Why do some in the Linux community treat ndiswrapper with such contempt? It seems to me that if something works, especially as well as ndiswrapper, what damned difference does it make if the device drivers were written for use in Windows? I rather think that the concept should be expanded to include the universe of devices that are not currently useable with Linux due to unavailable drivers. Fred
Why do some in the Linux community treat ndiswrapper with such contempt? It seems to me that if something works, especially as well as ndiswrapper, what damned difference does it make if the device drivers were written for use in Windows? I rather think that the concept should be expanded to include the universe of devices that are not currently useable with Linux due to unavailable drivers. hi Fred, The problem with ndiswrapper (or any other binary driver) is that it breaks the linux kernel philosophy, and also poses a threat to the stability of linux in general. One reason Windoze is *broken* is that way too many vendors (binary drivers) are trusted to run at ring zero (root level). This is not going to happen with Linux... and thankfully so far, the linux kernel development
On Tuesday 19 September 2006 20:45, Stevens wrote: people are sticking to their guns on this. If you want to *taint* your kernel and take the chance that that binary driver you just loaded into your kernel is safe, robust, and can be trusted... go for it. However, the open source gpl concept says that the kernel is going to be open, stable, and untainted (unless the user decides to shoot his/her own foot) Another problem with ndiswrapper particularly, is that the wireless standards are badly exposed with security holes... and most of the windoze drivers for wifi are painfully vulnerable... and that makes linux vulnerable too.... for ndiswrapper users. And lastly, software should be open.... period. The days of proprietary hardware drivers must end.... now. I am getting to the point that if a piece of ah--- hardware, doesn't have an open linux driver then I don't need their hardware. Period> -- Kind regards, M Harris <><
On 19/09/06 20:06, M Harris wrote:
On Tuesday 19 September 2006 20:45, Stevens wrote:
Why do some in the Linux community treat ndiswrapper with such contempt? It seems to me that if something works, especially as well as ndiswrapper, what damned difference does it make if the device drivers were written for use in Windows? I rather think that the concept should be expanded to include the universe of devices that are not currently useable with Linux due to unavailable drivers.
hi Fred, The problem with ndiswrapper (or any other binary driver) is that it breaks the linux kernel philosophy, and also poses a threat to the stability of linux in general. One reason Windoze is *broken* is that way too many vendors (binary drivers) are trusted to run at ring zero (root level). This is not going to happen with Linux... and thankfully so far, the linux kernel development people are sticking to their guns on this. If you want to *taint* your kernel and take the chance that that binary driver you just loaded into your kernel is safe, robust, and can be trusted... go for it. However, the open source gpl concept says that the kernel is going to be open, stable, and untainted (unless the user decides to shoot his/her own foot)
With my previous sound card, the module (native Linux) distributed by SuSE tainted my kernel. Recently, I got a couple of these: Sep 5 20:34:55 static24-89-67-198 kernel: thinkpad: module not supported by Novell, setting U taint flag. Why the system thought this was a thinkpad is beyond me, it isn't even a laptop... so much for shooting *myself* in the foot.
Recently, I got a couple of these:
Sep 5 20:34:55 static24-89-67-198 kernel: thinkpad: module not supported by Novell, setting U taint flag.
Why the system thought this was a thinkpad is beyond me, it isn't even a laptop... so much for shooting *myself* in the foot.
Nono, that is different. No taint flag = phone someone Your options: Novell, LKML or ModuleAuthor U taint flag = don't try phone Novell Your options: LKML or ModuleAuthor T taint flag = don't try phone anyone in the opensource community for it Your options: ModuleAuthor Jan Engelhardt --
On 20/09/06 03:42, Jan Engelhardt wrote:
Recently, I got a couple of these:
Sep 5 20:34:55 static24-89-67-198 kernel: thinkpad: module not supported by Novell, setting U taint flag.
Why the system thought this was a thinkpad is beyond me, it isn't even a laptop... so much for shooting *myself* in the foot.
Nono, that is different.
