After reviewing the archives of the various lists I have found no discussion on the topic of wireless driver inclusion in FC.
I have checked with LICENSE and FAQ for the intel driver and firmware and was wondering if it could be included in core. Without those driver it makes most centrino based laptops useless without special handling.
Specifically since I know people will ask, the problem must lie in the firmware since the driver is GPL. The firmware is under a commercial licence that allows redistribution as long as a small number of conditions are met. Distributed by rpm and installed the only condition that needs to be met is that the LICENCE file needs to be installed with the firmware.
http://ipw2100.sourceforge.net/firmware.php?fid=1
"*Your rights to redistribute the Software shall be contingent upon your installation of this Agreement in its entirety in the same directory as the Software.*"
So, is this a non-starter, or can this be accomodated by FC? I know this would be a bog help to the laptop users out there. At least if we can get the driver support in core so that each kernel release does not break laptop support I would be most grateful.
Cheers, Eric
On Mon, 2005-02-28 at 12:09 -0500, Eric Warnke wrote:
After reviewing the archives of the various lists I have found no discussion on the topic of wireless driver inclusion in FC.
I have checked with LICENSE and FAQ for the intel driver and firmware and was wondering if it could be included in core. Without those driver it makes most centrino based laptops useless without special handling.
The actual _driver_ has been included in Core kernels for some time already.
Specifically since I know people will ask, the problem must lie in the firmware since the driver is GPL. The firmware is under a commercial licence that allows redistribution as long as a small number of conditions are met. Distributed by rpm and installed the only condition that needs to be met is that the LICENCE file needs to be installed with the firmware.
http://ipw2100.sourceforge.net/firmware.php?fid=1
"*Your rights to redistribute the Software shall be contingent upon your installation of this Agreement in its entirety in the same directory as the Software.*"
This is a non-starter. If the package is not under a FOSS license, which this is not, then it cannot be included in Core. The Intel firmware is not under such a license.
Dan
The actual _driver_ has been included in Core kernels for some time already.
Having just purchased a laptop with PRO/Wireless 8915 chipset I'm understandably keen to see it supported out of the box in Fedora.
I'm content to download the firmware as a separate package as it isn't FOSS and I think Intel could do better here. Are you listening Intel?
I'm thrilled to see that both the drivers for the ipw2100 and the ipw2200 included in the kernel.
BUT... (come on you knew it was coming ;-])
... it seems the the ipw2200 driver in the kernel is 0.13 (based on a less of /lib/modules/2.6.10-1.1155_FC4/kernel/drivers/net/wireless/ipw2200/ipw2200.ko,
and then looking through the strings and seeing this:
<6>ipw2200: Intel(R) PRO/Wireless 2200/2915 Network Driver, 0.13
that the current version used it 0.13. Given that the current version is 1.0.1 and that there's been about 8 or nine releases since 0.13, what needs to be done to get a more up-to-date version included in the kernel?
Rodd
On Wed, Mar 02, 2005 at 11:41:46AM +1100, Rodd Clarkson wrote:
... it seems the the ipw2200 driver in the kernel is 0.13 (based on a less of /lib/modules/2.6.10-1.1155_FC4/kernel/drivers/net/wireless/ipw2200/ipw2200.ko,
and then looking through the strings and seeing this:
<6>ipw2200: Intel(R) PRO/Wireless 2200/2915 Network Driver, 0.13
that the current version used it 0.13. Given that the current version is 1.0.1 and that there's been about 8 or nine releases since 0.13, what needs to be done to get a more up-to-date version included in the kernel?
go to your nearest neigborhood friendly apt/yum repositories!
http://atrpms.net/dist/fc3/ipw2100/ http://atrpms.net/dist/fc3/ipw2100-testing/ http://atrpms.net/dist/fc3/ipw2200/ http://atrpms.net/dist/fc3/ipw2200-testing/
so, e.g.:
apt-get install ipw2200-kmdl-$(uname -r)
firmware will get lumped in as well.
HTH, Tim
On Tue, 2005-03-01 at 16:59 -0800, Tim Fenn wrote:
On Wed, Mar 02, 2005 at 11:41:46AM +1100, Rodd Clarkson wrote:
... it seems the the ipw2200 driver in the kernel is 0.13 (based on a less of /lib/modules/2.6.10-1.1155_FC4/kernel/drivers/net/wireless/ipw2200/ipw2200.ko,
and then looking through the strings and seeing this:
<6>ipw2200: Intel(R) PRO/Wireless 2200/2915 Network Driver, 0.13
that the current version used it 0.13. Given that the current version is 1.0.1 and that there's been about 8 or nine releases since 0.13, what needs to be done to get a more up-to-date version included in the kernel?
go to your nearest neigborhood friendly apt/yum repositories!
http://atrpms.net/dist/fc3/ipw2100/ http://atrpms.net/dist/fc3/ipw2100-testing/ http://atrpms.net/dist/fc3/ipw2200/ http://atrpms.net/dist/fc3/ipw2200-testing/
Useful to know, but it still leaves the question open, why are the current FC kernels (I'm using the rawhide kernels so the files above aren't a lot of use) sporting such an old driver for ipw2200, and what do I need to do to encourage the kernel maintainers for FC to update the driver?
