[ILUG] Help with nVidia drivers install FC2

[Bryan O'Donoghue]

Paul Jakma wrote:
> On Wed, 25 Aug 2004, Bryan O'Donoghue wrote:
> 
>> In my defence, what I'd respond to there, is that, in the real world, 
>> it's much easier for me to pop out to the shops in a Town like Tralee 
>> and *just buy* a card that works under Linux.

> Right, essentially you dont care at all about binary-only drivers

No that would be you, drawing your own conclusions. I find what you just said 
quite contradictory.

On the one hand you say that Open source exponents can't live in the real world 
without compromising sometimes, yet equating using the Nvidia driver (for 
example) to that stated belief on your part, doesn't seem to equate. Perhaps 
that's overharsh, but that seems like a logical disparity.


> pragmatism-comfort-level, on the scale of short to long term is set all 
> the way to 'right now, this minute'.

Not true. While *you* might have a good experience buying stuff from the 
internet, my choice is to be safer and buy locally, because there's an easier 
avenue of recompense for me if the thing doesn't work. Fine *you* don't have a 
problem with buying your hardware in this manner, but, it's certainly not my 
first choice.

> I'm honestly not trying to be personally insulting or anything. just 
> pointing out what is almost a fact: If it comes to "binary only 
> hardware, today" vs "open, next week", you choose the former.

No that's an incorrect statement. I distrust my ability to redress problems 
with hardware easily from the retailer over the internet. My experience with 
hardware is that *sometimes* it is necessary to redress problems with it, with 
the retailer. Furthermore, if one rules out internet buying (which is the crux 
of your "you prefare the easy option argument", then from the available cards, 
the *only* PCI card found in the four computer shops in Tralee (during the 
buying period), was the Nvidia. Irrespective of the OS, to be used, the Nvidia 
was my only choice. I could have gone to Dublin, for sure, however sometimes it 
would be *months* between Dublin visits and sans-transport, I'd be limited to 
the shops in town to satisfy my needs with no gaurantee at finding what I 
needed to buy.

Essentially I stand by the choice I made, not as a "I have the right to do as I 
please", but as the only 'reasonable' option available at the time, at least in 
my opinion.. which I'll admit is as subjective as anybody elses.

>> didn't do what I needed them to do or were just broken.. so buying 
>> from shops is much *safer*.
> I've never had problems with online retailers. I've had far more 
> problems with brick-and-mortar retailers.

Really? That's seems logically unsound. What do you do with an online retailer 
in Germany when you want to return goods, but he/she/they doesn't agree with 
you that said company is culpable for the return?

In a local shop, you march inside and start making noise... until ... in order 
to shut you up and stop scaring off the other customers... the shopkeeper relents.


>> Yes I'd prefare to have that theoretical control over the system. From 
>> where I saw it, there was a known Linux and FreeBSD driver, which 
>> would satisfy my 'need' on a pragmatic level, if not 'higher 
>> theoretical' desire to have a totally open and controlled system.
>  
> What theoretical? You have it, today.

It is desireable for have all aspects of a system open to scrutiny as an ethic.
In practice sometimes that is not possible, as is the case with some vendor 
supplied magic potions, that you can't easily walk away from.

>> mentalities don't 'totally' equate.. I think there is sufficient 
>> correlation, to draw a comparison of sorts....
> 
> 
> Such as? ??

I made the case that Nvidia supplied binary only drivers are _as much_ an 
abrogation of Open Source paradigms as SuSE supplied Yast, before Yast became 
Open Source. However I pointed out that the corporate mentality of SuSE also 
led SuSE to support many Open Source projects financially, so the transgression 
  from our stated aim of a 'transparent system' was negated .. in terms of 
SuSE's corporate mentality versus that of Nvidia. Furthermore I was attemtpting 
to make the point that while one maybe *less* culpable then the other.. that 
logically speaking, using SuSE with Yast is a logically (!transparent) as is 
using Nvidia binary only drivers.

Perhaps you'd say that Yast != Nvidia driver.
However what I'm saying is that

Transparent_System != Transparent_System, yet I don't see you up in arms about 
Yast, perhaps *you were* up in arms about that at some earlier stage, I haven't 
checked... but a quick google for "Paul Jakma" yast, does not lead me to that 
conclusion.

