If said patents were filed and shelved for some time, they should hold no validity in court afterwards.
Edit: that said, it is interesting how they came up with said patentable ideas :) , and it does fit with what Asimov said about the intersection of components, and daring being encouraged (at IBM, take note!).
That’s how it works right now. The “some time” is 20 years after the filing date.
Still, much wiser heads have written lots about this :)
IBM should open up this game to the outside world, so other people (not just IBMers) can use it to make a buck, while IBM can benefit from a large stream of interesting new patents from a diverse audience.
I would consider to submit ideas there myself, because I usually don't have time anyway to implement them myself.
At a corporate employer, "Think Fridays" would be a logical name for a program which rewards employees who do the least creative thinking consistently 4 days a week.
Patents at IBM are very easy to submit and the process is somewhat like a game: basically you fill out a web form with the idea, which triggers a process to vet the idea and eventually get the patent written and filed if it's good. You get a small bonus for each accepted patent (and another after filing I think). You get a larger bonus after your first patent and after every fourth patent (called a plateau). You get a nice plaque for each plateau.
I have eight patents from this system, but there are IBMers who use it to enhance their income and have 100s of patents (they have a large stack of plaques). They have the discipline to use every Think Friday to come up with and submit a new idea- maybe one out of three will make it all the way through and be filed. Some form brainstorming groups. They often pay attention to new technologies and get ideas from how they could be used as components. Many ideas come from the interaction of components. (Many of mine are from hardware/software interfaces).