Summary: The Death, Life, Death, Life, Death, Life, Death, Life, Death, Life, Death, Death, Certain Death, and Continuing Life of OS/2

Every OS/2 user who reads the virtual pages of this magazine is aware of the infamous "Judgement Day" post that Brad Wardell sent to one of Stardock's newsgroups that was covering Stardock's attempts to license and redistribute an OS/2 client from IBM. Since that time there has been a furious cacophony of posts and pronouncements from OS/2 users, companies, personalities, and pundits, all arguing as to whether or not, this time, OS/2 is actually, once and for all, finally and truly dead.

I have my own opinions on this as well, of course. Everyone does. But before I get to those opinions, let me set up the story, and give you the background on all the events that led up to "Judgement Day," as I see them.

We all know that at a certain point last year, IBM made an announcement that they would not be releasing a Warp 5 Client, only a Warp 5 Server called OS/2 Warp Server for e-Business. If you remember, at that time, there was a lot of distress in the OS/2 community because of this.

There's very good reason for this, after all: most of us are probably using the client right now. A few of the lucky rich users have bought the latest Warp Server and are using it as a client, but for the most part we're using either Warp 3 or Warp 4 as our client OS.

IBM's decision not to release a client with the release of Warp Server 5 was translated as, "If they're not going to release it with this version, they're probably not going to release a new client ever again." This, in my opinion, was a wholly reasonable assumption on the part of OS/2 users everywhere. IBM was trying to change the perception of OS/2 from a client/server platform into a Web Server and e-Commerce server/thin client platform.

Around this time an internal memo was leaked to Loren Bandiera's OS/2 Warp News & Rumors page that stated, in frank detail, that IBM was planning on slowly migrating OS/2 users over to NT or Linux systems. This internal memo was promptly posted on the website, and a fairly high-ranking official within IBM asked the webmaster of that site to remove it - which he did, but, unfortunately for IBM, many of us had already made COPIES of that post, and it had been posted on Usenet, emailed to friends, and kept on our hard drives for further study. The news was out: IBM wanted OS/2 dead... the client, anyway.

IBM stated that the memo was a draft and had many innacuracies, and that it would provide us the final version in a few days. In a few days, the "final version" was posted on the web, and after reading it, all I could do was laugh.

The "final version" of this internal memo read like a piece of standard IBM marketing drivel. Where the "draft" memo was filled with strategy suggestions and information on how their objectives would be reached, the "final version" was filled with mindless exhortations about how OS/2 would be an integral part of an e-business solution, but with no specific instructions on how that would be achieved.

In short, from where I'm sitting, IBM tried to pull a bait-and-switch on the entire OS/2 community: They replaced their internal memo with some piece of marketing blather that was full of sound and fury but told us absolutely nothing concrete on OS/2's future. The authenticity of both documents was the subject of bitter debate between OS/2 users: Some said the first document was a fake, some said the second document was meaningless. I'm among the latter.

But don't read this and think that all IBM was trying to kill off OS/2: Throughout the battle there were (and still are) OS/2 supporters from within IBM itself trying to protect and enhance the product. During the time just before OS/2 Warp Server for e-Business was about to be rolled out, some IBM'ers posted on Usenet that they were trying to get IBM to strip out all the server aspects of the server OS and sell that stripped-down version as an OS/2 client. There have always been OS/2 supporters within IBM, but it appears that, of late, their political power is weakening: IBM ultimately nixed the idea, despite the rumored support of some big OS/2 clients out there.

The OS/2 client was still supported to the point that OS/2 Warp 4 and Warp 3 fixpacks were released, and we did get our version of Netscape Communicator 4.61, Java 1.1.8, and the Java swing components. But it didn't look as though IBM was going to do too much to build on to and expand the parts of OS/2 that made it great. No improvements or fixes to SOM, the building blocks of the wonderful Workplace Shell. No journaled file system or support for multiple processors. No fully 32-bit TCPIP stack (unless you paid for it separately). In short, OS/2 client had been relegated to the status of "legacy system."

The OS/2 community seemed unsure what to do. IBM was sending a mixed message of two extremes: On the one hand, OS/2 was apparently an integral part of its e-business strategy, on the other hand, the strategy didn't seem to include the OS/2 client, and it certainly didn't include the people who used the client and not the server. So it seemed, at that point, that the OS/2 client users were being left to twist in the wind.

And most of us didn't appreciate this very much, thank you kindly.

At that point, an idea began to surface. Some people thought that it might be possible for a company to license, from IBM, a stripped-down version of OS/2 Warp Server for e-Business and resell it as a third-party client operating system.

I had actually heard about this plan from a group not associated with Stardock in any way shape or form. It was being considered months before Warpstock '98. It was only in the formative stages, and apparently some feelers had been sent out to some IBM-ers with encouraging results.

