“Co-developed by IBM and Microsoft, it was intended to replace DOS, the aging software that then powered most of the planet’s microcomputers. It never did. Instead, Microsoft’s Windows reinvigorated DOS, helping to end IBM’s control of the PC standard it had created. By the mid-1990s, IBM had given up on OS/2 – a major step in the company’s slow-motion retreat from the PC industry, which it completed in 2005 by agreeing to sell its PC division to China’s Lenovo. But while OS/2 never truly caught on, it’s also never gone away. Even if you believe that you never saw it in action, there’s a decent chance that you unwittingly encounter it at least occasionally to this day.” The last time I took a look at eComStation was way back in 2007.
This one made me smile a little. I was one of the early users of OS/2 2.0 though I moderately despised the gui. Under the hood OS/2 did shine but the interface developers just couldn’t seem to get a grasp of a consistent and simple user experience. Even Warp 4 though better was still lacking in usability, it was simply too “busy”. By that time Microsoft was getting Windows whipped into shape so I was left with little choice but to abandon what I felt was a technically superior OS.
I am saddened that IBM left the carcass with eComstation, they have squandered their opportunity to make something more modern and useful of the old OS. I cannot justify throwing away money to buy what amounts to a legacy OS that will nor even recognize all my hardware, especially printers and wireless cards.
OS/2 should have been placed in the public domain where legions of developers could tinker with the thing in their spare time.
Banks seemed to love it. I remember pulling up to an ATM that had been scrambled by a lightning storm and revealed an OS/2 desktop about 6 years ago.
Back in the day I ran OS/2 warp as my primary OS. It was much nicer than windows at the time… but the lack of native software choices was a big weakness. It should of been the future… but IBM was massively incompetent in promoting it.
Yes, I’ve also seen that OS here in Germany’s banks, but most ATMs seemed (and still seem… uh scary…) some kind of “Windows 2000” or “Windows XP”. OS/2 has been used on workstations used in banks.
Furthermore, OS/2 was very prominent in governmental installations. Because it worked well in regards of networking (and also integrated with IBM mainframes that are still common in those installations), it was in use for a long time. There were many products (e. g. in financial administration) for interacting with the mainframe, such as data analysis tools, programming tools and office communication suites.
As I’ve mentioned mainframes: IBM mainframes were typically IPLed (PC-speak: booted) from a service element (cf. hardware management console, HMC) which was a PC running OS/2. Within its GUI tools you could select which image and hardware configuration to init from, and then have the machine power up by a mouseclick.
I still have my original package with box, CDs, disks and manuals of OS/2 Warp 3 (german language). 🙂
Some german PC vendors even had OEM contracts of delivering OS/2 preinstalled on new PCs. That was part of a good marketing strategy, but with the “higher benefits” of preloading MICROS~1 software, OS/2 quickly disappeared from user’s minds, and finally off the market, even though it had much potential in that time.
Serenity Systems is like any other “legacy support” company out there: they just want old support contracts for themselves, not actually engaging in serious software development, and then suck the maximum amount possible from the product until his dead.
They never had the intent of actually making eComstation a competitive product to get new end users (they do not even/ever has/had the resources to do so), and the development that still goes on is just the minimum to keep their legacy support business running.
That is probably fair enough, but I think it overlooks a factor more significant than Serenity Systems business model – they don’t have the source code.
Any and all changes they have made have been through SOM/DSOM extensions, driver additions, adding/changing art assets, icons, etc., or patching binaries as the machine code level. There is only so much one can actually do without the source code.
Not making any excuses for them, your points are all valid. Just saying that without the source code their ability to do anything constructive with OS/2 outside of simple legacy support is rather limited.
you act like this is a BAD thing, at least they do offer new licenses and support. do you HONESTLY think that if IBM couldn’t hack it that eComstation could compete with MSFT? Do you have ANY idea how many new pieces of hardware are released monthly? Linux has dozens of corps contributing and there is still a decent lag between when new hardware is released and when it has GOOD support, not some buggy beta.
Finally don’t forget that nearly 30% of OS/2 belongs to MSFT. it has pretty much the entire Win16 subsystem built into it which means that while IBM can license it out to anybody they want they can NOT Open Source the code, its not theirs. I remember reading back in the day that figuring up the amount of time and money required to go through OS/2 line by line and remove or replace any infringing code would have cost tens of millions of dollars. Honestly? OS/2 isn’t worth that kind of expenditure.
So at least eComstation allows those institutions that need to keep OS/2 running to buy new licenses and they do add generic drivers here and there. but expecting them to actually try to compete with Apple and MSFT? that’s madness.
You are overlooking the fact that IBM deliberately killed OS/2, when they choose to shift focus to Linux.
