? How blue do you want it? (Apache now, and keep watching) Intel -- Committed to supporting Linux on Merced. I'm always amused by this line -- when did Microsoft support suddenly get so good? very low awareness within corporations. Cite your survey. Who did you ask? My first intro to Linux was in 1994 from a middle-management type -- yes, on the technical side, but no propellorhead. My current and prior contracts (Visa Int'l and a small medical devices firm) are both running and/or strongly considering Linux as a replacement for both NT and commercial Unixes on cost, performance, and TCO bases. 2.) Windows Has Overwhelming Momentum. Because Windows was the standard platform when computing went global (during the 1990s), it will be very hard to get rid of. This is true. Companies just don't upgrade all their PCs every 18 months. This is not. I'm running an operating system at home which is rock solid, dependable, and has a full range of applications available for it. I have trouble reading the latest MS file formats. I'm running an OS at work which is slow, buggy, and unstable, and cannot read the latest MS file formats. RH Linux 5.0 and Win95 (Office95) respectively. Microsoft works the upgrade treadmill, you know it. NT 5.0 won't be backward compatible on the *file system* level with 4.0. Go figure. Even this year, Microsoft will sell millions of copies of Windows 3.x to companies expanding the number of legacy systems without upgrading them. Do you ever wonder why? Do you wonder how many of those systems will never leave DOS? Very dependable little single-tasking OS. Future versions of Windows must support Windows 3.x and Windows 9x applications or they'll fail. And that's why only a platform that fully supports Windows applications can succeed. If it allowed, there are several emulation projects for Windows, which cannot be ported to Linux due to contractual obligations (Insignia's SoftWin for one). There are emulation projects which may succeed anyway -- Bochs (http://world.std.com/~bochs/), WINE, and DOSemu. If you really need to emulate, you can. Most of the 16 bit apps are supported. Almost anything standard in DOS will run. There will be a need to support legacy, but supporting legacy is not a growth market. The other killer would be a vb2perl or vb2foo (foo is what you will) -- or a VB for Linux (I think it exists) which would allow businesses to port their existing Excel & Access apps to the new world. The folks who stuck with Borland Paradox are already in luck. This is a concern, however. 3.) Windows Is Microsoft's Golden Goose. Despite all the hype, Microsoft is actually a medium-sized company, ranking about 400th in 1997 revenues worldwide. I don't dispute your revenue figures. If you're trying to make an anti-monopoly case, please remember that MS covers 90% of the desktop, and that this is amazingly powerful real estate. IBM dwarfs Microsoft with 1997 revenues of $76 billion, nearly seven times as much as Microsoft. Even co-conspirator Intel has twice the revenue and three times the staff. And yet, in terms of market capitalization, Microsoft is No. 2 (after GE). Why? Microsoft is valuable to stockholders because it's the most influential company in the world's most important industry; its intellectual property-software-is in almost every PC user's face all day, every day; and the company is well-positioned to continue its dramatic growth well into the future. Microsoft's importance, ubiquity and golden position stem directly or indirectly from Windows. Don't expect Microsoft to give that up without a fight. They're not. But this is a hot seat, and nobody's held it forever. The very fact that one company holds so much power makes them a natural enemy -- in any market. This happened to IBM and AT&T, it's happening with Microsoft and Intel. Again, Visa and the banking industry see Microsoft one of their top three threats (sources: Fortune magazine earlier this summer, Visa All-Hands meeting). Does the banking industry really want to put the cards in the hands of the competition? Besides, the stars are lining up: - DOJ - Y2K - Financial meltdown (oh, where is the Down Average today?) - Piracy -- 90 - 98% in eastern Europe and Asia. It's just another distribution channel to Linux. - Linux - NT 5.0 woes - Merced woes - Doesn't play well with others ...Microsoft can't count on NT 5.0 rolling out in a big way until 2001 (source: Gartner) due to several of the above concerns. The company is based on growth, and they have only two ways to go: up (servers) or out (transactions & finance). The road in both directions is rough and the natives are unfriendly. Meantime, Linux, OS/2, AS/400 and Novel keep truckin' along. The last is the fatal indictment for MS -- they got where they are by sharing, they've decided they don't want to anymore. Incompatible mail systems. Incompatible networking standards. Insistance on ruling the roost as a network server. Invite, invent, dominate plan of software aquisition -- bring someone in to build an app, copy it, quash the competition. After a while you find folks don't want to play with you anymore, and you wonder why. Until someone comes up with an open, wildly popular Windows-compatible alternative backed by a huge company, Windows will remain No. 1. Read Nick Petreley's forum introduction here (http://forums.infoworld.com/threads/get.cgi?66001) which explains why having an OS backed by a single company isn't such a wonderful idea. It helps if you're a Tolkein fan (or at least know who he was). Things you missed entirely: - Linux's support is not based on a huge marketing effort by any one company. It's grass-roots. No other OS (aside from the original Unix -- old BSD days -- can claim this. - Support for Linux in the commercial world is exploding as of this summer. My mom is calling me up with articles in the NYT and WSJ. She's a great mom, she's hardly a propellorhead. - Linux and NT are the only two OSs with growing market share. - Samba -- better NT than NT Suggestion: Spend a week hanging around Slashdot (http://slashdot.org). Don't mind the flame, it's generally a pleasant little hell . -- Karsten M. Self (kmself@ix.netcom.com) What part of "Gestalt" don't you understand? Welchen Teil von "Gestalt" verstehen Sie nicht? web: http://www.netcom.com/~kmself SAS/Linux: http://www.netcom.com/~kmself/SAS/SAS4Linux.html 10:21am up 80 days, 7:50, 2 users, load average: 1.69, 2.00, 1.82 --------------EA9A4A2BFB257D53008EA919 Content-Type: message/rfc822 Content-Transfer-Encoding: 7bit Content-Disposition: inline Return-Path: Received: from keystone.cmp.com (keystone.cmp.com [192.155.65.22]) by ixmail8.ix.netcom.com (8.8.7-s-4/8.8.7/(NETCOM v1.01)) with ESMTP id KAA21987; for ; Tue, 1 Sep 1998 10:14:16 -0700 (PDT) From: melgan@winmag.com Received: from NotesSMTP-01.cmp.com (gw58-246.cmp.com [192.155.58.246]) by keystone.cmp.com (8.8.8/8.8.8) with SMTP id NAA05360 for ; Tue, 1 Sep 1998 13:12:49 -0400 (EDT) Received: by NotesSMTP-01.cmp.com(Lotus SMTP MTA v4.6.1 (569.2 2-6-1998)) id 85256672.005E8C8F ; Tue, 1 Sep 1998 13:12:43 -0400 X-Lotus-FromDomain: CMPNOTES To: kmself@ix.netcom.com Message-ID: <85256672.005E80DD.00@NotesSMTP-01.cmp.com> Date: Tue, 1 Sep 1998 13:12:34 -0400 Subject: Re: The Death of Windows Mime-Version: 1.0 Content-type: text/plain; charset=us-ascii Content-Disposition: inline Can you elaborate? (Please reply to mike@elgan.com) -mike --------------EA9A4A2BFB257D53008EA919--