When Stardock came on the scene announcing that it had entered into talks with IBM to do the same thing, I was surprised - but not upset. Stardock was an OS/2 company that had a presence outside the OS/2 community, had made some very popular OS/2 products (Object Desktop, Galactic Civilizations) and was well known within the OS/2 community. I didn't always agree with the opinions of Brad Wardell, Stardock's president, but I couldn't fault him for the things he's done for the OS/2 community in the past, and I liked Stardock's enthusiasm when it came to promoting their products. It seemed to me that if Stardock were able to market and resell an OS/2 client, they'd be able to tap into the enthusiasm that OS/2 users have for OS/2 and communicate that very effectively.

Stardock's announcement made it seem as though they'd already come along a fair distance in their talks with IBM, and suddenly the OS/2 community seemed energized - especially on Usenet. Comp.os.os2.advocacy was split over whether or not a version of OS/2 released by Stardock would be a good thing. Stardock tends to provoke strong feelings, both good and bad, and there were people convinced that Stardock was about to repeat the miracle of the fishes and the loaves, and there were people convinced that Stardock distributing an OS/2 client was one of the seven signs of the impending apocalypse. This back-and-forth was entertaining to watch for a while, and the things Stardock was proposing for this client were interesting as well. Some of them were covered in a previous interview with Brad Wardell in OS/2 e-Zine!

Meanwhile, other projects in the OS/2 community continued. The OS/2 Netlabs, a group of people working on providing all kinds of free software projects to the OS/2 community, had taken the reigns of the old Win32-OS/2 project and was continuing its development. They were also working on a project that would allow OS/2 users to integrate Xfree86 with the Workplace Shell, allowing us to run programs like the GIMP in a seamless Xfree86 session.

Also, Lynn Maxson and a few others were starting to make new noise about the Warpicity project, which I had heard him speak on during Warpstock 98. Warpicity is an attempt to create an OS/2 clone by feeding software specifications into a logic engine and generating code in machine language on the other end.

And finally, there was a lot of speculation as to whether or not IBM would "open up" the OS/2 code by releasing it as free software. While many people pointed out that much of OS/2's technology is licensed from other companies, making this difficult if not impossible to do, this speculation has never quite gone away.

Somewhere along the line, however, a lot of people started believing that the Stardock Client was the One Last Hope of a future for an OS/2 client. It's understandable considering that Stardock's proposal seemed well organized and somewhat complete. Brad had visions of how the OS/2 community would be able to get involved in the new client's upkeep and development, without having to have access to the source code. It was a much more coherent vision than many had heard from IBM itself concerning OS/2, and a lot of us thought that it would really be great to see the client in the hands of someone who had plans, and whose plans included us.

Stardock's project seemed to gain momentum, and it looked, to many of us, as though they'd get the OK from IBM to go ahead. There was even a contest on one of Stardock's newsgroups where we were trying to think of a name for the new client. Finally, Brad announced that a meeting was coming up that would decide, once and for all, whether or not Stardock could move forward.

The meeting day came, the meeting day went.

The next day, Brad posted a message on his newsgroup that said, essentially, "They decided. We're not going to be able to do this, and it doesn't look as though they're going to do it either."

I was surprised at how disappointed and depressed I was when I read that news. So disappointed, in fact, that I quickly uploaded a new Help Desk comic strip to reflect precisely what I was feeling at the time: OS/2 was dead, because IBM had murdered it.

I spent the rest of the afternoon and evening in a state of depression. "OS/2 is dead," I told my wife. "OS/2 is dead," I told my parents (who were visiting that day). I went to sleep miserable, because the only thing I could really think about was that OS/2, an operating system that I had learned almost everything about computers from, was dead.

The next morning I noticed with some amazement that despite OS/2's apparent death, when I turned on my computer I continued to be greeted by the Workplace Shell. All of my applications, I noticed, still worked - regardless of IBM's negligence. I could still get my email, browse the web, and create my comic strip using my OS/2 software. In fact, Stardock's announcement that they wouldn't be able to license and redistribute a new OS/2 client in no way altered the way my computer functioned.

I guess I overreacted bit.

Don't get me wrong - I don't think Stardock's loss is a nonissue, here. I think that this pretty much guarantees that, unless IBM is secretly planning to release all of OS/2's source code under some kind of open source license, it plans to ditch the OS/2 client and never look back. We'll get fixpacks for another year or two and that'll be it. After that, we'll be on our own. And add to that IBM's recent decision to eventually close of the Software Choice website to all but subscription members, and suddenly we're faced with an operating system that isn't supported that we'll have to RENT instead of BUY.

And yet, OS/2 is still sitting on my machine, chugging along fine. Of the four operating systems I have installed on my computer: OS/2, Windows 98, SuSE Linux, and BeOS, OS/2 is the one I always go back to when I'm really trying to get something done. And I suspect this will continue to be the case for a long time even IF our worst-case scenario comes true and tomorrow IBM says, "OS/2 client is no more, unsupported, go away and go to hell."

At this time, that doesn't matter... and there are some very interesting things going on in the OS/2 world that, when finished, will give OS/2 capabilities that other operating systems don't have. When that occurs, we'll see just how archaic and outdated our favorite OS really is.

Meanwhile, my computer still boots OS/2 without any problems at all whatsoever. And I suspect yours will, too.