No taint flag = phone someone Your options: Novell, LKML or ModuleAuthor
U taint flag = don't try phone Novell Your options: LKML or ModuleAuthor
T taint flag = don't try phone anyone in the opensource community for it Your options: ModuleAuthor Bottom line, this all means what to the average user? To the average user, the difference between a tainted kernel and ndiswrapper is non-existent: It will all come down to him being unable to find support from the people who wrote the product he is using, and possibly being unable to find support from anyone. The average user will simply
Translation: "If you have a thinkpad, and are foolish enough to install SuSELinux on it, don't call us." Can you think of any reason why Novell doesn't support a SBAwe32 (which is the card I referred to before)? The source (included in the kernel source package) certainly *seems* to be open-source. translate either situation into "Novell doesn't even support it's own products" and switch back to <the OS from Redmond>, or some other Linux distro. The switch back to <the OS from Redmond> will come even faster, if refusal to support comes from the OSS community generally. Indeed, if a few of the recent threads on this list are any indication, such a user would be subjected to a great deal of derision, with the occasional suggestion that he should just go buy another (whatever device he bought) from a more Linux-friendly manufacturer, or simply go back to Windows. BTW, does all this disdain for non-GPL device drivers mean that anyone who ever installs a Blu-Ray or HD-DVD into his Linux machine (any distro) will no longer receive support for anything from the OSS community? Because I can assure you, if you will be waiting for open source drivers for either of these, you will be waiting longer than this lifetime. Of course, if you are an absolute purist, like Mr Harris, then you will just boycott those products, and refuse to use them at all -- even past the day when one of them becomes the de facto standard, and the present-day DVD has gone the way of the dinosaur.
Can you think of any reason why Novell doesn't support a SBAwe32 (which is the card I referred to before)? The source (included in the kernel source package) certainly *seems* to be open-source. Perhaps its because the card is ISA circa 1994 [ talk about wanting to keep your dinosaur around... ] :) But seriously, I have an ISA SBawe32 also... one of my favorites--- which is why I keep my old HW around... you know, pretty soon we aren't going to be able to find a machine to plug em into!
Of course, if you are an absolute purist, like Mr Harris, then you will just boycott those products, and refuse to use them at all -- even past the day when one of them becomes the de facto standard, and the present-day DVD has gone the way of the dinosaur. Oh, please. Trust us here... if the device is going to be a de facto standard there will certainly be open drivers for it. On the other hand, many of us are just plain through with being strong-armed (really held hostage) into running Redmond's junk OS just because "this silly card only runs on Windoze, with a proprietary Windoze driver, and an appropriate kick-back to Bill's favorite charity--- himself and his Windoze company". We're done... I
On Wednesday 20 September 2006 16:45, Darryl Gregorash wrote: think the final and last straw for me was the WinModem... and the WinPrinter... Bill can have his WinJunk... cause some of us aren't going to use it any more... period. And it isn't because we're purists who have climbed an ivory tower... its because Windoze is an evil affront to freedom of choice, computer science, innovation, and the American way! -- Kind regards, M Harris <><
On 20/09/06 17:15, M Harris wrote:
On Wednesday 20 September 2006 16:45, Darryl Gregorash wrote:
Can you think of any reason why Novell doesn't support a SBAwe32 (which is the card I referred to before)? The source (included in the kernel source package) certainly *seems* to be open-source.
Perhaps its because the card is ISA circa 1994 [ talk about wanting to keep your dinosaur around... ] :) But seriously, I have an ISA SBawe32 also... one of my favorites--- which is why I keep my old HW around... you know, pretty soon we aren't going to be able to find a machine to plug em into!
I thought about that, and while you may be right, that always seemed to me to be a rather underhanded way to suggest that I should cause that thing to disappear :-) Actually, the real reason I am no longer using it is because the 2.6 kernel doesn't seem to like the Via Apollo chipset/K6-2 combination I had it in (read: frequent system lockups). The current motherboard has no ISA slots, or I'd probably still have the SBAwe installed.
Of course, if you are an absolute purist, like Mr Harris, then you will just boycott those products, and refuse to use them at all -- even past the day when one of them becomes the de facto standard, and the present-day DVD has gone the way of the dinosaur.
Oh, please. Trust us here... if the device is going to be a de facto standard there will certainly be open drivers for it. On the other hand, many Don't hold your breath :-) (and a Blu-Ray equivalent of libdvdcss doesn't count).
of us are just plain through with being strong-armed (really held hostage) into running Redmond's junk OS just because "this silly card only runs on Windoze, with a proprietary Windoze driver.... This reminds me, the ATI and nVidia stuff is all proprietary.. you gonna start running SiS chipset video cards? :-)
(duck'n'run)
On 21/09/06, Darryl Gregorash
On 20/09/06 17:15, M Harris wrote:
On Wednesday 20 September 2006 16:45, Darryl Gregorash wrote:
Can you think of any reason why Novell doesn't support a SBAwe32 (which is the card I referred to before)? The source (included in the kernel source package) certainly *seems* to be open-source.
