Intel to Increase Linux Support, Release Centrino Drivers

jonman_d writes "ZDNet UK is reporting that Intel has promised to increase Linux support by releasing Linux drivers at the same time it releases Windows drivers for its hardware. According to the general manager of Intel's Software and Solutions Group, Intel wants Linux users to be able to use their hardware as easily, or easier, than any other hardware on the planet." Pingla writes in with more good news: "Intel promises to release Linux drivers for its Centrino chipset at the same time it releases drivers for Windows. An article featuring Lindows (aka Lin---s) on CNet has more." Sadly, the Centrino support will most likely be a proprietary driver, but it's better than nothing.

Proprietary drivers

Sadly, the Centrino support will most likely be a proprietary driver, but it's better than nothing.

I'll take proprietary drivers if it means I can use the hardware I like with the OS I love to get work done.

Re:Proprietary drivers

I'll take proprietary drivers if it means I can use the hardware I like with the OS I love to get work done.

People should not accept this or we'll get into another situation like you have with NVidia. Get a brand new box and you can't even do a net install on your Nforce chipset box because you need the nvnet driver which is a proprietary binary-only module and the manufacturer of the motherboard may or may not have included a pre-compiled binary on a floppy for you to use, but it's most likely only for Red Hat 9, etc. Screw all binary drivers, I insist on open source drivers for everything. The only thing I've had to relent on lately is the graphics card since the Nvidia stuff is the only decent graphics card out there but the modules are binary only. Sadly, my Nvidia card is also the most unstable part of my Linux box and it crashes (hard locks up) at least every 2 weeks or so and I have to power cycle the box. Fscking Nforce craptastic Asus A7N8X-Deluxe piece of shit motherboard.

Re:Proprietary drivers

People should not accept this or we'll get into another situation like you have with NVidia.

Which is what? A ompany providing kick-ass drivers that give superior performance than that same hardware would give in Windows? What do you suggest using as an alternative to NVIDIA? Ati? HAH, good luck trying to get those drivers to work, open-source or not! And if you do get them to work, what kind of performance are you getting from them? And how about their AMD64-support? NVIDIA has AMD64-drivers available right now. Where are Ati's drivers??? Where are open-source AMD64-drivers for Ati?

Get a brand new box and you can't even do a net install on your Nforce chipset box because you need the nvnet driver which is a proprietary binary-only module

One word: Forcedeth.

Sadly, my Nvidia card is also the most unstable part of my Linux box and it crashes (hard locks up) at least every 2 weeks or so and I have to power cycle the box.

You know, you CAN use the open-source NV-drivers that ship with Xfree. Or you could use the standard VESA-drivers. So it's not like you are forced to use those drivers. I for one haven't had any problems with NV-drivers.

Re:Proprietary drivers

Regarding the 100 -> 300 fps needs for games. I don't know of a monitor with a refresh rate above 125Hz which means that your fps is limited to 125.

But yeah, 300 is better, definitely.

Re:Proprietary drivers

"A ompany providing kick-ass drivers that give superior performance than that same hardware would give in Windows?"

Superior performance as in shorter time from bootup to lockup?

Re: Proprietary drivers

He was talking about "NVIDIA card". Since when do you call Motherboard "a card"? He was obviously talking about his vid-card

Well actually he was talking about both -- the references about not being able to do net installs with Nvidia motherboards is because the module for running the ethernet card has to be downloaded from Nvidia's web site and complied for the kernel you are using before you can connect to the net via ethernet -- a situation that is really lame and sucky...

Re: Proprietary drivers

Well, When I installed my Gentoo-box, there were nForce-drivers (forcedeth) right in the Kernel (2.6.3-mm1), so I'm not sure what more do you people want? He then complained how his vid-card was unstable, and I merely suggested that he could use some other drivers. I also mentioned that I haven't had any problems with the NV-drivers.

No-one is being forced to buy NV-hardware. Hell, to be honest, NV is under no obligation to provide Linux-drivers for their hardware AT ALL! Yet they choose to release kick-ass set of drivers, yet people complain.

Would I like it if the drivers were OS? Sure. But I'm not losing any sleep because of it. If you don't like the current situation, feel free to spend your money on hardware that does have OS-drivers available.

Re:Proprietary drivers

Restart X? whatever for? I often run two different X sessions on my system, with different configurations.
startx -- :1
I've also done it with kdm, in fact when I was playing with the kdm source code I had two different versions of kdm running on my machine at once!

But he does have a point

If you want to play the latest games your stuck with these proprietary drivers. This is only tolerated by many in the community because its either use the binaries are don't play the latest games under linux. btw yes I am aware of drivers for ati's older cards. When it comes to linux and gaming nvidia is the status quo.

Now my main point is this could lead to some problems for us linux users. Like he pointed out its possible in the future that we'll all be stuck with mobo's that don't work unless we load a dozen proprietary drivers. We did without in the 90's and we can do without now. The nvidia, now the Intel, next the VIA chipsets, its a dangerous trend. You tried to deflate his point at the end by saying just the free nv or vesa etc. What about when that's no longer possible?

The way I see it is this. You should be able to install your OS, have it support your mobo chipset, video card, mouse+keyboard, and ethernet card all with Free software. You should be able to surf the web, get email, use a calendar and contact list, play movies and music, and be able to create Office documents all with Free software. Those are the basics. Anything less is a failure. Right now all of the above is possible. Start throwing in a Nvidia card, a centrino chipset, and the truly Free desktop starts disappearing. Right now its the not the end of the world. But if in the future proprietary binary drivers become the standard a Truly Free Desktop won't exist and there will be no point in using Linux. After all if I need binary drivers for my hardware like in Windows and I continue to use all of my Windows apps via WINE, wtf is the point? Just stick with Windows and the closed source model. Throwing an opensource kernel on top of all that proprietary software is a lost cause.

Re:Proprietary drivers

I've got that same motherboard. While I haven't fiddled with the other onboard lan card, the onboard 3com one can used with only a vanilla kernel, 2.4.22 or higher. (Don't think it's gotten into the 2.6 series yet) Just use the 3com driver.

Otherwise, haven't had any problems with my board. I'd hardly call it 'craptastic'. I've gotten much better and more reliable performance out of it than my previous boards, while using Windows or Linux.

Re:Proprietary drivers

You have a choice, restrict yourself only to hardware that provides open source drivers, use what is made available, or use another OS. Centrino users don't really have the option of demanding open source, if they chose to use the hardware that they want to use. You made the same choice with your video card.

If Intel's choices boil down to "release a binary driver or ignore Linux", which realistically, they do, I'd prefer that they support Linux in any way that they realistically can.

Intel is obligated to its shareholders to protect it's technology. Open source drivers could tip their cards to AMD or some newcomer could gain the upper hand. That is the REALITY of how the hardware business works.

I have had no problem with Nvidia drivers and stability, but I stayed away from the Nforce chipset due to the ongoing support problems it has had.

I too would prefer open source drivers, but I'm not going to threaten to hold my breath until I turn blue just because all they want to give me is binaries.

