Is OS/2 Coming Back?
mstansberry writes "Is IBM considering relaunching OS/2? One source close to IBM says Big Blue plans to repurpose OS/2 services atop a Linux core. IT managers ask, why now?" Hey, back in simpler times OS/2 was super badass. Both of the guys who ran it were hard core.
Object Desktop was quite popular - their launchpad replacement was much prettier and more capable than the original. I have no idea how well-recieved the Windows port was, but many of the people on the old IRC channel used OD (and you see it in screenshots just about as much as you'll see the vanilla desktop).
For every problem, there is at least one solution that is simple, neat, and wrong.
You seem to miss the thousands of banks and financial institutions that were using it as well. OS/2 was far more prevalent in large businesses than it ever was with home users.
.technomancer
OS/2 is still running ATMs, train systems, all kinds of important things. It never went away.
There's no -1 for "I don't get it."
I am one of them :D
Ran it, wrote code for it, supported it for 10,000 users from version 2 to version 4.
Unfortunately, they kind of pulled the wind out of the sails around the time Win 95/98 came out, so it didn't really make sense to stay with it.
I still miss little things like being able to reset the video to the default driver with a key combo, SNA/3270 support (which matters if you're not addicted to using a VB front-end for your mainframe), the first graphical remote desktop support, and a really great CDE style dock.
Oh, and REXX. I loved REXX... that was a great language.
An operating system should be like a light switch... simple, effective, easy to use, and designed for everyone.
The only reason they are bloatware on windoze, is they had to write a lot of code to get some WPS functionality. Their original product, for OS/2, just added a layer on top of WPS.
I'd be wary of suggesting that we ever will or should have an official desktop. Some competition and cross-pollination helps us share interface ideas that work after having separate communities really find out what doesn't. Those of us who actually used OS/2 generally also find the very idea of "IBM will save us" to be ridiculous. IBM long neglected, ignored, and occasionally kicked the OS/2 community. They're not really the poster child for sanity. We liked the product, but were very wary of big blue itself.
Also, as a general hint to other people, whenever somebody says "let's face it", it's a good clue that they're being a douche. It's an empty, self-congratulatory phrase.
For every problem, there is at least one solution that is simple, neat, and wrong.
Although there are a lot of virtues in UNIX programming, some people just don't like it. They prefer richer APIs that Windows and OS/2 provide.
That's why there are richer toolkits that sit on top of POSIX and X11, such as Glib/GDK/GTK, Qt, wxWidgets, and Winelib.
SOM programming was a pain in the ass: code an IDL, precompile and get C header file from Hell (it was akin to the first C++ precomilers that would implement everything in C), link, and then there was a binding operation - IIRC. For the WPS, you'd create a dll that would extend it - your application was really a dll that was run by the desktop. It did allow multi threading BUT it was all in the same address space meaning, a bad app took out the whole desktop.
In a nutshell, GNOME and KDE is better than what IBM had invented 18 years ago.
RIP America
July 4, 1776 - September 11, 2001
When I worked for the state there was a company contracted to develop a whole suite of Windows applications to move us off the old VAX green-screen interfaces into the modern world. Most of the department ran on Windows NT 4.
So naturally, the contractor developed all of their applications on a Windows NT 3.51 emulator running under OS/2.
Aaaaand after millions of dollars spent, the contractor demonstrated their applications (working flawlessly under the emulator in OS/2) got their money and high-tailed it, leaving us IT schlubs to implement the applications. All the apps immediately crashed when we attempted to run them in the real NT 4 environment. We never did get them working, except on the few workstations actually running OS/2 with an NT emulator.
Your tax dollars at work. Remember kids, watch your specifications when hiring a contractor!
Because more people used OS/2 than any other dead GUI OS I can think of at the moment. The other (and more important reason) is that many companies are still using OS/2 for critical applications. If they were able to build WPS (which by the way is not what this story is really about) on Linux, your concern about 3 major desktop environments would go away within 18 months anyway. WPS was a better desktop environment 10 years ago than Gnome or KDE are today. If they spent some time actually updating it, the other two would fade into obscurity. Linux has come a long way, but it is no where near being a serious threat in the desktop market. Would OS/2 services and GUI change this, no probablly not.
Why? Because in order to support a desktop OS today, you either have to control the hardware platform or have a significant enough install base to compel every hardware manufacturer to release updated and supported drivers in a timely manner. This is why you have Windows (big install base) and OSX closed platform. Linux works on most every platform, but there are nearly always tradeoffs and limitations, no one devotes the same level of engineering to their Linux drivers as they do their Windows drivers for desktop hardware. In the server space there has been considerable progress made in driver development, in many cases Linux driver support far exceeds Windows on enterprise server hardware.
Desktops remain a difficult nut to crack. Revere engineered drivers are not a viable solution for a consumer operating system, drivers must be engineered and supported for consumer hardware , just like they are for server systems , before you will ever see Linux make any meaningful inroads into the desktop market. Since IBM does not make desktop hardware anymore, it is unlikely they will be the ones to bring a closed platform Linux solution forward (essentially like Apple did with BSD), but an OS/2 Linux hybrid could be interesting if they could partner with Lenovo (for example) and provided a fully integrated and supported solution.
