well Arm evolved as a risc processor. maybe the PB G5 is coming!
No, but the fact that ARM originated as a RISC architecture is significant, and why people with PPC experience (also RISC) are good acquisitions for future ARM products (like PA Semi and the hiring of Papermaster).
Do you think part of the exceptions would arise from the ARM licensing both IBM and Apple posses?
Not likely. The issue is whether Papermaster, while at IBM, was deeply involved with IBM trade secrets, proprietary technology, and particularized expertise--all of which IBM owns and continues to own and control--and would, either intentionally or unintentionally, divulge those secrets or provide Apple with the direct benefit of that technology and information.
This is the essence of unfair competition. Apple would unjustly benefit from research financed by IBM. This is the kind of case where NCAs are sometimes upheld as exceptions to the statute. Without knowing what he did specifically at IBM and what he does at Apple, there's no telling how things would turn out.
can you elaborate on this, not arguing the point just curious.
It's actually a very fascinating area, albeit a bit dry.
seems a very grey area to me,
Quite grey. But there's no other way for it to be. Any black and white standard would create unfairness in implementation on both sides. It's fuzzy because it really must be done on a case-by-case basis to make sure that the original employer can protect its investments and competitive arsenal, but also that the employee isn't unfairly prevented from having the career he wants and is skilled at. It's hard to quantify, because oftentimes a person is hired because of their former position--and that's valid, unless the hiring was a poaching for corporate espionage. But that person has to respect professional boundaries.
One of the great problems plaguing software development, in my experience, is that unlike doctors, lawyers, financiers, and accountants, professional ethics is not a formal part of the enterprise. People honestly don't know what the limits are because they were never taught in a systematic, consistent manner--because there's no system in the profession. But there is a legal system, which presumes that people know the rules.
But surely if the IP in the code is not patented, the ideas can be used in job B?
You can bring your skills and publicly-available knowledge to a new project, but there's a sliding scale, and at some point, ideas stop being ideas and start being specific implementations, which you might not be able to disclose.
Even if there's no NCA involved, you don't want to put yourself and/or new employer in a position to be sued for copyright or patent infringement or for trade secret misappropriation. It is always a bad sign (for the defense) in IP litigation to have a former employee of the plaintiff working for the defendant, directly on the product at issue.
Avoid verbatim reproduction or very similar implementations unless you own the code or have a license. If you're not sure, always ask someone in your legal department. They'll be able to provide more specific help. If you're self-employed, you can avoid a lot of the mess by meeting with an attorney and making sure that your project contracts leave you with the level of ownership and control you want. After that, you only have to talk to a lawyer when a client wants you to sign their contract instead of yours.
Basically, if it belongs to your employer, in whole or in part, then you're not free to disclose those portions to a new employer. When your job and that information become so tangled in each other that knowing the boundaries becomes almost impossible, that's when you get NCAs. If it prevents you from working in the field at all, it's probably not valid. If it only prevents employment in direct competition with the proprietary information, for a reasonable time, it may be valid.
