Computer history is all about patent prior art searches. You could be one of the guys lawyers hire to find examples of the way things used to be, so that you can demonstrate that something someone is trying to patent is not really a new idea at all.
They can't revoke it and then demand you pay royalties on stuff you made before they revoke it. Some judges might let them revoke it and then sue people who practice the patents in software made after the revocation, but even that is suspect.
But look, none of this matters. A patent isn't the right to do something, it is the right to exclude others from doing something. So you have a license from this guy -- big deal. If IBM has a license that covers a similar area, and you are infringing that, your license to this guy's patents won't help you at all.
The "open source" solution to patents should be to take your invention and publish everything you know about it, immediately, in the United States and in other countries. Then it's prior art which can be used to kill other patents.
Slashdot Mirror
Take EE in college and you'll be on your way.
But look, none of this matters. A patent isn't the right to do something, it is the right to exclude others from doing something. So you have a license from this guy -- big deal. If IBM has a license that covers a similar area, and you are infringing that, your license to this guy's patents won't help you at all.
The "open source" solution to patents should be to take your invention and publish everything you know about it, immediately, in the United States and in other countries. Then it's prior art which can be used to kill other patents.