On this day, 25 years ago, Microsoft released Windows 3.0. It was a very big deal. You see, for almost a decade before that, Microsoft had only managed to provide MS-DOS. It was an 8086 clone of the CP/M operating system, written by 25-year-old programmer Tim Paterson. Just to be clear, all Microsoft did was buy what Paterson had written. And this may explain why for the next nine years — 2.0, 3.0, 4.0, 5.0, 6.0, 7.0, 8.0! — Microsoft hardly managed to improve it. Of course, Microsoft did manage to come out with Windows 1.0 in 1985 and Windows 2.0 at the end of 1987. They were pathetic attempts. The only thing more pathetic was that Apple (Of course!) sued Microsoft for copyright infringement — unleashing on the world the “look and feel” lawsuit. (Note: the lawsuit that started this kind of anti-freedom nonsense actually dates back to Broderbund Software Inc. v Unison World, Inc. Courts should not be allowed to rule on technology or science when they don’t understand it — which they rarely do.)
But with Windows 3.0, Microsoft had finally managed to create something that was usable. It certainly wasn’t great. And a lot of it was just that computers were more powerful by that time, so the program could work reasonably well. Even still, the earlier versions of Windows did not multitask well. But let us remember: Windows 3.0 was a program that ran on MS-DOS. It was not until Windows 95 (version 4.0), that Microsoft managed to combine MS-DOS and Windows. And it wasn’t until Windows 2000 (version 5.0) that it became a modern operating system. (For the record, Windows XP is Windows 5.1. I find it annoying that people think that Windows XP was this major release. But all they did was put a different UI on it. If you wanted to save some resources, you could make it look just like Windows 2000. Indeed, my current Windows 7 computer looks just like Windows 2000. I don’t like change.)
So there you go. You may think that I’m opinionated about politics, but clearly computer history trumps that. And that’s especially true when I’m talking about Microsoft and Apple. Bill Gates is (usually) the richest man in the world for one reason: he got the contract to provide the operating system for the new IBM personal computers. He used that advantage to stifle innovation for years. But mostly because of computer heroes like Mitch Kapor (creator of 1-2-3), Gates was able to become an incredibly rich man despite the fact that he was only hurting the industry. But it after Apple lead the way with the graphical user interface (which they stole before they started suing everyone who did the same thing), that Gate got immorally rich based upon his control the operating system that so many people used because of the great programs created by people who were not Bill Gates nor worked for Bill Gates.
Happy birthday to the first decent product that Microsoft managed to produce!
Man, that’s harsh! The different versions weren’t just cosmetic changes, and the product got more capable over time. There’s still people who think of msdos 6.22 (and pcdos 2000 and drdos 7.03 and freedos 1.1 and others) with fondness, even in this Age of Windows — just as there are Linux and FreeBSD users who’d rather work with consoles than xWindow GYIs like Gnome and KDE. Really, you can’t call yourself a Power User until you’re addicted to command lines!
And you’re oversimplifying Windows’ evolution, by neglecting the various Windows NT op systems. I’d say Win 2K was much more the descendent of NT 4.0 and 4.5 than it was of Win 95/98.
And for all of Bill Gates’ flaws and Microsoft’s lack of innovation … it could have been worse. Imagine a world where IBM’s mainframe programmers were providing our operating systems, where every day, four billion suffering peons labored at workstations pumping out JCL statements on punched cards to feed their silicon-and-iron masters!
I will grant you a little bit, but it’s just a 500 word rant — not a detailed history of Microsoft. The changes in DOS are almost completely limited to improvements to the programming interface. And I don’t ever recall someone saying, “Man, I really want to run whatever but first I need to upgrade Dos!”
You are right: I did neglect NT 3.1 and on. And you are certainly correct that Windows 2000 is the direct offspring of NT — workstation and server. Windows ME was the last of its kind. (May it burn in hell!) It was for the NT project that Microsoft really invested in coding talent. But the company could have done that long before. There was an amazing amount of great talent that had been doing wonderful stuff with much less hardware.
