Date: Thu, 2 Oct 2008 10:11:32 -0400 From: Robert Huff <roberthuff@rcn.com> To: Da Rock <rock_on_the_web@comcen.com.au> Cc: freebsd-questions@freebsd.org Subject: Re: Questions drivers for VGA and NIC Message-ID: <18660.54932.394443.382443@jerusalem.litteratus.org> In-Reply-To: <1222951733.3927.27.camel@laptop1.herveybayaustralia.com.au> References: <e465331f0810011118i536c3e4kc1027c792bcc0754@mail.gmail.com> <20081001232502.G56202@wojtek.tensor.gdynia.pl> <20081001180424.56e6ca69@scorpio> <200810011436.27018.lists@rhavenn.net> <20081002072622.0284a05d@scorpio> <1222951733.3927.27.camel@laptop1.herveybayaustralia.com.au>
next in thread | previous in thread | raw e-mail | index | archive | help
Da Rock writes: > I apologise for jumping into this thread mid way, but wouldn't > the problem be simply a case of nil NDA? If an FOSS programmer > signed an NDA with say NVidia, then wouldn't the hardware > supplier be more willing to supply more specific details? > > Anyone with experience in the legalities here? It's not just the legalities, it's the philosophy. Accepting the N.D.A. would allow the writing of a driver ... which would - based on what I know about siilar N.D.A.s - have to be released as a binary. Now that could happen, and be a working solution; it's worked for other products. (Examples are left as an exercise for the reader.) But it's the (rare) exception and not the rule for a reason. Ignoring philosophical disagreements, it makes it harder to find and fix problems. Case in point: FreeBSD supports Intel non-CPU hardware; in at least one case (the /em/ network driver) Intel not only writes the code (for free) but releases it under an appropriate license. This gains Intel a lot of cred. (The fact that it's superior code on a superior card doesn't hurt, nor does tha fact that the writer is available, responsive, and friendly.) Robert Huff
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?18660.54932.394443.382443>