Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1753186AbWLYAJw (ORCPT ); Sun, 24 Dec 2006 19:09:52 -0500 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1751909AbWLYAJw (ORCPT ); Sun, 24 Dec 2006 19:09:52 -0500 Received: from inti.inf.utfsm.cl ([200.1.21.155]:48379 "EHLO inti.inf.utfsm.cl" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1753186AbWLYAJv (ORCPT ); Sun, 24 Dec 2006 19:09:51 -0500 Message-Id: <200612250009.kBP09jlE004582@laptop13.inf.utfsm.cl> To: Rok Markovic cc: LKML Subject: Re: Binary Drivers In-Reply-To: Message from Rok Markovic of "Sun, 24 Dec 2006 22:05:08 BST." <458EEB84.30509@kanardia.eu> X-Mailer: MH-E 7.4.2; nmh 1.1; XEmacs 21.5 (beta27) Date: Sun, 24 Dec 2006 21:09:45 -0300 From: "Horst H. von Brand" X-Greylist: Delayed for 00:14:44 by milter-greylist-3.0 (inti.inf.utfsm.cl [200.1.21.155]); Sun, 24 Dec 2006 21:09:47 -0300 (CLST) Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 3334 Lines: 71 Rok Markovic wrote: > It seems that debate around cars and drivers has gone too far (IMHO). Ditto. > I > do not think that there is a question if we have any right to demand > details about hardware from manufactorers -> we are NOT. OK. > But I think that > we have right to know how to use it. The card is designed for use with /Windows/, you get a /Windows/ driver. That your PC is able to also run, say, Plan 9, is not the manufacturer's business at all, it is /your/ choice, and /your/ problem if what the manufacturer provides doesn't help you there. The manufacturer is keeping his part of the deal. You don't like the deal, tough luck. > I will translate this to CAR > language - We have to know how to drive the car but not all the details > how is this done, so we can drive a car without the "driver". We do not > need an exact knowledge about engine, ECU, suspension,... Exactly. > Now the real question: > > Why are manufatorers afraid to give us this information? Not "afraid". It is more expensive to them: - Because they would have to pick developer's brains and write it down, make sure it is complete, check it for publishing, etc. That costs money (and ties up key developers, and...) for /very/ little gain (open source systems is something like a 5 to 10% of all systems, mostly servers where highest performance isn't asked for, so the market for current high-end cards (where the brains are there at all for picking) is /extremely/ small). - Because they would have to get permission to do so from third parties, that means paying real money for little gain - Because sometimes it is just "we tried several combinations, this one(s) happen to work, dunno why". How do you write something like that down? - Because wrong settings might break the hardware, people fiddling around would then want a replacement, a very real cost - Because you can't write any software at all without violating some patent. In this area, it is probable that everybody is violating everybody else's patents, and making that easy to find out (via source code or specs) opens you up to lawsuits. Lawyers (and common sense) will tell yo not to go there unless it is very worthwhile. And it isn't (see above) - See the bletcherous workarounds for many device bugs (or downright design braindamage) in the in-kernel drivers. They might be just embarrased by the junk they shove out the door (We all know it happens with software, right? Hardware is much the same...). And they can't just work a year or so longer to get them ironed out, by then they could be right out of business. -- Dr. Horst H. von Brand User #22616 counter.li.org Departamento de Informatica Fono: +56 32 2654431 Universidad Tecnica Federico Santa Maria +56 32 2654239 Casilla 110-V, Valparaiso, Chile Fax: +56 32 2797513 - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/