[tzs]

First to invent can screw the first inventor too, because the very notion of "first inventor" is kind of fuzzy. Suppose this is the time sequence:

   I come up with an idea.
   You come up with the same idea.
   We separately work on implementing the idea.
   We encounter serious problems with making it work.
   You solve the problems and actually get the thing to work.
   I solve the problems and actually get the thing to work.

Which of us is the "actual first inventor"? I had the idea first. You solved the practical problems necessary to actually make it work first.

In the US under first to invent the way it works basically is that your invention date is the earliest date such that you were working on the invention on that date and you worked diligently on the invention from that date through filing the patent.

So, in my earlier example, if you and I both worked diligently from the time we conceived the idea to the time we solved the problems in making the thing work and both filed patents, I would win.

However, if the timeline was this:

   I come up with an idea.
   You come up with the same idea.
   We separately work on implementing the idea.
   We encounter serious problems with making it work.
   I take a few months off to work on unrelated things.
   I resume work on the invention.
   You solve the problems and actually get the thing to work.
   I solve the problems and actually get the thing to work.

You would have priority under first to invent now. Since you were working diligently from you date of conception, that is still your invention date. My invention date would now be the date I resumed working on the invention after my break.