Re:Proprietary drivers

If Intel's choices boil down to "release a binary driver or ignore Linux", which realistically, they do

No. Full stop, no.

• Knowing how to drive a car does not mean I, myself, can build another just like it.
  
• Knowing how to replace a light bulb does not mean I can make one myself.
  
• Knowing how to read a map does not mean I am a cartographer.
  
• Knowing how to drive the roads does not mean I can repave Rome in a day.
  

On the contrary, all of those examples show how I am more likely to use and buy a product if I know how to use it.

If I could look at a product's manuals, and from that, figure out how to copy the product, then you can be quite certain I knew 99% of what I needed to know to make such a product beforehand.

For example, if I hand you a black box that takes two numbers as input and outputs a third, and you deduce that it's a multiplication box, you knew everything you needed to know to make a multiplication box before I even handed it to you.

On top of this, if you simply copy a competitor, you're a year behind them and dead meat anyway.

Re:Proprietary drivers

Giving away their shnazzy algorithm that's three times as fast as any other technique is just dumb

Algorithms? Techniques? Sheesh. Look at the average C header file or a reference document for some trivial microprocessor some day. It's mostly enumerations of functions and constants, with explanations for each one. That's the level of information we need, and historically, vendors who care about their customers give it to them for the asking.

It's as much "moving this bit over there" as posting this comment to slashdot is "moving this bit over there" - but we wouldn't be having this discussion right now if TCP/IP, HTTP, and HTML were undocumented and released as some binary-only implementation.

Who cares about their optimizations? We're smart, we can figure that out ourselves.

e100/e1000 began proprietary

and now they are in the kernels, and pretty much edged out the eepro100 drivers for intel nics.

So, even if they are originally released as proprietary, who cares, I bet the source will sooner or later be released.

Re:Proprietary drivers

What sort of "insistance" are you practising here when you are openly admitting that you use one of the prime offenders in the binary driver category? I'd say it's more like you "prefer" open source drivers but will take whatever you can get. That puts you squarely in the same boat with the others who don' really care if the driver is open or not, as long as it exists and they can use their hardware.

Re:Proprietary drivers

I believe that ideas like this, the parent, are the kind of ideas which halt progress. Their cries of "All or Nothing!" will much more frequently bring nothing than all that they ask for. This is what is known as a compromise. They knew that in an ideal world, we would like the full source code to the intel (and NVidia) drivers, however due to their corporate stance, they are unable to comply, due to their need to protect their shareholders and their IP. Instead of ignoring the userbase as 80% of the other hardware companies do, they instead make a peace offering. "We can't give you the full open source, but what we can do is let you use our device by giving you drivers written by the same organization who created the card". How many windows drivers are open source? Do you think that they have the ability to check their code? You mention that when you have to powercycle your box (every 2 weeks), it's the closed-source NVidia graphics card that causes it. That's so sad, I could cry for you. Reboot your computer every 2 weeks because you can't look at the source code of the driver for your top of the line video card. Other people tied to other operating systems should be so lucky. Stop whining, grow up, and learn how the real world works.

Re:Proprietary drivers

But there is something very morally wrong with proprietary software. RMS and co. have already written plenty about it; I have little to add, save that I feel it is one of my fundamental human rights to examine and improve on the source code for any software of my choice -- and if I do not agree to others exercising a similar right over any software I may write, then my remedy is limited to not writing that software in the first place.

Back in the days of drum memories, computer manufacturers would often offer you software gratis -- in return for which, you were expected to offer them something you had written, so they could pass it around to their other customers. This was a form of policed open source. With no such things as high-level languages, there was no distinction between source and binary; one line of assembly language translated directly to one word of machine code. Experienced programmers could read the ones and zeros as ups and downs on an oscilloscope screen and understand them as an instruction.

Then, somehow, sometime it all got stuffed up, when people began trying to treat ideas as property and earn money by litigation .....

Re: Proprietary drivers

People should not accept this or we'll get into another situation like you have with NVidia. Get a brand new box and you can't even do a net install on your Nforce chipset box because you need the nvnet driver which is a proprietary binary-only module

I totally agree.

I built a shuttle box for my little sister a while ago not realising that the Nvidia motherboard's built in ethernet card will only run with a module from Nvidia, it took a while to work this out after installing Linux on it the first time...

Then 6 months later I have a chance to upgrade Red Hat 9 to Fedora on the box and after the upgrade I discover that the network doesn't work... and at this point I remember what I had to do 6 months before... Aarh!

So I have to go through the whole process again, find another computer that is connected to the net, download the Nvidia drivers, burn a CD... I thought I'd try the SRPM to make upgrades easier, well these don't build as a normal user so I gave up on them, so I then need to download the tgz version, burn another CD....

This results in a situation where the kernel can't be upgraded without manually rebuilding the Nvidia modules and this isn't something that I would want to suggest to my sister (she never uses a CLI)... So the local root exploits that all but the latest Fedora kernal have don't get patched... (not a big issue since it's behind a NATed connection and there are only a couple of user accounts, but still it's not ideal...)

The result of this is that I'll never recommend that anyone gets a Nvidia motherboard and I'll never buy one, it's far too much hassle.

Sadly I'm stuck with Nvidia video cards in order to play games such as Quake 3 in linux... I wish this wasn't the case...

What would be ideal would be if the manufactures either release enough info so that GPL drivers can be produced or if they release GPL drivers themselves so they can be included in the kernel.

Last year I wanted a IDE RAID card and after much googling I discovered that the 3ware ones have drivers in the kernel and no others do, so I brought one even though it cost me more money it has saved me loads of time because I haven't had to mess with installing modules from a hardware company every time I upgrade the kernel... I have no regrets about this bit of hardware... unlike the Nvidia motherboard...

Re:Proprietary drivers

I would put it this way: The brand "Intel" should be less important for the buying decision than a GPLed driver for the hardware. I think there are several, real benefits for using GPLed drivers:

- fix bugs/do workarounds for the hardware the manufacturer doesn't care about
- tweak the driver to your needs (this is not a joke: I'm glad that the tmscsim-driver for Tekram SCSI cards could be tweaked by me to work seamlessly with my old SCSI scanner!)
- have support for the hardware as long as YOU wish

Re:Proprietary drivers

If you take the proprietary driver, it means that at some point you may not be able to get your work done. That's great that Intel is going to attempt an on-time release for Linux drivers once. But what happens every time the kernel changes? Or some system library changes? Or the compiler changes?

And I'm not blaming Intel for this one either. Hardware installation under Linux is a nightmare of inconsistency. If the shipped kernel doesn't fully support your hardware, good luck! The typical Windows user is still not ready to compile a kernel.

I sort of like what Nvidia does with it's video cards: The 'compile a small kernel interface on-the-spot' type of script. I'm sorry to hear about the fellow with the Nforce chipset problems, but Nvidia's video card drivers are solid.

Re:Proprietary drivers

The other things is intel take on the cost of maintenance and testing. Or at least, prerelease testing.

