Slashdot Mirror
The Sad Parable of OS/2
Still-in-Mourning writes "IBM's first 32-bit version of its advanced PC operating system was released 10 years ago this month. It was better than anything around, yet it failed. Its hopes were pinned on many of the same things we hope today will bring Linux to the forefront. What lessons are to be learned? Will we learn them? A glimpse of a sorry chapter in computing history."
There's a few other simple reasons OS/2 might have failed. The first was that it was just too robust. You comment in the story that it was 10 years ago that it was begun. Well, think about the machines we had 10 years ago. Most people, if they even had a computer, they were in the 286 or 386 department. OS/2 is a heavy-weight. It compared more to what NT was soon to become back then. Yes, it had smaller hardware requirements, but most people's machines were just insufficient for running it. Other possibility was the amount of DOS software out there 10 years ago.
Games and multimedia software were mostly written for DOS because authors needed direct hardware access. OS/2, while having excellent DOS support, it was still too slow and unstable to play Falcon 3.0 or what have you (although, I admit I was able to get CrystalDream II by Triton to run... only after a LOT of hacking).
Aside from that, there were no direct hardware access API's available... ever (as far as I know). When OS had to start competing with Windows 95, Microsoft was introducing the WinG (Windows Graphics) library, the library that eventually lead to DirectX. I'm not saying that OS/2 had no multimedia support (it had a fantastic multimedia model), but it simply was not ambitious enough.
Too bad. OS/2 was never geared towards people with lower end (average at the time) hardware and those who wanted to play games.
Why bother.
From the article:
Microsoft found it all but impossible to develop a useful multitasking operating system for the 286. This was not Microsoft's fault -- the design of the chip simply wouldn't allow much useful to be done with it.
What exactly in the in the 286 architecture prevents the use of a multitasking operating system? I seem to remember MS once touted Xenix, and there were also other Unixen out there. There were multitasking versions of CPM before the 286. Is the article writer missing something, or am I missing something. You don't need to have built in multiple instruction pipelines in the proceessor to multitask. It is almost trivial to write that into an operating system. Remember Andrew Tannenbaum's Minix that came on floppies included in his book "Operating Systems"?.
It appears to me that the article writer is trying to excuse Microsoft's lack of skill by pretending that the task was impossible.
-- Another senseless waste of fine bytes.
Many, far too many, moons ago I started a contract at Aldus Corporation as a SE/T (Software Engineer/Test). My job was to work with one of the first commercial applications developed for OS/2, a new version of Aldus Pagemaker.
Because no-one at Aldus knew anything about OS/2 (they were pretty much all Mac-heads and sneered at PC's, DOS and Windows) they gave me a brand new computer, a bunch of sticks of RAM and a pile of floppies they got from IBM. "Go figure it out." So I did.
The developers (who I was never allowed to meet for some bizarre reason) got Yesler (the codename for OS/2 Pagemaker) running about the time I was getting really bored with playing Reversi (the only real application on the OS/2 distribution I had) and I got started doing what they were paying me for; figuring out how to crash Yesler and/or OS/2 and emailing formatted dumps with my comments to the developers. It wasn't hard to find said bugs, although I was told "You can't crash OS/2, it is too solid." Hah!
Just about the time they got Yesler stable enough that I could put together a demo script the marketroids could use to show off the program (they had to follow it exactly or it would crash) I found a way to make OS/2 have a complete spastic seizure. It involved a fairly complex series of actions that had to be followed exactly, but when you did the last one the computer would freeze and waves of color would wash over the screen. Kind of pretty in a psychodelic way.
We called it the Colorshow bug and the developers claimed it was an OS/2 problem. This kicked off a shitstorm of finger pointing that ended with the developers working around the bug instead of IBM fixing it. Remember, at this point IBM was actually pretty happy about the Yesler project because it gave their new operating system some street cred, so it really surprised me that there was so much rancor. An earlier problem with printer drivers was fixed in a day from my reporting it.
But the punchline happened about a week after I found the Colorshow bug. One of the marketroids came by and asked if I could demonstrate the bug for a group of suits that were waiting down the hall. No problem, bring them on (and, yes, I promise to watch my language). So the suits crowd around the desk and I walk the dog and pony (click, click, drag, click, drag, click, colors, "OOOOHH!", nervous laughter). The suits thank me and they leave.
Then the guy sitting across from me leans over and says "You know who that was?" I shake my head. "That was Paul Brainard," (the CEO of Aldus) "and a bunch of Apple executives up here for a visit."
OK...
Jack William Bell
- -
Are you an SF Fan? Are you a Tru-Fan?
When I was at the MIS center for Designs by Levi, I saw them configuring OS/2 1.0 text mode for one of the cash registers.
While on a contract job in 2000 for a cash register company, some of their cash registers still run under OS/2. And, they run their central control systems under OS/2.
Fight Spammers!
I read through the article, and it was full of weird conclusions. I am very familiar with what was going on in the computer industry during the time period discussed, and I disagree with much of the article.
The story of OS/2 is what taught me that in the computer industry, revolution is not what the customers want; they want evolution. You can sometimes pull off a revolution (Macintosh) but it is much easier to offer a smooth upgrade path.
OS/2 was not killed by some weird conspiracy by Microsoft. Some of the other causes of death listed were not doubt contributing factors, but the major cause of death was: incompatible APIs.
