I would remind you, that IBM employs some of the best lawyers on the planet. If IBM lost control of anything that they own, a number of them would be in BIG trouble. I am also pretty sure that Microsoft is in the same boat. On the other hand, IBM, doesn't bother with small indiscretions, that don't really cost them money (Microsoft may be a different matter, but it would still have to be worth their time and expense). I would not like to be the person who steps over the line. If nothing else, you would need a few million, just to pay your own lawyers.
OS/2 is still being sold to Serenity Systems (and possibly others), who have subcontracted Mensys to sell it for them. Therefore, OS/2 is far from "abandon ware", it is still a current product, even if it has not been updated "forever".
I still insist that the long term strategy should be try to clone the IBM components with open source alternatives. The first versions will suck and will have a lot of bugs, but we will have something that can be evolved, and not like close source software.
This was the goal of the Warpicity Proposal of Lynn Maxson:
http://www.scoug.com/pastpresentations.html
http://www.os2voice.org/VNL/past_issues/VNL0699H/vnewsf2.htm
http://www.os2voice.org/VNL/past_issues/VNL0699H/vnewsf3.htm
Interesting reading, particularly in today's context.
There have been a number of things that have been updated by Mensys, and now Arca Noae (among others). Some of it is based on IBM's work, and some is original. The idea has been around for a long time. Some of it has already happened, but only in small steps. As time goes on, there are fewer people who have the knowledge to even attempt the small steps.
Martin seems to want to clone the WPS, but there is no need to do that, because XWorkplace has already expanded the WPS capabilities, and shown the path to do more (if anybody really wants to tackle it). QSINIT has already made a new OS2LDR, that works well, and as a bonus, it provides a RAMDISK that uses memory above what OS/2 can normally address. That has also shown the way to use that memory for other things (if anybody wants to do it). The OS/4 project is working on a kernel replacement. Unfortunately, there are some legal questions about the way that is being done. Those questions need to be cleared up before OS/4 can be considered to be a replacement kernel. JFS has been changed from the IBM version to an open source version (also from IBM, I think). It has also been updated so that OS/2 can boot from it. There are other things either done, or in the works, to update the OS (including drivers) itself. Then, there are the porting projects, which supply a lot of new application software, and IBM has little to do with them. Unfortunately, there is very little new "native" development, which should be the backbone of any OS. One of the main "problems" is that everybody insists that new software needs to be "compatible" with software on other OSs. I look at a product like PMView (still being maintained), and I see something superior to what other OSs have (although PMView is available in a windows version).
It seems that the "dream" is under way, but if you want to kill OS/2, try to do too much with those who can do it. Small steps are the only way to do it, and those small steps need to be targeted at the most important parts (currently WiFi, and USB 3.0.IMO). User software (Firefox, Seamonkey, Thunderbird, OpenOffice, etc.) also need attention, because the OS needs software to be useful (although older software generally gets the job done). One of the BIG problems here, are the tools required to do the job. Many of them are stretched to the limit already.
The main problem today, is availability of knowledgeable programmers. There are far too few, and they are working way too hard to get some of this stuff done.