I worked on a C++ project for some future DVD+RW devices, and we wrote windows only last year, even though I did all my dev in VmWare under linux -I can tell if the technology takes off there will be complaints that we didnt bring out a linux driver.

But even a pure Win32 driver (a) reused lots of existing windows code (some with Win16/win32 #ifdefs to show its age), and (b) took a lot of engineering effort. I dont realistically think the company will rush to duplicate that effort for Linux, unless it is tangibly lost sales. Even then, it will take ages. The new code we wrote will be ok -its all std:: C++ stuff, but the public API (COM) and legacy stuff is a historic mess.

I hope the company does the right thing and just documents the new SCSI commands and let other people write the Linux stack on demand. No maintenance costs, no development costs, the first implementation starts of OSS and stays that way.

Setting an example

I don't think it matters if this is a proprietary driver, just yet. With big people like Intel and IBM showing an interest in Linux, its bound to encourage others to do the same. Then with time, open source drivers might just happen?

Re:Setting an example

No, it will probably be the opposite. As Linux grows in popularity, you'll see more and more vendors shipping proprietary drivers for their products. That's not a bad thing unless OSS is your religion.

Re:Setting an example

>That's not a bad thing unless OSS is your religion.

Or if you use something other than ix86 platforms. Vendors will probably not make binary drivers for CF-cards on my Yopy.
Although in this Centrino case this might not be a big issue.

Or you want to path your driver. I.e to allow TV-out on your graphics card. Or fix a bug.

Or you use a !Linux OSS OS, like BSD.

Or you use an old or experimental kernel.

Re:Setting an example

No, it will probably be the opposite. As Linux grows in popularity, you'll see more and more vendors shipping proprietary drivers for their products. That's not a bad thing unless OSS is your religion.

It's quite a bad thing, irrespective of religion, if the vendors don't release enough documentation of the devices to make open source drivers. We'll end up in a situation where it'll be difficult to install Linux/*BSD on a machine whithout proprietary drivers. As an example, for the NForce chipset I've to buy a NIC due to lack of driver.

As documentation goes, Intel network division is very bad : they release GPL drivers, but no documentation is given (without NDA). That makes it difficult to make good open source drivers. And now the same company wants us to accept more and more hardware components with only a vague promise of drivers, much less documentation?

Re:Setting an example

I disaggree. The problem with proprietary drivers is they never keep pace with OSS development, hell, I can't even submit a patch before somebody else has done it nearly 99.9% of the time. Things just move too fast.

You want to upgrade to new fancy-schmancy kernel 2.7.x and you can't because your CPU-centr-a-yummy needs 2.6.x to install properly. They never keep up or give anything more than half-assed support. I had an nVidia TNT2, and I gave up on nVidia stuff, because I hated being locked in to THEIR schedule... and it crashed a lot, would corrupt the video (you could log in remotely though) and the only thing I could do was reset. I moved on.

Re:Setting an example

And then you can vote with your wallet and show these companies that you care about having open source drivers. It's really simple; if they supply open source drivers then consider buying their hardware, otherwise, don't.

The people who don't care can do what they like at purchase time and they should have the ability to get their closed source drivers so they can use Linux too. It's all a stepping stone to going completely open source. That's not to say closed source should ship with the kernel, because it shouldn't - that position is reserved for open source only.

Re:Setting an example

Well, OSS is not a religion. But remember RMS started GNU because of just this -- closed source printer drivers.

What Intel is doing is doubly bad:

1) They are releasing a Closed Driver, killing future development, growth and porting of support to future systems.

2) Really only doing this to spite Lin--s. They are doing this to STOP Lin--s' open-source driver development.... probably not because they want to. Why couldnt they have been forthcoming "we are working on a driver. we intend to release it first qtr 2004. we are making it closed." Why keep the FreeSoftware universe in the dark..? Because they want to hold all the marbles, withholding information is dishonesty. Plain and simple. If you want to be 'trusted', keep no secrets.

Re:Setting an example

I don't think it matters if this is a proprietary driver, just yet. With big people like Intel and IBM showing an interest in Linux, its bound to encourage others to do the same. Then with time, open source drivers might just happen?

That will take much longer if non-free drivers are available. Intel said somewhere that they won't release the driver as free software because they fear that this would reveal too much information about the hardware itself. So when Intel is out, the driver has to come from a third party. And clearly, the urge to develop a free driver is much lower when there is already a proprietary one available.

Big Hurdle

Many claim that linux is held back by several factors including ease of use, interface, etc.. etc... I've always felt it was hardware compatiability. You could never be sure all of your hardware would work easily, and the average user can't try and go and build their own custom drivers, or even download them. This will certainly put pressure on the rest of the hardware manufacturors, and this could help linux take a few more points in the market share. No, it's not the magical answer, as their isn't one, but it's another start.

Lin---s

Looks like the censors finally beeped out the profanity.

Nice, but...

It's nice that one of the giants to adopt this position, but I wonder about the form of these drivers. Maybe it's me, but I find more convenient to have drivers that can be compiled as kernel modules, and diffently from, for example Nvidia drivers, that they're not closed source, and license-compatible with the Linux kernel, so people can contribute in order to improve them, and maybe who knows if they can be integrated in the Linux kernel tree. Maybe i'm being too idealist.

Reverse engineer the drivers!

If you really care about freedom, then help reverse engineer the drivers. Several drivers have already been reverse engineered (such as nvnet for example), whats so hard about a simple wireless network adapter!

Re:Reverse engineer the drivers!

IIRC one of the prime reasons Intel won't release open source drivers is because the hardware chipset is capable of broadcasting on frequencies that are reserved for police/fire/etc and at higher output levels than is presently legal. Open source drivers would ease the path to hacking and utilising these hardware features.

Re:Reverse engineer the drivers!

That's a security problem with Intel hardware don't you think ?

Security through obscurity doesn't work. It's possible to make a safe hardware, so that reverse engineering (as rightfully understood by the EU laws) is applied and legal for compatibility. And even (gasp) open-specced hw.

Even by hardware, I do not mean realy that : software doesn't mean driver OR app OR os, you have firmware too.

Dumb

> Thus far, the company has been hesitant to ship an open-source driver, based on its concerns that showing Centrino's underlying programming instructions might reveal previously unavailable information about the wireless networking technology.

Yeah. Because obviously no other companies have been able to produce wireless networking products. I can see the point of commercial secrecy when you have some l33t hardware that no-one else can make, but when you just have yet another implementation of something that's already widespread and implemented in lots of different ways it seems dumb to worry too much about protecting it through drivers. If the other companies cared enough about your particular methods they'd just get a team of coders to reverse engineer the closed-source drivers.

Embarassed Intel into it using NDISWrapper

Initial Centrino support in Lindows will be based on existing Linux adaptations of Windows drivers, Lindows CEO Michael Robertson said. "It works great," he said. "It's invisible to the consumer."

It's sad that Intel had to be embarased into fullfilling the promise it made to support Centrino (when they annouced the chipset) by having someone ship a NDISWrapper with their Windows driver in a distro...