Perhaps its because the card is ISA circa 1994 [ talk about wanting to keep your dinosaur around... ] :) But seriously, I have an ISA SBawe32 also... one of my favorites--- which is why I keep my old HW around... you know, pretty soon we aren't going to be able to find a machine to plug em into!
I thought about that, and while you may be right, that always seemed to me to be a rather underhanded way to suggest that I should cause that thing to disappear :-) Actually, the real reason I am no longer using it is because the 2.6 kernel doesn't seem to like the Via Apollo chipset/K6-2 combination I had it in (read: frequent system lockups). The current motherboard has no ISA slots, or I'd probably still have the SBAwe installed.
Of course, if you are an absolute purist, like Mr Harris, then you will just boycott those products, and refuse to use them at all -- even past the day when one of them becomes the de facto standard, and the present-day DVD has gone the way of the dinosaur.
Oh, please. Trust us here... if the device is going to be a de facto standard there will certainly be open drivers for it. On the other hand, many Don't hold your breath :-) (and a Blu-Ray equivalent of libdvdcss doesn't count).
of us are just plain through with being strong-armed (really held hostage) into running Redmond's junk OS just because "this silly card only runs on Windoze, with a proprietary Windoze driver.... This reminds me, the ATI and nVidia stuff is all proprietary.. you gonna start running SiS chipset video cards? :-)
(duck'n'run)
--
My ATI hardware runs fine on open-source drivers. I don't run games on my PCs, but if I wanted to run games, I would get a games machine. Jeff Rollin
On Wednesday 20 September 2006 20:43, Darryl Gregorash wrote:
This reminds me, the ATI and nVidia stuff is all proprietary.. you gonna start running SiS chipset video cards? :-)
(duck'n'run) I'm not a serious gamer... In fact I don't have time to play games with my systems... I have early 3D accelerated cards that work fine out of the box... although not nearly as fast as the nVidia GeForce stuff... like I would know... and my kids are using those without too much complaint [ it beats not having any computer at all--- which could happen if they whine... ;-) ] So for me its not an issue... but I hear you [ and it annoys me ] that the gamer community doesn't demand an openGL DirectX 3D accelerated monster from somewhere that has open drivers! <ahhh sigh.>
-- Kind regards, M Harris <><
On Wednesday 20 September 2006 20:43, Darryl Gregorash wrote:
Actually, the real reason I am no longer using it is because the 2.6 kernel doesn't seem to like the Via Apollo chipset/K6-2 combination I had it in (read: frequent system lockups). I will read it. And, I'll let you know what happens when I reinstall mine... my old e-machine box uses Via and the AMD-K63... we'll see what happens.... that machine is going to be dedicated to off-line midi-music attached to the Korg. I had the awe installed in an old 486 DX40 (believe it or not) running RH5.2 kernel 2.0.36 (with a Y2K bios wedge installed) and it ran until early this year when it finally died.... it was my oldest machine IBM Ps1 circa 1991. True story. Ps. I took the MB to the framing shop downtown and had it mounted on red velvet...
-- Kind regards, M Harris <><
On 20/09/06, Darryl Gregorash
On 20/09/06 03:42, Jan Engelhardt wrote:
Recently, I got a couple of these:
Sep 5 20:34:55 static24-89-67-198 kernel: thinkpad: module not supported by Novell, setting U taint flag.
Why the system thought this was a thinkpad is beyond me, it isn't even a laptop... so much for shooting *myself* in the foot.
Nono, that is different.
No taint flag = phone someone Your options: Novell, LKML or ModuleAuthor
U taint flag = don't try phone Novell Your options: LKML or ModuleAuthor
Translation: "If you have a thinkpad, and are foolish enough to install SuSELinux on it, don't call us."
T taint flag = don't try phone anyone in the opensource community for it Your options: ModuleAuthor Bottom line, this all means what to the average user? To the average user, the difference between a tainted kernel and ndiswrapper is non-existent: It will all come down to him being unable to find support from the people who wrote the product he is using, and possibly being unable to find support from anyone. The average user will simply
Can you think of any reason why Novell doesn't support a SBAwe32 (which is the card I referred to before)? The source (included in the kernel source package) certainly *seems* to be open-source. translate either situation into "Novell doesn't even support it's own products" and switch back to <the OS from Redmond>, or some other Linux distro. The switch back to <the OS from Redmond> will come even faster, if refusal to support comes from the OSS community generally. Indeed, if a few of the recent threads on this list are any indication, such a user would be subjected to a great deal of derision, with the occasional suggestion that he should just go buy another (whatever device he bought) from a more Linux-friendly manufacturer, or simply go back to Windows.