Rodd
On Wed, Mar 02, 2005 at 01:29:42PM +1100, Rodd Clarkson wrote:
On Tue, 2005-03-01 at 16:59 -0800, Tim Fenn wrote:
On Wed, Mar 02, 2005 at 11:41:46AM +1100, Rodd Clarkson wrote:
... it seems the the ipw2200 driver in the kernel is 0.13 (based on a less of /lib/modules/2.6.10-1.1155_FC4/kernel/drivers/net/wireless/ipw2200/ipw2200.ko,
and then looking through the strings and seeing this:
<6>ipw2200: Intel(R) PRO/Wireless 2200/2915 Network Driver, 0.13
that the current version used it 0.13. Given that the current version is 1.0.1 and that there's been about 8 or nine releases since 0.13, what needs to be done to get a more up-to-date version included in the kernel?
go to your nearest neigborhood friendly apt/yum repositories!
http://atrpms.net/dist/fc3/ipw2100/ http://atrpms.net/dist/fc3/ipw2100-testing/ http://atrpms.net/dist/fc3/ipw2200/ http://atrpms.net/dist/fc3/ipw2200-testing/
Useful to know, but it still leaves the question open, why are the current FC kernels (I'm using the rawhide kernels so the files above aren't a lot of use) sporting such an old driver for ipw2200, and what do I need to do to encourage the kernel maintainers for FC to update the driver?
it's been on my todo list for a while. I'll get to it this week. rawhide will get it in the next day or two hopefully. fc2/fc3 will probably be in the update after next.
Dave
On Tue, 2005-03-01 at 21:31 -0500, Dave Jones wrote:
On Wed, Mar 02, 2005 at 01:29:42PM +1100, Rodd Clarkson wrote:
On Tue, 2005-03-01 at 16:59 -0800, Tim Fenn wrote:
On Wed, Mar 02, 2005 at 11:41:46AM +1100, Rodd Clarkson wrote:
... it seems the the ipw2200 driver in the kernel is 0.13 (based on a less of /lib/modules/2.6.10-1.1155_FC4/kernel/drivers/net/wireless/ipw2200/ipw2200.ko,
and then looking through the strings and seeing this:
<6>ipw2200: Intel(R) PRO/Wireless 2200/2915 Network Driver, 0.13
that the current version used it 0.13. Given that the current version is 1.0.1 and that there's been about 8 or nine releases since 0.13, what needs to be done to get a more up-to-date version included in the kernel?
go to your nearest neigborhood friendly apt/yum repositories!
http://atrpms.net/dist/fc3/ipw2100/ http://atrpms.net/dist/fc3/ipw2100-testing/ http://atrpms.net/dist/fc3/ipw2200/ http://atrpms.net/dist/fc3/ipw2200-testing/
Useful to know, but it still leaves the question open, why are the current FC kernels (I'm using the rawhide kernels so the files above aren't a lot of use) sporting such an old driver for ipw2200, and what do I need to do to encourage the kernel maintainers for FC to update the driver?
it's been on my todo list for a while. I'll get to it this week. rawhide will get it in the next day or two hopefully. fc2/fc3 will probably be in the update after next.
Dave,
Thanks for your quick reply letting me know where things stand.
Much appreciated 8-]
Rodd
On Tue, 2005-03-01 at 21:31 -0500, Dave Jones wrote:
it's been on my todo list for a while. I'll get to it this week. rawhide will get it in the next day or two hopefully. fc2/fc3 will probably be in the update after next.
Dave
Dave,
Noted that 'strings ipw2200.ko | less' on kernel-2.6.11-1.1166_FC4 now shows:
<6>ipw2200: Intel(R) PRO/Wireless 2200/2915 Network Driver, 1.0.0
so I think it's time to say, "Thank you very much."
Thank you very much!
Rodd
Noted that 'strings ipw2200.ko | less' on kernel-2.6.11-1.1166_FC4 now shows:
<6>ipw2200: Intel(R) PRO/Wireless 2200/2915 Network Driver, 1.0.0
so I think it's time to say, "Thank you very much."
Just a quick question. Was there a reason why the ipw2200 drivers weren't updated to 1.0.1, instead of 1.0.0?