I'll believe Intel, when I see it the driver...

Intel gets it IMO

I have used several Intel Linux development products for several years. The C++ compiler, performance primitive library (IPP), and VTUNE are all extremely excellent products, and well supported under Linux.

It would be nice if VTUNE would be brought up to equal footing with VTUNE for Windows, but it's pretty good as is.

Why would 'Proprietary Drivers' be so 'sad'?

Seriously, what makes it so sad?

Intel can do what they want. They are the owners of their hardware designs and the drivers to make that hardware function.

If it's so sad that Intel is going to provide proprietary drivers, do you get sad everytime you get into your automobile? (The computer under your hood mosty likely uses proprietary drivers to interface with the autmobile.)

There is room for both open and closed software in this world. I for one envision a world where the Operating System is wide open with all the tools one needs to make whatever changes they wish to it and to develop whatever they want to on it. If hardware manufacturers want to keep some or all of their drivers 'secret' that's fine, let them. If application developers want to keep their 'Whiz-Bang 2.0' application proprietary, let them.

Believe whatever you want. I have and still use quite a large amount of both proprietary and open source software and in some cases, the open source software is better, in other cases, the proprietary software is better, even for the same task.

What needs to end are silly proprietary APIs put into an OS by particular vendors to allow their other applications to run like the dickens while making competitor's applications less capable.

Re:Why would 'Proprietary Drivers' be so 'sad'?

Actually I do get sad when I get in my car with a proprietary computer under the hood. I enjoy "tinkering" and doing minor maintenance to my car, but something as simple as an oxygen sensor now requires a $50 trip to the dealer to tell me this is the reason why my check engine light is on.

Re:Why would 'Proprietary Drivers' be so 'sad'?

There is room for both open and closed software in this world.

Yes there is. That does not mean that the choice is value neutral, however.

The licensing of the relevant code is a part of the feature set just as much as the checklist items for the hardware is. It is another item that the customer needs to evaluate and contrast with competing offerings.

This is why the anguished cries of some manufacturers against governments requiring open source rings so hollow. Just as a customer can require for instance Word file import capability, or three year installation and upgrade support, they can require open source compatible licensing. It is another feature that may carry more or less importance depending on the customer.

So, if someone says they will not consider hardware without open source drivers, that just means they, for various reasons, value the feature of open source relatively highly, and are ready to pick another supplier to get the feature they want. Note that it really is not just about whether open source or proprietary software is better; the licensing is in itself one (sometimes major) factor in determining the "betterness" of a piece of software.

Re:Why would 'Proprietary Drivers' be so 'sad'?

Those willing to give up a little liberty for a little hardware compatibility deserve neither drivers nor liberty.

Give us documentation... instead of closed drivers

"...Intel has promised to increase Linux support by releasing Linux drivers at the same time it releases Windows drivers for its hardware"

I doubt that they will open souce their drivers. So the Linux developers will write their own anyway, whenever they can.

And personaly, as a user, I find open source drivers much more convenient.

proprierty drivers

what is so secret about them, really?

To use them for your own hardware, don't you have to create the exact same hardware? So no use there, since you have your own hardware...
To use the hardware independet part of the code? Well.. that ought to be a lot of code.

To use their algorithms? Well, there are a lot of code already they can have a look at (without telling they looked at it, if they are evil)..

And if they are to stupid to come up with an algorithm of their own, how expencive would it be to hire someone to do it?

I don't get it...

Re:proprierty drivers

The problems reside in keeping trade secrets and sometimes even in using licensed code in your drivers. Imagine if you licensed some sort of technology from someone for a cost for use in your drivers. I'm sure they wouldn't be happy if you distributed (for free) their code along with yours since it would be required to compile the code.

Re:proprierty drivers

There are issues with regulations: Many WLAN chipsets are within regulation boundaries only due to software control. With a completely open driver, people could use channels which are off-limits for WLAN or boost the signal strength. Many drivers are therefore split into a firmware part which is kept proprietary and executed on the card and a host part which is open and runs on the system processor. Sometimes this is not possible because some chipsets rely on the system processor even for low-level control. Splitting an existing driver into proprietary parts and open parts can take some time too.

I just don't get the idea of the Centrino anyway

Who cares if wireless is built onto the CPU. I sure don't. Plus the Centrino is outdated technology. I wouldn't buy a new laptop that didn't support 802.11g anyway.

Re:I just don't get the idea of the Centrino anywa

Centrino is a set of 3 chipsets.
Intel(R) Pentium(R) M processor
Intel(R) 855 Chipset Family
Intel(R) PRO/Wireless Network Connection.

These 3 parts make up the Centrino, it's not just the wireless part.
There is already full support for the processor in the 2.6 kernel
The 855 Chipset is also supported
The PRO/Wireless is what this is all about. Intel has been saying that they will be supporting the wireless for the last year and we have not seen a thing. The best chance we have currently is running a wrapper for the Windows drivers, this is not bad but not good either. If Intel can deliver a driver that gives Linux users FULL functionality with the Wireless, I know I will at least be happy.

open source drivers

having this drivers open is not only for nerds or geeks to be happy. it would probably help the kernel people to include a better support for it.. possibly having it better integrated in the kernel..

proprietary drivers

A bunch of people in this thread have already posted responses that say things like "I don't care if a driver is bianary, I want to use my hardware, the only people who care are free software zealots."

Bullshit. Proprietary drivers are a bad idea for linux. Now I have to say, the licensing issue does matter to me. Even if you don't care, there are plenty of technical reasons to avoid them and pester a company to release the source for their drivers. First of all, the code is usually sub-par. EEs right them, they're smart people, no doubt, but most of them aren't programmers and lot's of bugs and race conditions show up. The OSS community can't help debug them because we don't have the source. Furthermore, on a more personal level, most of the kernel hackers don't give two shits about proprietary drivers, because of that, they generally stay buggy and improperly maintained. Intel is a big enough company that they'll properly produce high-quality drivers; however, it is simply a fact that letting the OSS community have the source would increase their quality, more eyes looking at the code, and they would be the same people that have written the kernel. These debates flair up all the time on LKML. I was too lazy to go look for links to specific discussions, if you're interested in the issue however, they wouldn't be hard to find.

- Ryan, who can't remember his password right now, and so posted AC

Re:proprietary drivers

And Security. Dropping someones' closed drivers in your kernel means you cannot do an effective audit. You can *never* be certain you've not bee backdoored.

Would Intel do this? maybe, maybe-not. But no one expected it from Borland's interbase [com.com]

is this paranoid? maybe, maybe-not....

Intel Centrino Drivers: A Series of Rumours

Intel is announcing plans to release Linux drivers for the WLAN part of their centrino technology from the time beginning. Though there are no facts yet, no release date, no statement whether the drivers will be binary only or Open Source, no information which chipset generations will be supported eventually and so on. See details of the story and How to Get Linux Running on Centrino Laptops [tuxmobil.org] at TuxMobil. So don't miss to sign the Linux Support On Centrino Petition! More at the link above.