Cutting Gates and company slack because they acted differently from the way a vertical systems company did doesn’t mean anything. At least IBM used its brand power to launch the PC revolution with an open architecture. Regardless, if we want to talk about what might have been, we need to talk what actually was thanks to a truly great man, Richard Stallman.
But I don’t have it out for Microsoft. The company did good and bad things. MS-Office, in particular, really was great. And the NT project was good. But I have decades of resentment stored up!
“The changes in DOS are almost completely limited to improvements to the programming interface.”
That suggests to me you’re thinking of MSDOS 4 and 5, which I found comparatively uninteresting at the time, and are neglecting the increases in hardware capabilities which came with the advent of MSDOS 3.0, 3.1, and 3.2. I mean, 1.2 MB 5″ floppies! WOW! And 1.44 MB 3.5″ floppies! Gaw-DAMN! And then MSDOS 6, with Stacker and other disk compression methods, and support for gigabyte-sized memory and disk partitions! Damn, that was exciting! I can recall The Real Old Days when I had to chop up a 50 MB hard drive because DOS could only deal with 32 MB partitions. And the Very Special Day I drove a hundred miles or so to the distributor that had just moved into the outer San Fernando Valley that marketed such marvels as that 50 MB hard drive — which cost me $800 that day and which was worth every penny, especially since the only drive in their office was supposed to be their demonstrator and I insisted on buying it right then and there because I NEEDED it for my Kaypro 286. (Why yes, I used to be a bit of an asshole. What made you guess?)
Ah, such memories! Possibly I date myself ….
As for Stallman … well, he’s fated to be a neglected character in history, a John the Baptist analog unfairly shadowed by the Christ-like figure of Linus Torvalds. (Why are you snickering? Why is the whole wide world snickering?) So sad! But OTOH, no Linus and no other evangelists like Bill Joy, and UNIX-style operating systems would probably have superseded in time by VAX/VMS and DG-UX and the like, and Stallman would have inevitably faded away into insignificance, like all those once-important but now forgotten Old Testament minor prophets. (Emacs buffs might disagree, but I rather prefer vi as an editor, so I know I’m objective.)
Someday, somebody ought to write a good technical history — a Ph.D. dissertation? — on the ins and outs of early Unix-style OS’s, beginning with MULTICS and working up to System V and BSD and Digital Equipment Corporation’s variants and so on. An Old Testament, so to speak, with Linux appearing as the Gospels, and AT&T’s System 9 tagging along at the end as the Book of Revelations, or maybe the Quoran.
I fear that there is some partitioning of your defense of DOS between seriousness and parody, and that you aren’t even sure where you land. And I’m with you on that. I too remember being excited about many of these things. But it is kind of like a guy who is locked in a casket 23 hours a day: he really enjoys that hour outside! Actually, I would say that version 3 was primarily an update to software support. The 1.2 MB floppies were around for approximately 2 weeks in the Summer of 1988. As for the compressed file system, well, that was not that big deal and if I recall correctly, it was just something a third party had already been selling that MS bought and added standard to DOS.
I’ve never thought much of emacs. I’m a vi fanatic. I still use it for everything on every platform.
I disagree about Stallman. His most important software contribution is gcc, but I understand that most people remember him for emacs. But he is most important not for his software but for the framework that he created and for his own evangelism. I admire Linus Torvalds, but he was a given. There are a lot of good programmers out there. I remember a lot of people complaining, “Why can’t the GNU project manage to create a kernel?” But the truth is that it hadn’t been very long. If it hadn’t be Torvalds, it would have been someone else. And soon. And note that it doesn’t work the other way around. By the time a workable kernel was created, pretty much all of the unix tools had been created. GNU was already hugely important without Linux. And it continues to be. I use gobs of GNU tools when I use Windows. Of course, Stallman isn’t a well known figure, but he’s hugely important.
Such a book would be interesting to you and me and a few thousand old techs around the world. But I think you are wrong about the Book of Revelation: it would be written by Google and the NSA.