It was not possible to take a Windows application and compile it for OS/2; you had to substantially re-write your app. It wouldn't be quite as much work as re-writing your app from scratch, but it was close. Microsoft didn't want this. Microsoft wanted to make OS/2's windowing API compatible with Windows, but IBM had some other API they thought was better, and they insisted it be used.
This had the effect of forcing companies to decide whether they wanted to write for Windows, or write for OS/2. That was totally dumb of IBM. If people could have just recompiled for OS/2 and offered an OS/2 version of their app, they would have done so. IBM was asking developers for a revolution, not evolution.
But let's go back to the first version of OS/2. Because it was written for the 286, its compatibility with DOS apps was poor. OS/2 1.x offered a "compatibility box" for running a single DOS app at a time; it worked poorly, and it was often called the "Chernobyl Box" because it would often crash (and it would take the whole OS down with it). So, any company that wanted to adopt OS/2 had to plan on getting new versions of all their applications.
But in 1990, Windows 3.0 shipped. It sold like hotcakes. The article makes some bizarre statements about Win 3.0, but the reality was that it would multitask your DOS applications very well. DOS applications were preemptively multitasked, not cooperatively, and DOS apps could very well crash but usually Windows would not crash with them. In other words, Win 3.0 allowed companies an evolutionary upgrade path: they could keep running the same DOS apps they were using, and then phase in Windows apps over time. The same companies that were unwilling to commit to OS/2 were willing to commit to Win 3.0.
Win 3.0 was what made Microsoft decide to walk away from OS/2. The customers were voting with their dollars, and what they were voting for was Windows. It didn't hurt that Microsoft had covered all bets: they had applications for DOS, Windows, OS/2, and Macintosh. (They even flirted with a few other platforms: my favorite word processor for the Atari ST was Microsoft Write.) When Win 3.0 took off, Microsoft was ready, and sold lots of Word and Excel.
So, to review: IBM forced developers to choose whether to develop for OS/2 or Windows, and Windows became a runaway hit. That's it right there. That's what killed OS/2.
steveha
lf(1): it's like ls(1) but sorts filenames by extension, tersely
I use OS/2 just about every day at work, and i can tell you it's very much still alive and kicking :) when the software that runs our equipment was first designed, windows 3.1 just wouldn't cut it, so the programmers decided to use OS/2. since then, each new version has been built on the last, and new equipments' software borrows elements from the previous generation's. so to this day, all our machines use OS/2 version 3 in their embedded computers, and all field engineers are given laptops that dual boot into Warp 4 to run simulations. (ironically, OS/2 is not supported on these IBM laptops, so it's sometimes hard to find drivers.) it looks like the big push to port everything to windows NT is finally on, though. but it will still be around on all our machines in the field for quite some time to come....
Don't forget, in 1995 Microsoft did have a monopoly on PC OS's and what they did to make Windows 95 apps NOT run under OS/2 was anti-competitive and illegal. Microsoft built Windows 95 to load a few resources up at the 1GB memory address just to prevent OS/2 from running Windows 95 apps just like it ran Windows 3.x apps. You see, OS/2 could ONLY access 512MB of address space.
Let's not forget that in Nov 1994, at COMDEX, HP had 50% of their PC's running OS/2 the night before the show opened. Bill Gates made a phone call and by morning, NO HP computers were running OS/2.
The list goes on. I blame IBM 10% for OS/2 not gaining more market share and the other 90% was Microsofts anti-competitive nature to do ANYTHING to prevent the consumer from making the choice.
Speaking of choice, do you remember that Microsoft threatened to pull out of COMDEX because IBM was doing it's keynote speech about choices unless IBMs timeslot was moved to reduce the viewers. I think IBM dropped out of COMDEX the following year and all since.
Think about it. It took MIcrosoft 10 years to build a version of Windows that is close to OS/2 v2.0... well maybe v2.1 is a better comparison since it had better legacy Windows support and the 32bit graphics system updates. TEN YEARS!
WHERE DO YOU WANT TO GO TODAY??? With Microsoft?
Nyet.
LoB
"Anyone who stands out in the middle of a road looks like roadkill to me." --Linus
Actually with OS/2 the interface was seperate from the OS.
I remember dropping the GUI all together and using a text based switcher to run my BBS in.. didn't need a gui, just a alt-tab interface to the os2 cmd prompt so i can run PCBoard 15.1 and play sierra games at the same time.
I too have heard this---from someone at IBM, no less.
Interestingly, he told me that this was one of the obstacles to open-sourcing OS/2. The banks are worried about people having the source to the OS that runs their ATMs.
"Never bullshit a bullshitter" All That Jazz
"OS/2 1.30 (SE and EE) was the first version which was written entirely by IBM. There was still some Microsoft code in it - that would not go away for a couple years yet - but all of the new code and a good portion of the existing code for OS/2 1.30 was written by IBM. As a result, OS/2 1.30 was smaller and faster than previous versions, more stable, and there were far more device drivers available, though still not nearly enough."
This is from your own link #2. Many users agree that the quality of OS/2 peaked around version 2.1. At this point there was very little Microsoft code left in there. Reading the original article that \. linked to, these early releases weren't very good at all. Only after IBM re-wrote it and brought it beyond the simple *text-mode* support of 1.0, did it gain a following. If Microsoft was able to create a system as good as the OS/2 written by IBM, it wouldn't have tripped and stumbled through NT 3.51 and 4.0 before releasing something decent like 2000. This comment sounds real juicy at first glance but is highly misleading. Sheesh, if only I had some moderator points.