A new version of OS/2 is going to be released this year by Arca Noae, and all of this time I thought it was almost dead. And it is supposed to be able to run ported Linux apps as well. Now I will probably keep an eye on this, but I really don't see the point of this considering that IBM pulled the plug on OS/2 back in 1996. Is there a market for this product?
OS/2 never really left. After IBM effectively abandoned it, it was re-branded as eComStation
, and has seen a steady trickle of releases ever since.
It is unclear to me who actually owns the rights to it these days; according to that Wikipedia article, ongoing development is by XEU.com, and it is marketed by Serenity Systems. OTOH there doesn't seem to be any information regarding an actual transfer of ownership to either of those entities by IBM, so perhaps IBM has simply licensed it to (or partnered with) them. (Serenity was already an IBM business partner, so the latter is probably the case.)
Looks like Arca Noae is picking up where Serenity/XEU left off. Or maybe they're affiliated with those companies somehow; I have no idea.
Sure, there's a market for it, albeit a relatively small one. As Captain Ned points out, there are legacy vertical applications (originally written for DOS or OS/2) which are still in use behind the scenes. Eventually the hardware those legacy applications are running on will die, and the companies using them will need to migrate the apps to more modern hardware; in order to do that, they'll need a version of the OS which will install and run on that hardware.
Aside from the above mentioned use case, I don't see why you'd want to run it. No 64-bit support, and limited support for modern desktop applications.
The site says Linux support is for "ported Linux applications"... which I interpret to mean, they need to be tweaked and recompiled to get them to work. I'd be surprised if you can just drop a .rpm or .deb file on there and have it work, and the build scripts for most Open Source applications have likely never been tested on OS/2 (and so will themselves need debugging and modification to get them working). You'll also need a lot of patience to get anything that uses libraries beyond the basic C runtime built, as you'll quickly descend into library dependency hell (this can easily happen even on native Linux distros, and it sure ain't gonna be better on an oddball legacy platform like OS/2).
Captain Ned wrote:
OS/2 was a damn good project/product that ran into the fundamental dichotomy between the IBM vision of software selling hardware (or, preferably, leasing it) and the MS vision of universal software just as Win 3.0 started grabbing market share, causing MS to pull out of the joint development project.. It was one of the early wake-up calls to IBM's command/control mind set.
...and while IBM was late and slow to heed those wake-up calls, they are working very hard (still) to turn the ship. They know their future depends on it, as their traditional businesses continue to wither away. They've been taking the axe to (or selling off) a lot of their legacy divisions, went on a buying spree of "Big Data" and Cloud-centric businesses, and have been playing up Watson/AI and Cloud services big-time.
POWER is still a thing too; POWER9-based HPC systems are slated to come online soon, and a POWER10 architecture is on the roadmap. POWER looks like a solid product, but it remains to be seen whether it is a viable long-term player in the datacenter/HPC space. I don't think it'll ever take over, but I think there's a niche for it.
As an employee of one of those Cloud businesses they recently Hoovered up, I'm cautiously optimistic that they've turned the corner, since my job depends on their turnaround strategy now...