To everyone saying whats wrong with proprietery.

Read this. Here is the problem. The kernel developers arent GPL zealots like RMS is, but closed source modules are a problem for them. If a kernel module crashes, and it is the propreitery modules fault, then they can't find out whats wrong and unable to sort out the bug. That is why since 2.6 the kernel developers discourage acccess to the kernel. By opening the drivers, drivers can be more stable on your system.

To those who say, but Windows DRivers are closed. They are not to the kernel developers. When installing new drivers you may of had a warning that a driver wasnt signed. A signed driver means one that has had its source code audited by MS for bugs, and is more stable than a unsigned one. Microsoft dosent like closed source (unsigned) drivers, and will warn you if you try to install it.

So if you want a stable Linux, don't load closed source modules into it. Dont take unstabllity for short term hardware support over stabillity in the long term. Encourage companies to open their source, or reverse engineer and stablise their drivers!

Will other organizations follow suit?

Kudos to intel. Even if the drivers are proprietary, at least they are releasing drivers tailored for their hardware to run under linux. This has been a concern of mine ever since recently switching from the Windows world to linux. Although I may sound like a typical end user when I say this, when I switched from Windows to Linux, but it was an extreme pain in the tail to even configure a sound card in Linux. I know there are things like ALSA and similar projects, but hardly any organizations were packaging any of their own drivers customed suited to their hardware to be delivered in Linux. The result is that the novice user loses from this because he/she cannot use the hardware he/she chooses to use with the software and/or OS he/she chooses to use.

With that said, this is a step in the right direction and I hope other hardware manufactures do what Intel has pledged to do. Closed source, proprietary drivers are better than no drivers at all.

what's wrong with Intel?

I cannot see any excuse from Intel to wait a whole year to get out a drivers for linux.

It might be a small marked, centrino together with linux, but they are pissing off a lot of people unnecessary. Many of these people have influence in companies buying computer hardware, not only laptops but servers and workstations. Good way to make the bias towards AMD stronger.

My job gave me a dell laptop where I am not using the wireless at the moment (I don't dual boot). I am reminded everyday why the next server will be opteron since I am in charge of buying the new one.

Easiest

If they want to make it easiest then they should submit code to the Linux kernel. That way the next version of almost every Linux would support that hardware straight away automatically.

HAL: Proprietary, SML: Open

Though it's not an open-and-shut simple approach, one can imagine a closed hardware management layer, driven by an open, developer-manageable O.S. software management interface layer. This doesn't solve the instruction-set incompatibility problem, but it is possible to let open maintainers handle the work they are (very) good at: Accommodating changing kernel interfaces, races, etc.

Linus is on record stating that as uncomfortable as it is, proprietary binary-only software can be linked into the kernel as long as it is not a derived work, meaning not depending on any interface provided by the kernel.

So Intel can preserve their private, secret register settings, providing a controlled abstraction of the hardware, and still tolerate, to some extent, varying kernel interface requirements.

Intel once again behind the 8-ball

Centrino? You mean 802.11b-onna-chip?

That is so 15-minutes-ago.

802.11g is all-the-rage, there are proprietary (I cannae give ye much more, cap'n) extensions to g which give it even more KickAss throughput and already Intel even are trying to jumpstart "more wireless speed than you would know what to do with" mode AKA UltraWideBand based technologies.

Somebody releasing half-assed (in the sense that we have to rely on them to provide timely updates, because it's not open source) drivers for last-years wireless technology is not in any sense of the phrase "stuff that matters".

On this kind of timescale I expect we're soon going to have our own OpenSource (we worked it out for ourselves, thanks for nothing) drivers.

Intel is a large enough company making enough profit that they could easily afford to provide current-and-up-to-date drivers for their wireless technologies as they release them not whenever they're no longer busy doing "important stuff".

Intel, you're half-assed. Period.

Behind the 8-ball when it comes to 64bit (busily playing catch-up to AMD) and can't be bothered getting out drivers for your technologies.

Here's a clue

• hardware with drivers sells more units
• more sales = more profits

Intel, please just plain get up off your fat hairy ass and deliver drivers (we'll live with proprietary if you insist) as soon as the hardware is available on the shelf and provide timely updates for new OS releases (dammit man, it's not like we're releasing a new MAJOR kernel every month) Yours truly The Community (aka Your Customers)

Graphics specs too please! Not just wireless

The situation is pretty infuriating with the video drivers for laptops with integrated graphics on 855GM chipset. Many of these come with a 1400x1050 SXGA+ lcd display but a bios that does not know how to switch to this mode. (No kidding, it can do 1024x768, 1280x1024, etc, but NOT the native lcd resolution...) Intel has not released specs to let the XF86 developers program the video modes from the driver, so X Windows is entirely dependent on the BIOS.

Result is your spiffy new SXGA+ laptop with Intel integrated graphics can only do a fuzzy interpolation at lower effective resolution. Needless to say, the Windows driver authors had all the info they needed to program the driver.

And you guess what trouble you will have getting the laptop to display on an attached external monitor....

Intel needs to provide specs to the XF86 developers, so that they can provide good drivers for Linux!

I don't mind the proprietary*(sp) drivers.

Open source is the way to go for software, but it may not be the best way for hardware. There is far fewer hardware makers than software makers. If somebody doesn't like the closed source software somebody has already made an open source version. If somebody doesn't like the hardware being closed source then they can use a different peice of hardware. The only problem with that is there is a limited ammount of hardware for some applications.

Let companies provide us with their drivers in any form that they choose. Chances are they will be better than using wrappers or an opensource driver that does not utilize the full capability of the hardware.

I have already paid for my hardware, be it new or used. If I can't find a driver to use it under Linux or *BSD then I won't buy it. I can't afford a license for a competing operating system, as a result I can't afford some hardware. With intel supporting hardware under linux this give me and many other college students a break on our wallets. Now all we need to do is purchase our student copies of Codeweavers Crossover office (www.codeweavers.com) and MS Office if needed. I have a database class and all the databases are Access so I need MS Office to make sure my databases are 100% like the professors. I would use an opensource program but sometimes they don't save the database correctly.

So let the hardware companies support linux in any way they choose. It makes Linux look more attractive to the average user or company. The more users, the more (and better) software out there.

~ryan

Want your Cake and Want to Eat it Too

You can't have it both ways. Either all you have is open source and you scare off a lot of commercial interests, or you have to live in a mixed open/closed environment. Do you want support for the most hardware on your box? Do you want desktop Linux to succeed?

Intel, NVidia, etc. have spent millions of US$, Euros, etc. developing their hardware and software. They need a competitive edge to their products (real or imagined). If they can protect their interfaces, at least for a short period of time, they can stay ahead of the competititon (or try to).

On the other hand, Open Source including Linux needs the broadest support possible. Restricting the O/S to only closed drivers will scare traditional companies away (and already has in come cases, think Canon printers). It will limit the accessibility to state of the art HW and SW. Much of the performance gains in modern hardware are due to the software drivers (graphics comes to mind). If you give away all your software, you weaken your position in the market and it can affect your bottom line.