If IBM really wanted to keep OS/2 alive, i’m pretty sure that there was companies large enough interested in that great product back in the day. But keeping a competitor alive (even if originally developed by yourself) against a new product line is bad for business, so they choose to simple let OS/2 fade away.
It was not about competing against MSFT and Apple for PC market, but to keep the niches that OS/2 dominated expanding.
But the new strategy of IBM was quite flawless, and i cannot argue against it: get a new flagship OS (Linux), with a vibrant developer community, offset some development cost to the community that also profit from these shared work, replace all niches that OS/2 once filed by Linux, and get access to all niches that Linux himself excels.
Thus, what really killed OS/2 was not Windows, but Linux. IBM never saw OS/2 as a Windows replacement in the last years of OS/2 as their product, but as a niche market workstation OS.
IBM do not wanted to just leave the niche markets that OS/2 had to another company. Anybody know pretty well that for IBM there is no such thing as “too small market”. They wanted all niches of OS/2, the niches of other mainframe/workstation OS and several new niches, all consolidated around a single OS for support cost reasons. The elected OS for this task was Linux.
Now the OS/2 is near his total dead. =(
And i don’t regret it because i love Linux. =D
No, OS/2 was a desktop OS, and IBM has not pushed Linux on that space, at all. What killed OS/2 is that IBM saw the writing on the wall and reformed itself into a services company during the early/mid 90s.
IBM mainly produces technology that can lead to either service contracts or vertically integrated products. OS/2’s value proposition over windows was minimal when it came to either consideration, so it was killed.
Right. Linux as an alternative to OS/2 came later, OS/2 was already in the process of becoming legacy software…
Some videos related to OS/2 and eComStation
http://www.youtube.com/user/eCoTVstation?feature=watch
The current state of O/S 2 is not ‘perfect’ but the people (esp Mensys) are doing a very nice work there.
I *really* hope that the hardware compatibility will improve at the future (like adding support for wireless, printers, ACPI, etc), as for the s/ware I would like to see a better NTFS driver (mainly for external USB drives) and an Odin32 that is comparable to Wine.
I believe, also, that the price of the eCS for personal use should be way LESS than the current price.
Ecomstation has had wireless and ACPI support for a while.
yea, I should mention ‘better’ ACPI/wireless…
I still cannot run the eCS on a 2011 system, due to ACPI problems, as for the wireless….not everything is supported
They actually didn’t want to replace dos, they wanted a better Windows OS. OS/2 was designed to fix many of the original limits of dos, hence windows. Microsoft’s only interest in the venture was to gain access to thousands of IBM patents that they later put into windows products. IBM messed up by trying to sell it only to commercial companies. MS was way less expensive (since they stole it) at something like $30 per Windows 3 versus $200 per OS/2
Edited 2012-04-02 15:50 UTC
What exactly did MS steal for windows?
Joint development of OS/2 between Microsoft and IBM begin in August 1985… Windows 1.0 wasn’t even released at that time. You can’t really start working on a “better” version of a product when the product doesn’t exist yet and you have no idea if it would be successful on it’s own.
Only interest? My take on it is Microsoft started out the venture very interested in building what they thought would become the eventual successor to DOS. It unraveled for a multitude of reasons (on both sides).
It was the unexpected success of Windows 3.0, and the difficulty they had with working in IBM’s bureaucracy that did the whole thing in, among other things. I don’t think that initially it was bad intentions on either party’s part, although things definitely broke down and got heated as time went on.
Besides… The eventual winning product for Microsoft was Windows NT. That was Dave Cutler’s project. He thought OS/2 was a train wreck and wanted nothing to do with it.
I seriously doubt patents where the original motivation for Microsoft. I don’t think the Microsoft of 1985, even for a second, considered taking IBM on head on… The practical purpose of a cross licensing of patents is make sure the parties can’t sue each other. You make it sound like it somehow gives Microsoft access to “secrets” or other such nonsense. If it is patented, it is already public knowledge – that is the whole point of a patent… All the development agreement did was allow Microsoft to use patented methods without having to negotiate for licensing separately – something they would have done and IBM would have been open to based on their relationship at the time.
That is simply not true. Windows 3.0 Sold for $149.95 (upgrades from Windows 2.0 were $79.95). That was on top of the price of MS-DOS. Sure, it may have been cheaper to PC manufacturers, but to be honest that was the difference maker, not the price point of the software. Windows came bundled with cheap computers that sold well. OS/2 came bundled with expensive computers that didn’t (PS/2).
And… Microsoft stole Windows 3.0 from IBM??? That’s just silly…
Edited 2012-04-03 00:17 UTC
If you consider preloads, Windows was ridiculously cheap to the PC seller and free to the consumer. You really have to move to 1995 to appreciate the difference. As Norris testified in the antitrust trial, IBM’s strategy to offer OS/2 on some of it’s PCs meant they had to pay 4-5x for Windows licenses over what they would have paid for offering only Windows. This was even after Microsoft’s per-processor licensing agreement had been deemed illegal by the court.