BTW, does all this disdain for non-GPL device drivers mean that anyone who ever installs a Blu-Ray or HD-DVD into his Linux machine (any distro) will no longer receive support for anything from the OSS community? Because I can assure you, if you will be waiting for open source drivers for either of these, you will be waiting longer than this lifetime.
Of course, if you are an absolute purist, like Mr Harris, then you will just boycott those products, and refuse to use them at all -- even past the day when one of them becomes the de facto standard, and the present-day DVD has gone the way of the dinosaur.
-- Check the headers for your unsubscription address For additional commands send e-mail to suse-linux-e-help@suse.com Also check the archives at http://lists.suse.com Please read the FAQs: suse-linux-e-faq@suse.com
I can see no reason for complaining that people who do not comply with the Linux licensing requirements are not given quarter; if a company disclosed confidential Windows specs or code, in breach of a Windows licence, in order to write a driver under the GPL (or any other), the response from Redmond would be a lot heavier and nastier than a snide remark or 20 on a mailing list. Jeff Rollin
On 20/09/06 17:58, Jeff Rollin wrote:
<snip> I can see no reason for complaining that people who do not comply with the Linux licensing requirements are not given quarter; if a company disclosed confidential Windows specs or code, in breach of a Windows licence, in order to write a driver under the GPL (or any other), the response from Redmond would be a lot heavier and nastier than a snide remark or 20 on a mailing list. Your point being what? That the disclosure of confidential and/or copyright and /or patented hardware specifications or driver code is somehow in the same league as the lack of open-source code for the same device?
Get real.
On 21/09/06, Darryl Gregorash
On 20/09/06 17:58, Jeff Rollin wrote:
<snip> I can see no reason for complaining that people who do not comply with the Linux licensing requirements are not given quarter; if a company disclosed confidential Windows specs or code, in breach of a Windows licence, in order to write a driver under the GPL (or any other), the response from Redmond would be a lot heavier and nastier than a snide remark or 20 on a mailing list. Your point being what? That the disclosure of confidential and/or copyright and /or patented hardware specifications or driver code is somehow in the same league as the lack of open-source code for the same device?
My point being, that you seem to have no problems with Windows licences limiting you to closed-source drivers, therefore I can see no reason to object to Linux licencing which limits you to the reverse. Since it's precisely the "lack of hardware specifications or driver code" which prevents people writing open-source drivers for operating systems whose licences require them, I'd suggest that it's you who needs to: Get real.
I might add, that despite your complaints about how people who want to run
ndiswrapper/closed source drivers are treated, so far in this thread the only person I've seen being rude or abusive is you. Jeff.
On Tuesday 19 September 2006 18:06, M Harris wrote:
On Tuesday 19 September 2006 20:45, Stevens wrote:
Why do some in the Linux community treat ndiswrapper with such contempt? It seems to me that if something works, especially as well as ndiswrapper, what damned difference does it make if the device drivers were written for use in Windows? I rather think that the concept should be expanded to include the universe of devices that are not currently useable with Linux due to unavailable drivers.
hi Fred, The problem with ndiswrapper (or any other binary driver) is that it breaks the linux kernel philosophy, and also poses a threat to the stability of linux in general. One reason Windoze is *broken* is that way too many vendors (binary drivers) are trusted to run at ring zero (root level). This is not going to happen with Linux... and thankfully so far, the linux kernel development people are sticking to their guns on this. If you want to *taint* your kernel and take the chance that that binary driver you just loaded into your kernel is safe, robust, and can be trusted... go for it. However, the open source gpl concept says that the kernel is going to be open, stable, and untainted (unless the user decides to shoot his/her own foot)
Er, wouldn't it be better to explain this with out tossing FUD around like some microsoft salesman? Shoot Yourself in the Foot? By loading a wireless driver? For god sakes give it a rest man! You sound like a tainted kernel is a hanging offense. 80% of linux users are running tainted kernels. (I pulled that statistic out of my ass, - go ahead and prove me wrong). Taint simply means one is doing more with the machine than the original kernel developers were able to supply. It simply means you can take a module designed for another platform and run it in a sandbox and make it work in Linux. The solution is brilliant. If ndiswrapper is the only way you get you network card working you either open you wallet and fork over $35 for a card that works or climb down off your ivory tower and load ndiswrapper and go about your business.
