Date: Mon, 10 Feb 2003 18:49:33 -0800 From: Terry Lambert <tlambert2@mindspring.com> To: Matthew Dillon <dillon@apollo.backplane.com> Cc: Joe O <joeo@cracktown.com>, hackers@FreeBSD.ORG Subject: Re: Another EPIA M 9000 update (was Re: More compartive power/performanceresults (was Re: Lower power SMP boxes?)) Message-ID: <3E4864BD.EE2370A7@mindspring.com> References: <20030210120212.G47233-100000@clubfoot.cracktown.com> <200302102042.h1AKghcu023195@apollo.backplane.com> <3E4855B5.F8C90526@mindspring.com> <200302110204.h1B24IUD027279@apollo.backplane.com>
next in thread | previous in thread | raw e-mail | index | archive | help
Matthew Dillon wrote: > :> The "vga" driver works in low resolution modes. The "vesa" driver > :> does not work. Via has a linux driver on their CD for X, called "via", > :> which linux people seem to be using successfully, but I can't find > :> sources anywhere. I don't understand why these companies don't just > :> include sources for their X drivers, it would make life so much easier. > : > :The do not because then people could leverage their work by > :building hardware which does not license anything from them, > :but operates compatably. The same reason Adaptec developed > :their "HIM" layer, to prevent people from using Adaptec SCSI > :drivers with non-Adaptec hardware, and getting all the work > :they did to get the driver into the Windows base OS, for free. > : > :Basically, it's done to amortize non-recurring non-developement > :related collateral business costs. > > This doesn't make any sense to me. There are a huge number of > open-source drivers available, why would a third party want to > "steal" the hardware layer to VIA's hardware just to emulate it? > Why not some other hardware abstraction that is already available > in open-source form? From a business perspective I just don't > see how this could possibly effect VIA's bottom line. It isn't > rocket science we're talking about here, it's a sodding frame > buffer. These people operate on very low margins. They can not afford to give things away. If they did not have to worry about producing documentation, or even drivers, internally, then they reduce their amortizable R&D costs by 0.5%, which is a significant fraction of their profit margin. It's the same reason a proprietary software vendor would not release R&D results under Open Source license: by spending $1M on R&D, and then giving the results away, they bootstrap any competitor to themselves by removing just that much R&D costs. Using the $1M number, if I'm a business operating on a 5% margin, and I expect to make $5M over a 1 year product lifecycle, then I give away the $1M in R&D, I now have two problems. First of all, if 5% is $5M, then the lowest I can possibly afford to go is a 1% profit margin, because that's required to recover my R&D costs, while a competitor can take 0.5% or 0% -- either way, they can undersut my prices, and I can't afford to compete. Second of all, if I didn't feel that I was building a product that could win in the marketplace -- and that product includes not only the end-user product, but the support systems and business systems behind it -- then I would be building something else. So I *honestly* believe I have intellectual property tied up in the interface design, and I *honestly* believe that I can attach a monetary value to this. Third, I have business processes which cost me to develop, which are generally matched to my product design. The closer the match, the lower my operating costs, the higher my profit. Part one of this is that I want to make it hard to copy these processes outright and be successful. I do this by not disclosing information about the processes, and by not disclosing information about the product: like a binary weapon, neither of these can be used against me, if my competitor does not have the other. Part two of this is that some of these business processes are preemptible, and I need to prevent that happening, or my employees are, in effect working for my competitor. For example, if my hardware design has a known flaw (it might even be intentional, so that if the design is copied, I have a "fingerprint" proving it), then any of my dealers would be able to correct this flaw for my competitors customers, and may in fact not know they are cutting their own sales margins by doing so. Best case, a customer called in Pirate Enterprises with a problem, and then Pirate Enterprises calls *my* support people and gets a fix for their problem -- I have become Tier II support for my competition! Worst case, the driver indicates my company, and the customer calls me directly with the problem -- I have become Tier I support for my competition! Basically, there are a lot of business reasons for this, and they all have to do with protecing myself from being screwed by my competition. Yes, I agree, in many cases, the belief-in-value-proposition is not justified: a framebuffer is a framebuffer, after all; but even if that's the case, you can't expect everyone to understand "Enlightened Self Interest", or every piece of software with tactical value, but no strategic value, would be Open Source, and there would be clear lines of division in interface definitions between strategic and tactical components. Companies whose belief-in-value-proposition is not justified will, eventually, lose out to some other company. 9 out of 10 new businesses fail in the first year, and 8 out of 10 of the remainder fail in 5 years. Until then, they will do annoying things like treating tactical data as if it were strategic, etc.. -- Terry To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-hackers" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?3E4864BD.EE2370A7>