Humor value noted, but for those wondering, he's talking about Extended Attributes, the big database of stuff about files, stored on HPFS. Kind of like a Resource Fork on a Mac file. EA corruption was one of the more annoying things you'd have to deal with on an OS/2 system. Examples of EA data would include the file's icon, data type (which would refer back to which program to open it), etc. Without it, a lot of the system would get really unhappy. There was even a hack IBM came up with to let you have EAs on FAT volumes, but that was a little less nice.
Nope. "She" hasn't gotten the final surgery done yet.
Actually, there was some exchange of technologies going on between Amiga and OS/2. IBM gave Amiga REXX in exchange for some general desktop enviroment tech; something like that.
One that hath name thou can not otter
Now called z/OS, it is still popular too, mostly as the backend to all those OS/2 ATMs. However, neither will see a resurgence. The PC market is 'mature' and will not have room for another general purpose OS. The future of operating system is in the mobile device, then in dedicated purpose devices such as cars, appliances, and gadgets.
OS/2 was a basterd child. I had the first OS/2 developers kit. It cost $3,000, had no GUI (PM came later), and wouldn't compile "Hello World." The day after I got the SDK, I drove from SF to Seattle to attend the first OS/2 developers' conference at the Westin. Balmer was there but Gates was not. I wondered why the head geek did not show up for such a "big event." Now we all know why.
> ...the first graphical remote desktop support...
The X Window System was first.
Warning: this article may contain humor, sarcasm, parody, and perhaps even irony. Read at your own risk.
I don't see any reason at all to dig up WPS for any reason except nostalgia. It might be nice if GNOME/KDE had a better support for "live" objects much like SOM but I hoped they learnt from the mistakes of WPS if/when they implement them.
Up until rather recently, a large majority of bank ATM's ran OS/2.
Many call centers ran software that used OS/2.
OS/2's attempt to reach the consumer market were laughable - they sponsored the OS/2 Fiesta Bowl in the 1990's, without explaining to the public what OS/2 even was - but the software was everywhere in the corporate world it seemed. (for those slashdotters who don't know what the Fiesta Bowl is, it's one of the biggest college football ball games.)
Ford car dealerships ran a satelite uplink system that required OS/2.
I used it to ran a multiline BBS. It was good stuff. Even today, many of the guts (and filenames) of Windows stem from MS's long ago partnership with IBM....the more stable portions of Windows.
Not sure what the relevance of it today would be, but it was more widespread than you might think.
From what I remember about those days, EA corruption was oly problem with people running OS/2 on FAT. With HPFS there wasn't such problem.
For those who don't know, a Workgroup Folder allowed one to put a group of programs and/or documents in a single folder and then open/close those elements as a single logical unit. Open the folder, and all of your programs and associated documents popped open. Close the folder, and everything closed as a unit. It was very slick...
Mainframe/UNIX Bit Twiddler and long time Windows/Linux Hobbyist.
The Theorem Theorem: If If, Then Then.
It ran useably with 4 megs, and it flew with 16. All versions.
OS/2 worked quite well on "business-class" hardware ... 3Com and Intel NICs were almost always support, Matrox was well known for its quality OS/2 drivers, Creative Labs soundcards were well-supported through the AWE64 until they completely changed the chipset and stopped writing drivers for OS/2, etc.
With a little research, it wasn't difficult at all to get a PC to run with OS/2, but sometimes that meant replacing a component. OS/2 had the same issue that Linux did at the time ... most hardware manufacturers tended to provide drivers for Windows only, so you were somewhat limited in what you could use. But sometimes the hardware switch was well worth it ... the original Matrox MGA Millenium was one of the fastest cards around on all platforms for a while, for example, and its OS/2 drivers were second to none.
OS/2 also had EXCELLENT SCSI support, and that plus SCSI's performance advantages over IDE drives of the early/mid 90's made Adaptec controllers the priimary choice for many of the OS/2 folks I knew. Linux had good support for those SCSI cards, too.
Mainframe/UNIX Bit Twiddler and long time Windows/Linux Hobbyist.
The Theorem Theorem: If If, Then Then.
As one of the OS/2 beta testers of every release and participant on several beta-conferences held by IBM and then later using the OS in real life situations with customers, I can truly say you have no fucking clue.
If you mod me down, I *will* introduce you to my sister!
Yeah: vessels going to sea today ...
Nuclear wessels, even.
Hey, Windows users, there is no such thing as "forward" slash, there is only slash and backslash.
No, EA's weren't stored in any "big database of stuff about files" on HPFS, unless you count the filesystem a "big database of stuff" (which would seem rather pointless). The "databases" were for filessystems which didn't have native support for EA's in the filesystem. HPFS stored EA's attached to what we could call "inodes" (they weren't inodes in the proper sense, but for the sake of the argument..).
I wrote a sector editor for OS/2, which could parse the HPFS structures, so I'm not just guessing here.
And if you ever thought EA corruption was a problem on HPFS, you had a broken system (software or hardware). Broken EA's on HPFS was not common a problem. But it was a common problem on FAT.