The primary objective of a company is to maximize shareholder's wealth. Put these problems in this context.

Linux is the best thing out there. Mozilla and OpenOffice rock. I love open source (free and otherwise) software and support it whenever I can. However there is a market for state of the art hardware (Nvidia) and software (Intel compiler, Oracle database, high-end applications, etc.). We live in a mixed environment.

Do you want to be paid as a programmer? Do you want to have some worth to your products? There is a strong market for commercial, closed software (specialized software, industrial databases, custom solutions, high-end games). Not all can be free and open, nor should it be. It is far harder to make money on just services. Do you want programmer jobs to go to India like the mass of consumer hardware now made in the far east? Are the US and Europe becoming consumers and service organizations with few products of our own?

I can't resist mentioning Microsoft in this context. Much of what they do is now a commodity (operating system: use Linux, word processing/presentation/spreadsheet: use open office, servers: use Linux/BSD with Samba, etc.). They are the competition in the desktop, server, and embedded spaces. They are getting scared (think trapped beast). How can we compete with Microsoft with their nearly 100% (until recently) closed products? By working with vendors that can't or won't open their products. By getting commodity and older product drivers released (for example Canon printers - hint, hint). By working with hardware/software vendors on state of the art drivers but letting them keep their core IP if it helps them with a competitive edge (and gets us drivers).

Mmmm..

ZDNet UK is reporting that Intel has promised to increase Linux support by releasing Linux drivers at the same time it releases Windows drivers for its hardware.

Pingla writes in with more good news: "Intel promises to release Linux drivers for its Centrino chipset at the same time it releases drivers for Windows.

OK, that first piece of news is nice, but the second one really gets my heart racing.

well

I never bought the centrino hype in their commercials "internet anywhere" of course there was a * clause in their ads.

I have always thought they will follow the profit. They do not have much to lose by supporting centrino for Linux, only everything to gain. More sales because they know (at least some) people buy laptops to use Linux on them. And by adding WIFI support to linux they are hoping to sell their centrino technology outside of the windows market.

After all, indicators show that Linux is growing and can only gain more users in the years to come.

Cameras

Great. So where are the frickin' drivers for all the Intel USB cameras?

Releasing at the same time...

"ZDNet UK is reporting that Intel has promised to increase Linux support by releasing Linux drivers at the same time it releases Windows drivers for its hardware."

So, does this mean that Intel will stop releasing Windows drivers? :)

Having open-source drivers helps the hardware...

When I bought this new computer in Nov 2003, my options were basically (1) all intel + integrated graphics or nvidia (2) all AMD + nvidia graphics card --- since older ATI cards with full open-source drivers are hard to obtain here, I will not consider them. I chose intel+i845G because it is well supported under linux without all those closed-source driver hassles, although it is quite a bit more expensive than an AMD solution, and the 3D performance advantage of a low-end Geforce4 versus i845G (whose performance is about the state of the art five years ago according to my experience) would be somewhat useful to me. Now, seeing all those people having trouble with nvidia drivers (even though they are probably the best closed-source drivers around), especially those tinkering with new kernels (I am one), I think I have made the right decision.

Therefore, I think the availability of open-source drivers should help the hardware sales quite a bit, in that people like me are willing to accept somewhat worse price-to-performance ratio for a open-source (therefore well-supported) driver. Considering that more and more people are trying to install linux on their desktop, and most distributions are unlikely to include proprietary drivers anytime soon, closed-source drivers will be a significant minus for people planning to install linux on the system.

Don't underestimate the value of having the drivers open-sourced, Intel...

I blame Linus Torvalds.

Seriously. This is not a troll, so hear me out here. I love Linux and I won't use anything else, including on my desktop.

The real problem here is Linus's stubborn refusal to freeze the driver API's. At the very least, the driver API's should be frozen during each major release cycle; i.e. a driver which loads on 2.6.0 should continue to work properly on 2.6.999. If there are big new exciting things that force an API change, it should wait until 2.8.0.

I say that this is Linus's fault because it's well-documented that the moving-target API's are his clear decision. And it's a bad decision. If he wants large-scale adoption of Linux at the end-user level, he's going to have to realize that most end-users aren't smart enough to do their own driver integration -- but they might be able to download a driver off the 'net or from a CD, and see "Gruntle FOOset driver for Linux 2.6" and expect that it'll work on any Linux distribution that includes a 2.6 kernel.

Until the driver API is stabilized, Linux is going to have a hard time finding users outside the hacker set.

Re:I blame Linus Torvalds.

I say that this is Linus's fault because it's well-documented that the moving-target API's are his clear decision.

He says so repeatedly in his posts, so it's not like it is a secret.

And it's a bad decision.

No you are wrong here. As a practical matter binary drivers lead to buggy unstable kernels. The people writing these drivers have no contact with or support from experienced kernel developers due to the closed nature of the process, and code quality suffers. And people posting about binary drivers waste everyone's time, including their own.

Until the driver API is stabilized, Linux is going to have a hard time finding users outside the hacker set.

Linux has a lot of users outside the "hacker set". Did you miss the part about Linux overtaking MacOS and it's current share of the server market?

Palladium

Does this mean that we're more likely to get Palladium aka Trusted Computing to work with Linux? If Intel is interested in making sure that their boards work with Linux, this seems like a good start to keep Microsoft from tying up the hardware...

Didn't they say this last year?

Last march I bought a Centrino notebook, only because chipset support seemed to be there, powerstep looked about to be implemented AND wireless LAN drivers were promised to be released very soon by Intel. Now, almost a year later, still no wireless driver and they still say "real soon now". I guess I am too gullible.
Oh, I also believed them that their crap keeps cool. Even at 600MHz (instead of 1300) and doing nothing this thing gets freaking hot and makes lots of noise.
I am MUCH happier with my Crusoe (Toshiba Libretto) notebook. I guess my next one will be an Efficeon.

Legality of these binary drivers?

Ok, what I would like to know is: are these binary drivers even legal?

I did some reading on the linux kernel mailing list and the general concensus between the developers seems to be that binary-only drivers as modules for the linux kernel are not legal.

The only case they sited as a legal binary only module so far was the nvidia video card driver because the driver was not written for linux, it was written for windows and merely repackaged into linux.

The concensus seemed to be that a driver written specifically *for* linux is a derrivative work and therefore must be GPL'd.

Fedora Core 1, Suse 9.0 etc etc

This will be a binary only release, pretty much hands down, pretty much precluding the more esoteric and non US centric distros getting a driverset. Still the big deal for me isn't distro, OS lockin because of drivers is no news to me.

I sit here typing this on my Presario X1000 which would not agree to function with the DriverLoader hack. The only way I'll be able to get reliable support for mini-PCI wifi will be to replace the intel card with something like this. [agere.com]

Hell I'm not even worried about the wifi drivers until I can actually get decent battery life. Maybe if the speedstepping was 100% complete and verfied by an intel OSS coder then I'd take this to heart. Until then, this is just more of the same empty promises [intel.com] Linux drivers are "under development" and have been for nearly a year for the wifi, from intel's page anyways.

