Frank Mittelbach <[EMAIL PROTECTED]> wrote: > Walter Landry writes: > > Frank Mittelbach <[EMAIL PROTECTED]> wrote: > > > Note that above we also addressed the concern by (I think Walter) > > > concerning 5a2 so that it now only requires run-time identification > > > if the original used runtime identification > > > > Thank you. It is extremely close. It doesn't quite allow me to take > > out parts of the program and use it in "grep" if the original program > > identified itself originally. > > yes it does. because if you use parts of the original work as part of > something else you would do so under 5a1 so 2 would be not required at all. > remember 5a are alternatives.
5a1 is not a free alternative. 5a2 approaches that, but it has to cover _every_ occasion where 5a1 fails, not just most of them. Regards, Walter Landry [EMAIL PROTECTED]
