Manoj Srivastava <[EMAIL PROTECTED]> writes: > But then everyone else who is saving their time by using Sven's > driver would have to duplicate it, and that may be a significant > amount of time lost that culd have gone towards something more > useful (anyone who can generate a new driver has a modicum of > technical competence). Even worse, someone transitioning to Debian, > helped by that driver, shall no longer have access, and not yet > being able to compile their own drivers, leave -- and we have lost > a future contributor.
I've also wondered lately about about cases where the software is legally required[1] to be nonfree. What got me to thinking about that was the madwifi drivers for the Atheros a/b/g chipset. Presuming I understand the issue correctly, there is a small bit of code that is not, and can not legally be made publically available, at least not in the US. That's because the Atheros chip is a flexible piece of hardware, capable of doing all kinds of things in the radio spectrum that the FCC (at least in the US) forbids. So although I believe the developers of the madwifi drivers actually have the source to the bit of code that configures the chip to operate within the regulations, they can only release the resulting object file. The rest of the driver code (nearly all of it) is completely free. I suppose one could claim that Atheros just shouldn't have created such a flexible chip, and should have hard-coded the the chip for the current (each?) regulatory domain. I suppose in that case, you would be making a strong distinction based on the actual storage (or perhaps execution) medium -- of the code EEPROM (or ROM) vs RAM. [1] Of course IANAL, and I may well be misunderstanding the state of affairs -- i.e. is distribution in this case actually illegal in the US, or is Atheros just covering themselves? -- Rob Browning rlb @defaultvalue.org and @debian.org; previously @cs.utexas.edu GPG starting 2002-11-03 = 14DD 432F AE39 534D B592 F9A0 25C8 D377 8C7E 73A4