Google and the NSA
+1 for you!
Actually, 5″ floppies lasted a bit longer than a few weeks. I had a 5″ drive on my main PC up through 2005 or so, and still have a couple lying around in my spare parts box. You never can tell when they’ll come in handy again. Suppose I have to reinstall XyWrite III from scratch?
Beyond that I’ve kind of a … not love-hate, but love-neutral … relationship with DOS. I used to love that I had the tools to read and alter every single bit on my MFM hard drives — not just the data in the sectors, mind you, but the bytes on the track that lay between the sectors. Ah the feeling of power! Maybe there’s some disk forensic s/w that would let me attempt that on a 1-TByte SATA III drive, but I don’t have it, and to be honest I’m not especially dismayed by that loss of control.
I’ve adapted to modernity, I guess, and while I still keep PCDOS 2K on one elderly PC, I don’t use it all that much. Some games, some writing … there’s DOSBox and vDOS for most of the applications I ran on DOS, or maybe a virtual machine. All the maintenance tasks I might want to run are more easily done with various flavors of Windows. And I never was blown away by DOS, or even CP/M, or Windows 3, the way I was when I first fired up Windows 98. An operating system which looked pretty! OMG! And a mouse-driven user interface which was almost intuitive, rather than requiring extensive use of a manual. I knew when I saw that, that I was looking at The Future — that finally an operating system had been created which would make millions of people — billions of people! — into computer users rather than a few mere hundred thousand geeks. Even Linux — and I’m planning to install Debian 8 this coming week — has never given me an experience that surpassed that Moment of Awe.
I have great memories with all of this. I wrote my first assembly language program under DOS. I loved being in complete control of everything. But soon, I discovered unix (SunOS) and learned c and then there was no going back — except that it eventually led to some basic Windows programs and then an extremely large MFC multiuser application. And now everything is web-based. I feel so old. The truth is, I’d find it more interesting to build a word processor on a Commodore 64 then anything we get to do now.
I agree that Windows 98 was pretty, but no more so than Motif on X-Windows, which beat it by a long time. Admittedly, it was only running on relatively high-powered workstations (nothing compared to the worst we have today). But I was charmed in that regard.
But I don’t care what I work with. It’s all a matter of style at this point. But I do require vi on all my machines. But the brilliance of the open software movement is that it is available on every machine!
See, told you I was reading your archives. And I understood almost nothing since, and this is so sexist of me, computers are men’s work to me. In my life, men deal with computers and I deal with the law/history.
In the histories of Microsoft, they mention that the company would purposely short the number of programmers (mainly to be jerks from what I could see) and that had a huge impact on the quality of the programming. So I have noticed that always go 3rd generation at least with them.
Microsoft put out total crap for a very long time. The main reason that DOS evolved so slowly is that they weren’t willing to hire great programmers. That all changed when they started Windows NT. They brought in great people — programmers and managers both. But before that, it really was pathetic.
Of the three people in the main company I work in (for), the two guys are very much the beta to the one woman. We are all incredibly opinionated, but I feel if you tested us all for testosterone, she would come out on top. But in general, I think it is just a social thing. I have noticed over the years that the big difference between men and women in tech is that the women are more likely to see it just as the job and the men as a way of life. But I’m with the women: during the tech boom, I just wanted to do my work and go home. All the guys liked to goof off most of the time and spend 16 hours at work. I had better things to do than play nerf ball in the cubicles. That’s another thing: ageism. When I was 35, they considered me old.
As I said, I didn’t understand anything y’all were talking about but that is okay, as long as you do.
I can believe it-women are not big on the workaholic thing as a general rule for our gender. Especially as women get married and have kids-they have other things sucking up all of their free time. Maybe it is also a maturity thing-at 35, most of the even hard core Let’s Play gamers on YouTube are running businesses and have responsibilities with families so while they may still be workaholics, they are also more likely to take work seriously. Even if it involves playing games on YouTube.