>> Unfortunately, I think that when Linux becomes more proliferic, that 
>> there will be alot more closed systems that simply run on top of it... 
>> because many programming houses couldn't/wouldn't be willing to give 
>> away their code and move their business model to one of support. Ergo, 
>> close source will always be 'some' sort of fact of life.... well 
>> probably.
> 
> 
> Yes, indeed. And that doesnt matter, as you point out it's "on top of", 
> that's fine. It's the "below" bits that are the problem.


It is?

How is that? That seems like a snobbish view at best.

So would it be fair to say you're proposition is that running Microsoft Word on 
Linux or even Microsoft packaging and distrubuting Microsoft Word-Linux is 
fine, doesn't violate the principals of "Openess and control" you've stipulated 
thus far, while the Nvidia driver does?

To me, that is logically spurious.

Either a system is Open or it's not, after that all that is happening is an 
argument over a shade of grey.

I disagree that using something like a VESA frame buffer under Linux

http://encyclopedia.thefreedictionary.com/VESA%20BIOS%20Extensions

where the source code to the firmware bios on the video card in question is 
*not* available is anything but a semantic distinction in your argument. I 
don't think that makes me a 'heretic' but, I do think that the thought process 
is logical. Perhaps you don't and if you don't the please do point out the flaw.


>> I did search locally quite extensively, the Nvidia card was the only 
>> PCI card I found, and fortunately it also had S-Video, which was the 
>> only *actual* requirement...
> Quite extensively? If you spent days searching for this S-Video out PCI 
> card, you'd have been quicker just going to komplett.ie ;)

However I just exaplined in detail my aversion to buying from an online 
retailer, when it comes to hardware.

Books, clothes, fine, something which I have experiences of frequently 
'needing' to take back, not fine.

>> Yes in theory.
> 
> 
> No, in practice. The various bits exist already to varying extents. Eg, 
> the only thing my AP is missing is the trusted part, which exists too, 
> eg, see the X-Box for a "trusted" PC platform.
> 
> No theory about it, all the pieces exist.


No. You can 'prove' that the trustworthiness of the platform is false, since 
the vendor can't 'prove' by way of source code, that the platform is trustworthy.

However for you to be able to 'prove' that it is Trojaned somehow, you'd' have 
to provide evidence, which you haven't.

It *could* be Trojaned, but it can't be *proved* to be so. Unless you _can_ 
offer proof now that the AP is Trojaned.

I accept that it *could* be. I accept that it is _not_ trusted, but, I don't 
accept it has this flaw until you prove it!


>> The Intel X-Scale for example, comes with chunks of code that get 
>> downloaded to the CPU and act as a 'network core'...
> 
> 
> Err, no. I think you're confusing the Xscale with the IXP2x00, which are 
> Xscales with added network bits, "microengines", which are fully 
> documented by Intel btw, TTBOMK.

I happen to have an IXP425BD on my desk.

http://ixp4xx-osdg.sourceforge.net/howtos/uClinux-IXP400-AccessLibrary-v1.2.2-Integration-HOWTO.html

To quote
http://dwdm.cis.nctu.edu.tw/project/WiMax_Gateway/NP425.htm

It is an :

Intel XScale RISC core up to 533 MHz

I have the 533 model.

Whatever you choose to call them, the magic code from Intel could be Trojaned 
for Echelon...
I can't prove that though.

> Also, you arguing from a flawed position, you're countering a 
> "everything *must* be free or I wont use it" position, which is not my 
> position, nor that of anyone else here.

Then why argue the toss over how *hard* I looked in Tralee for a Linux 
supported video card, if your position isn't one of "everything must be free".

Is work really *that* boring?


> Given:
> 
> - Firmware X is closed source
> - Driver Y is closed source
> 
> Premise:
> 
> - Open software is preferable
> 
> Argument:
> 
> - X is closed source
> - Use of X is accepted
> 
> Ergo:
> 
> - Use of Y is ok
> 
> does not follow.

Not logically true and you know it. Lets reduce the arugment to absurdity.

X_Firmware
Y_Driver

B_ClosedSource=1 -- constant

What you just said was.

X_Firmware=0, Y_Driver=0

Print_Accept(bool bArgument){
    if(bArgument)
         print "Logically Sound"
     else
         print "Logicall spurious"
}

Then you said
X_Firmware = B_ClosedSource

Print_Accept(X_Firmware) -- and said that would print "Logically Sound"

then you said

Y_Driver  = B_ClosedSource

Print_Accept(Y_Driver) -- would print "Logicall spurious" , which is false

> Just because CPU microcode, or CDRW drive firmware or .... is closed 
> source, is not a good reason to accept that your Linux kernel driver and 
> XFree drivers for your graphics card should be closed.