Another problem with ndiswrapper particularly, is that the wireless standards are badly exposed with security holes... and most of the windoze drivers for wifi are painfully vulnerable... and that makes linux vulnerable too.... for ndiswrapper users.
There is ABSOLUTELY no evidence that open source drivers prevent the "wireless standards" from being "badly exposed with security holes". (whatever the hell that means). Open source drivers have to ad-hear to the same standards. If WEP is broken and weak, it will still be broken and weak with open source drivers.
And lastly, software should be open.... period. The days of proprietary hardware drivers must end.... now. I am getting to the point that if a piece of ah--- hardware, doesn't have an open linux driver then I don't need their hardware. Period>
Well how cute. Let me summarize: You don't like Windows. Ndiswrapper allows you to use windows drivers in Linux. Therefore, Ndiswrapper is evil. Time to grow up Mr Harris. Buy what ever card you want but don't trash a very clever solution to migrating a product from one platform to another just because of your political beliefs. -- _____________________________________ John Andersen
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 The Tuesday 2006-09-19 at 23:35 -0800, John Andersen wrote:
Taint simply means one is doing more with the machine than the original kernel developers were able to supply. It simply means you can take a module designed for another platform and run it in a sandbox and make it work in Linux. The solution is brilliant.
Not really. Tainted means that you loaded some "closed" software, which in turn means that if some thing fails there is nothing the kernel developers will do about it, because they can not check the code. As simple as that. It is like trying to repair a car with a black box inside. The scheme of peer review that has given Linux software the stability and security it has is thrown out of the window if you load closed source software. Not because it is evil - nobody said that but you - but because it can not be checked. - -- Cheers, Carlos E. R. -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.2 (GNU/Linux) Comment: Made with pgp4pine 1.76 iD8DBQFFERUMtTMYHG2NR9URAq9tAJ4ip/4XLl8Uush0kTIM3jeKp9ObqgCgj1LO WlCIDoZsAkwKzlvHgjhv7GA= =m/F8 -----END PGP SIGNATURE-----
On Wednesday 20 September 2006 12:16, Carlos E. R. wrote:
The Tuesday 2006-09-19 at 23:35 -0800, John Andersen wrote:
Taint simply means one is doing more with the machine than the original kernel developers were able to supply. It simply means you can take a module designed for another platform and run it in a sandbox and make it work in Linux. The solution is brilliant.
Not really. Tainted means that you loaded some "closed" software, which in
Not necessarily. The "tainted" flag has several different meanings. One (probably the one you're thinking of) is "non-GPL", but that doesn't necessarily mean closed. Apart from that you also have the "unsupported" tainting flag.
On Wednesday 20 September 2006 02:16, Carlos E. R. wrote:
Not really. Tainted means that you loaded some "closed" software,
No Carlos, Taint does not automatically = closed source. You can get tainted just by loading anything that the kernel developers don't have source for, even if the source is not closed. There was a thread here just last week mentioning someone's inability or disinclination to walk the mine field of getting something into the kernel. Its become rather an onerous task these days even if you are willing to donate the code. -- _____________________________________ John Andersen
Er, wouldn't it be better to explain this with out tossing FUD around like some microsoft salesman? ... you mean Linux salesman... me thinks Shoot Yourself in the Foot? By loading a wireless driver? For god sakes give it a rest man! (depending on your card, and which "blob" you load into ndiswrapper) anyone with a couple of cute scripts (in close proximity) can dump a root
You sound like a tainted kernel is a hanging offense. Of course not, don't be silly, nobody said that.
80% of linux users are running tainted kernels. (I pulled that statistic out of my ass, - go ahead and prove me wrong). But the point is that (unlike Windoze users) you don't have to. If you want an open stable system that you can check, rely on, and get support for from the open linux kernel community you can... unless you taint the kernel... then you're on your own. Nobody is gonna hang you... but nobody is gonna help you either.
Taint simply means one is doing more with the machine than the original kernel developers were able to supply. It simply means you can take a module designed for another platform and run it in a sandbox and make it work in Linux. The solution is brilliant. No way. Carlos answered it well... but here's my two cents also... tainted means that you have tacked a piece of code (call it a blob) that no one can check (we don't know what it really does under its black cover) and
On Wednesday 20 September 2006 02:35, John Andersen wrote: package on your "tainted" system without you detecting it... take your files, modify your data, and leave your machine trashed. This has been demonstrated at the latest black hat convention very effectively. The reason for this is that many of the current drivers for Windoze are not robust enough to handle data overuns. If you are running ndiswrapper at your local wifi cafe you are vulnerable... if you don't like calling it "shooting yourself in the foot" then call it something else.... but it ain't gonna happen on my system. therefore no one can support. If you have enough of those blobs running along side your otherwise stable kernel then the kernel is unstable. This is the primary problem with Windoze.
