Joerg Schilling writes: > James Carlson <[EMAIL PROTECTED]> wrote: > > > For Linux, there is no documented DDI other than the source code. In > > order to write a decent driver for Linux, you *have to* read the Linux > > code. In order to do that, you must accept the terms of the GPLv2, > > and, as you use the information you learn in your work, you've now > > produced something that's "based on" the GPLv2 work. > > > > The GPL does not require you to accept it, just to read the code.
Again, unclear. You have no right to use the code at all, except as provided by the license. What constitutes "use" is certainly open to some interpretation. I think it'd be a staggeringly useless license if it didn't cover at least reading the code and applying what you've learned from it. Does the derived software include anything out of <linux/*>? Does it have programming constructs that the author could have learned only by reading the source itself? These are murky waters. I hope the people swimming in them have good protection. Brian Gupta writes: > What if someone were to attempt to reverse engineer the kernel > interface for Linux strictly by reading what is available on the web. > (Commentary) Doable in theory, probably worthless in practice given the kernel interface churn. No DDI == no reason for stability. -- James Carlson, Solaris Networking <[EMAIL PROTECTED]> Sun Microsystems / 1 Network Drive 71.232W Vox +1 781 442 2084 MS UBUR02-212 / Burlington MA 01803-2757 42.496N Fax +1 781 442 1677 _______________________________________________ opensolaris-discuss mailing list [email protected]