In order to get OS/2 into enough PCs to get OS/2 into enough consumer’s hands to build demand, IBM would have had to subsidize that extra cost, provide OS/2 to the PC sellers for free, and provide free installation support. Instead IBM went for the hope that consumers would choose the better OS.
IBM got GUI tech from Commodore Amiga. Maybe it was the desktop(?). In exchange, Commodore got the programming language REXX, which was dubbed AREXX.
Not everything that you read in wikipedia is true…
The irony was that OS/2 largely lost in the marketplace to Windows because of the press. The press were very anti-IBM because it was the big evil monopolist. Of course, where are we today with Windows?
That is certainly hindsight for the future of Apple, or Google for that matter.
While Oracle still remains the big villain of the IT industry that nobody really cares about.
This isn’t entirely true. Byte magazine (and some others) praised OS/2 heavily. Of course in 1994 (1995?), Byte published an article that went the opposite and lambasted OS/2; it was so full of inaccuracies to be almost laughable and seemed to be part of Microsoft’s big FUD push before the release of Windows 95.
How good the reviews were depended largely on how much effort the reviewer put into having a system ready to run OS/2. Due to the Microsoft monopoly, drivers were limited, so you needed to do your research. If the reviewer was lazy, then the installation/operation was hit or miss. (Pournelle was guilty of this, and he also liked for support to fawn over him, which IBM didn’t.)
If your system was OS/2 ready, the install was easy.
I would encourage everyone posting to actually read the article, if you have not already done so. From reading the existing comments, it seems more like people are just adding their reflections based on their memories rather than actually reading the article.
I for one, was a bit floored by how much it contradicted my memory. I had a demo of OS2 Warp that we ran on our 386 dx. It was amazing! We really wanted to use it based off of that demo. We never did because of the cost. Why buy a computer and pay more for an additional OS, when it came with one already?
But the point is, the article reviled that the OS 2 warp I loved so much, would never have worked on that PC we demoed it on due to the limited amount of ram it had. How awesome is that the demo worked, but the actual paid for software would not have. Man, we would have been ticked off if we had actually bought it. That was low IBM.
I don’t know how much RAM your system had, but OS/2 ran just fine (albeit a little slow) on my 386 with 4MB.
I thought the article said that OS/2 2.2 needed 6mb of ram. I think we had four, or two. I can’t remember exactly, but it wasn’t 6 or above. I would imagine that OS/2 3.0 Warp would be the same or require more memory.
I ran OS/2 2.0 on 4MB quite comfortably on a 386 for quite a while back in the day. I have seen references stating that the minimum RAM requirements for version 2.1/2.11 were 6MB, but prior to that version it was always 4MB, and the version 3.0 (Warp) was also 4MB.
I believe this was because the version of Win-OS/2 (the Windows subsystem) changed from using Windows 3.0 (which used 8086 real mode) to version 3.1 (which used 286/386 protected mode) with 2.1.
This change involved some surgery on Windows itself. Parts of Windows 3.1 were customized in order to function properly on top of OS/2, where as Windows 3.0 could run practically unmodified on a standard DOSBox session.
Since Windows 3.1 needed more memory than Windows 3.0, that increased the minimum to 6MB if you used Windows Application. OS/2 itself did not ever need more than 4MB until Warp 4.0 (although 4MB was pushing it for v3.0, especially if you actually used TCP/IP).
For people (still) interested to O/S 2, check the “OS4 Phoenix Project” ;-}
Here it is the link.
http://www.os2.ru/
I actually remember OS/2 in a different way. When I was in high school, the company that printed our yearbooks subsidized our journalism department’s technology. They gave us these old 486 AST machines running OS/2 2.1. We had to run Aldus PageMaker in Win-OS/2 mode, but OS/2 had better networking support. It also worked much better with NetWare than Windows. I actually have one of the machines at home.
My dad also used OS/2 at work. They ran some AutoCAD-like program for designing landscapes and decks/porches. As far as I remember, it was a native OS/2 app. When Windows NT 4.0 finally became ironed out, they moved to NT. I kept his workstation from home to play with OS/2 on.
I actually have boxed copies of OS/2 2.1, Warp 3.0, and Warp 4.0. I loved the game Cat & Mouse that shipped with it.
I always thought it was a shame that IBM exited the operating system business. I thought they should revive OS/2 by building their own Linux distribution and putting an OS/2 VM on top of it for legacy support. They could have their own GUI instead of Gnome or KDE. They could call it “Blue”
IBM still has rather large OS development groups, AIX, OS/400, Z/OS, etc.