If ndiswrapper is the only way you get you network card working you either open you wallet and fork over $35 for a card that works or climb down off your ivory tower and load ndiswrapper and go about your business.
True enough... no one said anything any different. I have used ndiswrapper myself; until I could build (or find) a better driver. But I knew the vulnerability and worked hard to correct it. Sometimes taking risks is necessary, but it should never be the status quo. Correct design and open philosophy are the best. And by the way... the hardware manufacturers that provide open source drivers for linux *are* the ones that will get my money for hardware... its called voting with your bill-fold.
There is ABSOLUTELY no evidence that open source drivers prevent the "wireless standards" from being "badly exposed with security holes". (whatever the hell that means). Untrue. What it means is that a well written driver... and most of them are not... will not allow root packages to be dumped onto the machine by using a buffer overun vulnerability. A closed driver cannot be changed and
therefore cannot be corrected quickly by the community; however, an open driver *can* be corrected quickly by the open source community to correct for the problem (patch and recompile) and eliminate the vulnerability. Also, anyone with the knowledge can *check* the code to see what it is doing should they suspect vulnerability or naughty programmer tricks. the black hat conference pretty definitely demonstrated the problem with most wifi cards/drivers and very assuredly proved my point here. Check it out...
Open source drivers have to ad-hear to the same standards. If WEP is broken and weak, it will still be broken and weak with open source drivers.
Let me summarize:
You don't like Windows. True enough. Ndiswrapper allows you to use windows drivers in Linux. Therefore, Ndiswrapper is evil. No. Ndiswrapper is a vulnerability with known risks.
Time to grow up Mr Harris. How rude. This isn't an issue of maturity. Buy what ever card you want but don't trash a very clever solution to migrating a product from one platform to another just because of your political beliefs. Nobody trashed anything. Another user wanted to know why some of the community look unfavorably on ndiswrapper... I answered him openly and honestly and told him the truth. Philosophically (not politically) I believe in the open source gpl commitment of the linux kernel development community--- I will try whenever asked to explain why openly and honestly so
See above. that the person asking the question has fair data on which to base their decision. I will at the same time not try to hide my anti-M$ bias. The M$ company has become evil in my opinion, and there is a better way. I think it is best for the community when we can discuss the reasons why in a very open and frank way. At the same time many of us are working hard to make sure that the mistakes of M$ are not replicated in the open linux community. -- Kind regards, M Harris <><
Why do some in the Linux community treat ndiswrapper with such contempt?
Though ndiswrapper itself is open, ndiswrapper loads a closed-source proprietary windows binary blob, which makes the whole kernel potentially unstable again. Just like the NVIDIA blob.
It seems to me that if something works, especially as well as ndiswrapper, what damned difference does it make if the device drivers were written for use in Windows?
Because the bad code parts can even bring down the Linux part that works. Jan Engelhardt --
Why do some in the Linux community treat ndiswrapper with such contempt? It seems to me that if something works, especially as well as ndiswrapper, what damned difference does it make if the device drivers were written for use in Windows? I rather think that the concept should be expanded to include the universe of devices that are not currently useable with Linux due to unavailable drivers. IMHO, ndiswrapper is a good thing in that it allows us to use hardware that we otherwise would not be able to use. Hopefully, some of these chip manufacturers will provide native Linux drivers. On the down side, since we can use the Windows drivers on Linux, there is less of an incentive for
On Tuesday 19 September 2006 9:45 pm, Stevens wrote:
them to produce a Linux kernel driver.
For a very long time, we've seen many chip manufacturers who refuse to
produce open source drivers because they do not want to open up their
interfaces and secrets to the world. Companies like nVidia and ATI provide
binary Linux drivers, but no open source. Other companies may just not have
the software engineering staff to produce a driver. But, our open source
community will simply reverse engineer drivers.
the key to fixing this is in the enterprise. This is where the big
enterprise vendors, like IBM and HP support Linux on their products and
need to make sure that Linux can run on them.
--
Jerry Feldman
participants (9)
-
Anders Johansson -
Carlos E. R. -
Darryl Gregorash -
Jan Engelhardt -
Jeff Rollin -
Jerry Feldman -
John Andersen -
M Harris -
Stevens