I will hold fast to my no-intel policy...

Until after i actually see the crap they promise. I'll stick with AMD and superior add-on/pcmcia cards that have native linux support.

Intel is pschizo. They "support" linux, they don't support linux. They say one thing, do another. They are, in a sense, merely Bill Gates' and M$'s Poodle.

Boycott Intel until they pull their multiple personality head out of their anal sphincter and actually go OS neutral the way a CPU maker SHOULD be.

One idea

Proprietary drivers are not optimal, but they may become a fact of life for newer hardware. Too much of the functionality of many devices (and therefore their advantage in the marketplace) is now in the drivers.

One novel approach would be for the company, in this case Intel, to produce a binary driver and place the source code in some form of trust, to be released when they no longer support the driver or the company no longer feels that the source code would provide an advantage to other companies.

FCC regs

It is likely that Intel cannot release an OSS driver, if the driver itself controls a large portion of the radio hardware. This is probably the case, in situations I've used Centrinos -- the CPU useage is notibly higher when using the WiFi hardware then when not.

Software access to the radio control portion of the system would mean users could adjust the frequency and power output of the system -- something which would run them afoul of FCC regulations requiring that equipment of this nature be fixed and not changeable by the end user. And, the FCC would not take kindly to this. Both Intel, and the modifying user, could be liable.

Strange New World

I have many contacts in commercial hardware and software companies that produce proprietary products. I advocate open source and it's benefits as often as I get the chance. For the most part, the message is heard and appreciated.

In fact, in a recent job interview with a software company, the following question was put to me: "What process or organization should a product company use to build product on open source platforms?" I thought this was an excellent question and I believe my answer was pertinent to this thread.

I actually don't have a problem with what Intel is doing and here's why. Not everything on an open source platform has to be open source. I know many of us are purists and would rather not hear this, but I believe it's true. One niche that proprietary products fill very well is as niche-filler. :) In other words, there are many applications and products out there that just will not work as an open source project. Some products lack the broad consumer base for it to mature fast enough in an open source mode for it to be useful to anybody. Does that mean that those kinds of products should not benefit from an open source platform?

Getting back to my answer to the interview question. I told the interviewer that a commercial software company that wants to build on an open source platform should organize by having a community development group which adds to and supports the open source product that they are using and by having a proprietary development group that uses the open source product to do their work without changing or contributing to the open source product. I added that there should be a wall between them lest you create a mini-SCO fiasco. I mentioned several examples of this that appear to be working. I think a company like CodeWeavers is a perfect example and I think there are many big companies that are starting to figure this out (i.e. how to work with the GPL).

Get ready folks. The Man is coming. :)

tims

2007, Linux = Windows

Sadly, the Centrino support will most likely be a proprietary driver, but it's better than nothing.

No, it's not.

Ever heard of a precedent?

long-time FreeBSD support for internal wireless

When I bought my Thinkpad T40 [cr.yp.to] many months ago, I chose a Cisco internal (mini-PCI) wireless card instead of the Intel internal wireless card. FreeBSD already included a driver for the Cisco card. The driver worked perfectly.

Sure, this substitution means that the T40 is not officially a Centrino, but who cares? The OS uses the two antennas built into the laptop; it doesn't need a PCMCIA wireless card; it simply works.

How about a better i865 video driver?

The current XFree86 driver works, but relies on VESA video modes. The driver released two years ago or so by Intel is useless, as it just causes X to crash immediately.

I really hate it when vendors make one release of the driver then EOL the product. (Kinda like the Radeon M7 in my Dell Inspiron 4150!)

DMCA is the reason binary drivers are unacceptable

Forget for a moment the fact that free software drivers are mostly better made than their proprietary counterparts. Disregard the fact that anyone can fix bugs and make improvements to free drivers without restriction. Let's ignore the fact that you can't port binary-only drivers over to other free OSes. Let's even overlook the problem of getting binary drivers to work on different versions of the Linux kernel. One thing you cannot ignore are the implications of Digital Restrictions Management and laws banning reverse-engineering. If we do not set a precedent for all Linux drivers being released as free software, it may be impossible to run Linux with certain hardware devices without breaking the law (of course if the hardware has mandated hard-wired DRM features, it may be impossible to run Linux period!). The fight against anti-reverse engineering laws has already been lost, but at least the preceding scenario can still be avoided.

I don't believe the current response from the Slashdot community is in any way hypocritical, considering that most of them only want these drivers for the same reason they're using Linux - it doesn't cost anything. Unfortunately, this is just greed, and greed tends to make one's views short-sighted. The only way to ensure the growth of Linux development is to set a precedent for Free Software drivers - it's not much fun having an OS without hardware to run it on!

Format?

Will these be easy to deal with drivers like NVidia or ittatation ones like most winmodem drivers. If it's the first, that would be cool. If it's buggy, irritating frivers that only worh with red hat, then it's useless.

It's already working...

Thanks to Pontus Fuchs, Giridhar Pemmasani, Joseph Dunn and others from the ndiswrapper [sourceforge.net] project, I'm actullay posting this from my Thinkpad [ibm.com] using the Centrino Intel wireless network card!

Since I'm running Debian GNU/Linux [debian.org] stable (yes, that's right, I'm on woody), I had to install a newer version of iwconfig and modify my /etc/network/interfaces file to make it work well:

iface wlan0 inet dhcp
pre-up modprobe wlan0 || true
pre-up /usr/local/sbin/iwconfig wlan0 mode managed
pre-up /usr/local/sbin/iwconfig wlan0 enc 1234-789A-EF
pre-up /usr/local/sbin/iwconfig wlan0 essid WIRELESS

Of course, since ndiswrapper use the Windows XP drivers file, it does not resolve the problems about proprietary drivers. But at least, I was not stuck to wait (an eternity) for Intel to release their Linux drivers.

This space is reserved.

Help by supporting OLD hardware

Intel could really help by releasing information about old equipment so that the community can develop drivers for equipment.

For example, I have an Intel 2200 ADSL PCI card that doesn't have any drivers for Linux or BSD. This card was given out by the Qwest ILEC in the last few years and there are a lot of them out there. I can't do anything with mine though. I hang onto it hoping that someone out there will be more resourceful than I am.

http://www.intel.com/network/broadband/modems/ds l_ 2200.htm

linux support == something good

more linux support is always good.

Don't people see what is happening here?

Intel is not doing this out of kindness, nor out of love for Linux. No, Intel is doing this to screw Microsoft by supporting Linux because Microsoft is screwing Intel by supporting AMD64 with their new Windows XP 64-bit version. The former partners in crime are turning on each other and Linux is the beneficiary...at least for the moment. Microsoft and Intel could patch up their differences at any time and then that Intel support for Linux would disappear faster than a snowflake in July.

Will Intel really support Linux? Or just Centrino?