You’re right. I knew that and shouldn’t have worded it that way. I guess what I meant to say was that I thought it was a shame that they exited the consumer operating system business.
I also think it was a shame they exited the PC business. Their machines were built to last and incredibly stable (or at least as stable as they could be running Microsoft Windows). My family, school, and then clients, used a lot of IBM hardware… The PS/2, PS/1, PS/ValuePoint, and NetVista machines. Even the Aptivas were incredibly proprietary but durable and very easy to maintain. Yeah, they weren’t cheap, but they also weren’t cheap, and I’m always amazed when I see clients running some older IBM desktops that are in much better shape than their 3 or 4 year old Dell boxes.
Somewhat revisionist in it’s views; trying to claim that consumers embraced Windows 3.0, that multi-user was a big deal in 1995, and ignores a lot of the revelations in the Microsoft antitrust trial (pressure on the whole PC industry from hardware to software vendors and magazines).
As far as the user study, well things were pretty routinely “gamed” in Microsoft’s favor. e.g., the Windows PC would have a better processor, graphics card, and more RAM, or users would be chosen from a pool that were used to Windows (throw a new interface at anyone, and they’ll declare it’s hard to use.). This might not be the case, but considering PC World’s bias back then …
While IBM does shoulder a big portion of the blame, OS/2 lost the PC OS wars due to Microsoft’s monopoly. At best, IBM could have kept it a niche player on par with the Mac, which probably didn’t look like enough of a business case. Even Judge jackson’s remedies would have been too little and too late for the alternative PC market. i.e., OS/2, Linux, BeOS, Mac, Amiga, etc.
Although I never got around to using OS/2, I remember very well its “compound document architecture” OpenDoc. That sounded very promising and very “whizzy”. Sadly, like the rest of OS/2, it had trouble getting off the ground.
I remember working in a project that required something similar, but, in Windows 98. To be honest, I already forgot how it worked.
IBM never offered any discounts on OS/2 where MS did. At one place we purchased a site wide for windows 3 and 1. We purchased a site wide windows 95 for almost nothing per system. I can’t say how little we paid but it was close to nothing.
In the 90’s I was involved in IBM systems and tested them on a daily basis. I had a small engineers portable that had 2M or ram and ran OS/2 on it. I also had access to the cost of MS for very large companies that installed their OS on new systems. Again it was stunningly low.
Yes, MS did have a working GUI as basis for windows when they thought about working with IBM. They needed the agreements from IBM to make it work. Then they dropped out saying IBM isn’t going our direction. Fact it they never wanted to make a stable OS. IBM kept working on OS/2 and it was in no way an attempt to replace dos or any other command line OS that IBM sold. IBM continued to sell it’s DOS for many years after they offered OS/2.
Some anecdotes, first.
I had the chance to have OS/2 installed in some machines, at my University, as a “new technology”, sort of a contract between IBM & my university.
They didn’t allow the students, to work with them. Later, the academic people, change its mind, and invite students to work with them. But, with any support or teaching, just, take the machine, take the manuals & see what can you do with that.
Win95 just been released, and students, where getting used to. OS/2 machines work, but, where slow. Used a lot of resources.
A teacher used to work with Mac, said, that IBM wasn’t wrong about requiring more resources (32 MB of H.D. was expensive at that time), the same path, have been with Apple, but, they could it done better.
The Dean, had a OS/2 system folder icon on its desktop. He commited the common mistake, of thowing it to the trash icon, because he tought it was a game or old document.
Microsoft already have several anti Monopoly court suits. I also, remember mentioned in magazines and the early internet, that IBM tried to tie / restrict developers with agreements, (similar to Apple, these days), something that Microsoft wasn’t doing.
Later, I work in a large, manufacturing company, that had a lot of IBM servers, plus, some Windows 98 PCs, as clients.
Several times the programmers require technical info, not just OS/2, but other technologies as well, and had problem getting them. IBM already had its developer network site, but, a lot of stuff, couldn’t be available.
It wasn’t like a patent, or copyrights, like Microsoft did years ago, secretive about technical info. It was more like burocreatic stuff, plus like not having technical people available full time, or doing other stuff because they where contracted.
Even, Microsoft have people, or some of the technical info available at that time.
Summary
So, knowing Microsoft way of doing business (getting rid of competitors), and IBM bureocracy, OS/2 didn’t go as far as expected…
Note:
I forgot to mention, that I live a country where, there is a kind of “conspiracy theory” about IBM, that many competitors outside IBM & Microsoft where discouraged, in favor of IBM or Microsoft.
I don’t have probe for this, wish I had, but…
For many years, there where a lot of manufacturing companies that only use IBM servers, I haven’t see a lot of old Digital, HP, Sun servers, whatever, like the U.S. and other countries did.
And you know, when a company works like a monopoly, its get used to not giving good quality or support, to customers, and users…