Ingo Molnar writes: > * Alexandre Oliva <[EMAIL PROTECTED]> wrote: > >> On Jun 15, 2007, Ingo Molnar <[EMAIL PROTECTED]> wrote: >> >> > it is a false statement on your part that the executable "does not >> > function properly" if it lacks that part. Try it: take out the harddisk >> > from the Tivo (it's a bog standard IDE harddisk), put into a nice Linux >> > PC, mount it, modify a bit in the kernel image header and it will likely >> > still boot just fine on that PC. >> >> Ok, try this: take the disk out, remove/replace/modify the signature, >> put the disk back in, and tell me what it is that fail to run. > > you mean back into the Tivo? That is not support for what you claimed. > You claimed the "executable does not function properly" if it lacks that > part (and you did not qualify your statement with anything). That was a > false statement, because it still works fine in just about any > bog-standard PC. A true statement would be: "the modified executable > does not function properly _in the Tivo_". It still works fine on a > general purpose PC.
I claimed that. Unless I missed something, Alexandre did not. Ability to run on a standard PC is irrelevant. Tivo distributes the executable for the specific purpose of running on their hardware. Having the signature accepted by the hardware is a critical aspect of the executable. That purpose and function are what make the signature part of the work based on Linux. Courts consider purpose and intent when analyzing actions; except when one has bought the best available legal system, they would not follow your logic. (The role the signature plays in controlling access to a copyrighted work, per DMCA, might also separately identify it as part of the work based on Linux.) If I wished to distribute a kernel with extended functionality from a C file but not the C source files, under your logic I need not give them out -- a user could modify the binary and run it on a general purpose PC. Right? At most it would take clever linker tricks to make the change small enough. As to the suggestion that vendors would use another kernel: I would not mind. A huge fraction of the interesting and useful work in open source kernels happens in Linux (first or only). Using any third party software is a trade-off of what you get versus what you give up. Michael Poole - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [EMAIL PROTECTED] More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