I don't accept it... I *live* with it. My preference is for Open source. When 
you use all alternatives such as LinuxBIOS, over and above the propiatery 
formats, then well and good, you can logically lecture on this topic, not 
before, in my opinion...

> Like saying "we put up with Dictator foo doing bad things, so we can put 
> up with ....". One suboptimal situation, however tolerable, does not 
> justify tolerating the same type of suboptimality in another context, 
> especially when the consequences in this other context are far graver.

That's a subjective argument. In my opinion. And the alternative you've 
suggested was not practicle at the time.

Now you may try to castigate me from a great height for that, but, I don't 
accept the logical validity of you doing that, until you , for example use 
LinuxBIOS on your motherboard.

You are dependant on your BIOS to adequately configure your system for you, so 
you are 'more' dependant on that then you are on, a video driver.

Lets rule out the silly argument about Trojaning. Quite simply you aren't free 
to change the system as you choose and when you use a systems which come with 
RedBoot, LinuxBIOS and friends only and then let it be known, why you've done 
it... so you can 'train' the corporate types, on how to release source code?

That's hypocritical. Now you might not find it hypocritical... *people* rarely 
do recognise hypocracy in themselves... but, I think it is hypocritical to 
castigate me for that, and *not* make all the 'requisite' efforts to get a 
fully compliant eCos only setup.

> ie the context of proprietary firmware in a very limited environment 
> versus code running in a full blown Unix environment, with libc, full 
> networking APIs, etc.

You're talking context. I accept that context is very important, however, I 
don't accept that somehow I've *sinned* by buying Nvidia, as opposed to buying 
'nothing' while running a propiatery bootloader, boot time apm setup enabler 
et-al is *less* of a sin in terms of propigating what we allegedly all want.. 
transparent, controllable, verifiable systems.


> I'm pretty sure you would. Note that code might (or might not) be 
> closed, however that architecture is fully documented. Intel are 
> actually good in that respect.

Nope, I may not view/modify the magic code for the IXP425BD, without selling my 
soul to Intel.


> Right, except that the short-term pragmatism of you and others has a 
> detrimental on long-term health of Free Software, which will eventually 
> impact you too.

Less detramental I'd argue then using say, the IXP425BD, with it's closed magic 
code.
I mean what's the real market for the Nvidia Linux driver <=100,000 individuals?

What's the market for this brand of X-Scale and it's progeny? Roughly the same 
I'd venture .. somewhere in the region of a million chips?

How does a million IXP425BD chips with magic Intel closed code ... not damage 
Open source? Or how does it damage it less?

Why not shout the house down about that?

> I'm not dermot. And I dont see how libdvdcss is relevant to core 
> hardware support ;). But yes EUCD is a problem. However, existence of 
> EUCD doesnt make libdvdcss illegal. EU directives are not law, they are 
> directives to be implemented in legislation in member countries. That 
> legislation can potentially still contain fair use provisions. I've no 
> idea what state EUCD implemenation is in.

I was post to two people for brevity.

Last I heard something like 5 countries (including Germany) had attempted to 
legislate based on this directive.

-- 
Bryan O'Donoghue
Embedded Software Engineer

Europlex Technologies Ltd
Clonshaugh Business & Technology Park
Dublin 17
Ireland

T:+353 (0) 1 2500500
F:+353 (0) 1 2500590
E:bryano at europlex.ie
W:www.europlex.ie


This message (including any attachments) is confidential and may be privileged. If you have received it by mistake please notify the sender by return e-mail and delete this message from your system. Any unauthorised use or dissemination of this message in whole or in part is strictly prohibited. Please note that e-mails are susceptible to change.  Imetrex Technologies Ltd shall not be liable for the improper or incomplete transmission of the information contained in this communication nor for any delay in its receipt or damage to your system. Imetrex Technologies Ltd does not guarantee that the integrity of this communication has been maintained nor that this communication is free of viruses, interceptions or interference. This communication does not create or modify any contract, and unless otherwise stated, is not intended to be contractually binding.  Views or opinions expressed in this e-mail message are those of the author only.