Intel is throwing money at Centrino because it is an inferior wireless solution.

Let's see if Intel really supports Linux:

The main catch in getting the software is the driver for your mass storage, the hard drive. Unless you need to support only a small number of workstations--say, 5 or less--a SCSI or SATA interface is essential for good performance. Unlike the parallel IDE, neither the SCSI nor the SATA is a standard interface, and a special driver for Linux is needed. Moreover, the driver needed is for the SCSI or SATA interface and, in general, should be supplied by the motherboard maker or the interface maker. The SCSI driver I got from Intel was written for Red Hat 8.0, and as a result, I can choose only the LTSP software that was built on Red Hat 8.0. In fact, the driver from Intel didn't work with Red Hat 9.0, so I had to settle for LTSP 3.0, which was built on Red Hat 8.0. [linuxjournal.com]

When situations like this change, then Intel can boast about Linux support. Until then, its hot air.

And until then, I'll continue buying AMD, the best bang for my buck, and the new performance leader on 64 bit, which I can use on Linux NOW.

Also kind of off-topic...

But Lindash may not be in the clear - after all, MS calls its X-Box loader the MS Dashboard..

Then what about Linux?

Then I guess Linux been Free Software does not matter either? I am serious, what's the adventage if Linux is going to have the same issues than other OSes: have to get drivers every now and then from different places instead of just upgrade kernel or distro, "play but do not touch", "upgrade your hardware becuase we do not support it anymore", "we do not like your architecture, just x86-32", etc.

Re:**SIGH**

Who the hell cares besides RMS? I love using my machine and it has an nVidia card in it. I don't care that their "driver" is closed source, I can play a lot of heavy duty games with it.

It's not about RMS. Open source drivers benefit the development of the kernel, and also the users of the drivers and hardware those drivers support. Remember when the linux kernel was at 2.6, but we had to wait some time before nvidia released 2.6 compatible drivers? If they were GPL, the kernel developers could have incorporated the drivers into the kernel and development would have gone concurrently.

Even now, sticking a closed source driver in there is problematic if there's a kernel panic. How are you going to debug it? What about security? Nobody ourside of nvidia has audited the code. There could be a potential vulnerability that they missed. We negate the benefits of open source if only *part* of our program is open source.

Re:**SIGH**

Who the hell cares besides RMS?

That's quite a flamebait-inducing post you've got there...

What about other operating systems? Do we have to badger Intel to release drivers for BSD, and whatever other operating systems might be released in the future?

What happens if we release a new kernel, or decide to change something that breaks the rigid structure into which this proprietary driver is locked?

Releasing proprietary drivers like this seems to be no more than a "keep them happy" quick-fire solution, as this is by no means a long-term solution. And frankly, ignoring the long-term is a very short-sighted viewpoint indeed.

What's the ideal solution? Write your drivers so that they use a well-documented and open API that can always be well-supported, and make the code as portable as possible. Then what happens when you want to use your hardware with a different operating system? Well, so long as your operating system implements that particular well-documented and open driver API, then you shouldn't have any problem. Recompile, rinse, repeat.

Think ahead. We wouldn't be pushing for open source drivers without reason.

Re:Yeah right...

It's where the Windows source code is hiding...

Re:**SIGH**

> Who the hell cares besides RMS?

What do you do when the manufacturer goes belly up, or just descides to terminate the support for your product line?
Just throw it out and buy a new piece of hardware?

How about Linus?

Seriously, Linus doesn't like propretary drivers and will happily make source code compatible but binary incompatible changes. Linus does not want to have to deal with proprietary binary compatibility crap for drivers. It just clogs up the kernel with a lot of dead wood and dead weight. For open source, Linus can integrated the code and ensure that any changes will be automatically propagated to these drivers. With proprietary drivers, his hands are tied.

Re:**SIGH**

I do not want to execute code I can not audit.

Re:**SIGH**

I care. Simple: when the hardware is no longer supported my the manufacturer, what do you do if it becomes incompatible with new kernel versions? Or let's say that for some reason (I can think of one), you don't want to use XFree86. Too sad, your driver doesn't work with Y or freedesktop.org X servers... you have to stay with whatever the manufacturer chooses.

Re:**SIGH**

BZFlag? Unreal Tournament (including GOTY, 2003 AND 2004)? Armagetron?

Re:**SIGH**

I love using my machine and it has an nVidia card in it. I don't care that their "driver" is closed source, I can play a lot of heavy duty games with it.

Heavy duty games? On Linux? I knew this was a troll... why didn't you throw in a "and who cares about BSD drivers, it's dying anyway?" line?

Re:**SIGH**

Who the hell cares besides RMS?

Me. I have a ppc. (of course I'm not concerned about Centrino stuff this time, but I was when I couldn't buy a powerbook due to lack of 3D, when said powerbooks shipped with nvidia chips).

Re:**SIGH**

1. 1. Sadly, the Centrino support will most likely be a proprietary driver, but it's better than nothing.

   Who the hell cares besides RMS? I love using my machine and it has an nVidia card in it. I don't care that their "driver" is closed source, I can play a lot of heavy duty games with it.

[Raises hand] While I am not dogmatic about it, there are a few serious practical concerns about closed source drivers;

Can't use them out of the box; it's another set of steps.

1. Except for propriatory drivers, most hardware is well supported under any distribution you have. There
2. is no seperate installation step or set of directions you need to follow for the open source parts!

The closed drivers tend to be flaky.

1. Nvidia has done a great job with this, though it has taken well over a year to reach a point where they are stable. Few other drivers I use -- silently and without hassles -- have stability problems at all.

There are few reasons *not* to go with open source...and quite a few reasons not to.

1. Hiding shoddy code or protecting 3rd party licenced parts are the only reasons not to release the source.

2. Allowing the code to be reviewed and fixed, having nearly automatic support for non-x86 CPUs, having a much wider user base, and simple good will are reasons to release the source.

   In some cases -- and Intel and Nvidia specifically can do this -- a mix of 'firmware' style add-ins limited narrowly to a few 3rd party propriatory parts would probably work. Hiding the source to protect it from prying eyes isn't a good reason since everyone has debuggers and disassemblers...so if they want to know they probably already do know how the secret sauce is made and what it does.

Re:Closed source drivers can be very helpful

The problem with closed source drivers are
1. Time lag and
2. Platform support

How about putting proprietary interfaces in firmware instead. That way it can be updated, and open source drivers don't tell you anything valuable about the hardware IP. I don't mind proprietary firmware but frankly, its the reason I bought a radeon 9000 instead of an nvidia card for my linux box.

Re:Video Card/GPU/CPU with GPL'd drivers?

Intel seems quite timid about making any fast moves about this... Speaking of precedents (from another post): if hardware drivers are ever to become largely open source, someone is going to have to take the first step and release truly OSS drivers for their hardware. Has it been done yet? By ANYONE? The parent could have asked alternatively: would it be possible to put together a modern desktop PC with internet and gaming capabilities using only hardware with fully OSS drivers? I ask because I'd actually like to do it...