A look at a changes from 1.0.0 to 1.0.1 seem worthwhile and it was released on Feb 08, 05 so should be well tested.
Fixed #559: iwconfig rate support (thanks to Florian Hackenberger) NOTE: Configuring a subset of rates that excludes mandatory rates by your AP may result in not being able to associate with your AP. * Improved link signal quality calculation (thanks to Bill Moss) * Added additional debug output during scan to dump all 16 bits of capability field to IEEE80211_DL_SCAN output * Added support for Intel PRO/Wireless 2225BG Network Connection adapter * Removed trailing whitespace on lines in code (thanks to Henrik Brix Andersen) * Yanked script trace (-x) from helper scripts (thanks to Henrik Brix Andersen) * Fixed a problem with sensitivity threshold during association (thanks to Raphael Slinckx for troubleshooting with me on IRC for three days) * Added iwpriv for turning forcing long preamble support: iwpriv eth1 set_preamble 1|0 * Possible fixes for #542 and #377 support for short preamble * Fixed #563 compilation warning on 2.6.11-rc2 (thanks to Henrik Brix Andersen) * Fixed locked BSSID reporting channel number (thanks to Pedro Ramalhais) * Fixed type-o with scan watchdog timeout message (thanks to Pedro Ramalhais) * Changed logic for displaying get_mode output so the code is easier to follow (thanks to Pedro Ramalhais) * Added initial support for WPA (thanks to Yi Zhu) -- tested with wpa_supplicant (either tip w/ ipw driver, or with -Dipw2100) with both CCMP and TKIP * Fixed problem with CCMP not working due to uninitialized 802.11 header fields (thanks to Pedro Ramalhais) * Fixed unload script to unload the ieee80211 modules in the correct order
R.
Rodd Clarkson wrote:
Noted that 'strings ipw2200.ko | less' on kernel-2.6.11-1.1166_FC4 now shows:
<6>ipw2200: Intel(R) PRO/Wireless 2200/2915 Network Driver, 1.0.0
so I think it's time to say, "Thank you very much."
Just a quick question. Was there a reason why the ipw2200 drivers weren't updated to 1.0.1, instead of 1.0.0?
Probably becuase the odd ending number represnets a development driver, not stable. 1.0.2 will be the next stable release.
Cheers, Eric
On Sun, 2005-03-06 at 20:03 -0500, Eric Warnke wrote:
Just a quick question. Was there a reason why the ipw2200 drivers weren't updated to 1.0.1, instead of 1.0.0?
Probably becuase the odd ending number represnets a development driver, not stable. 1.0.2 will be the next stable release.
Ah, that's good answer. All these different version numbering schemes can get confusing (and I just assumed it was a little like gnome or the kernel).
Rodd
Rodd Clarkson wrote:
Noted that 'strings ipw2200.ko | less' on kernel-2.6.11-1.1166_FC4 now shows:
<6>ipw2200: Intel(R) PRO/Wireless 2200/2915 Network Driver, 1.0.0
so I think it's time to say, "Thank you very much."
Just a quick question. Was there a reason why the ipw2200 drivers weren't updated to 1.0.1, instead of 1.0.0?
1.0.1 is an unstable release.
To quote ipw2200.sourceforge.net:
Stable versions can be identified by the last digit of the version number being a 0 (zero). All other releases are development snapshots.
Regards, Owen
On Mon, 07 Mar 2005 11:49:08 +1100, Rodd Clarkson wrote:
A look at a changes from 1.0.0 to 1.0.1 seem worthwhile and it was released on Feb 08, 05 so should be well tested.
Although 1.0.1 has a number of bug fixes, it also contains a number of uncompleted new features, and it is considered unstable. The next stable will be 1.1.0. I hate this version scheme, but it's not under my influence.
However, 1.0.1 is the first release supporting WPA, but it breaks a lot of stuff for people using NetworkManager to connect wireless on less then ideal link quality.
Regards, Tomislav
On Mon, 28 Feb 2005 12:09:47 -0500, Eric Warnke wrote:
After reviewing the archives of the various lists I have found no discussion on the topic of wireless driver inclusion in FC.
Same with Atmel at76c50x wireless chips.
See http://thekelleys.org.uk/atmel/
Debian : non-free [http://packages.debian.org/unstable/net/atmel-firmware], SuSE : GPL, Other License(s), packaged. [http://www.novell.com/products/linuxpackages/professional/atmel-firmware.htm...] Ubuntu : In main distribution. [http://higgs.djpig.de/ubuntu/www/hoary/net/atmel-firmware]
Copyright File here:
/******************************************************************************/ /* Copyright (c) 2004-07-05 Atmel Corporation. All Rights Reserved. */ /* */ /* Redistribution and use of the microcode software ("Firmware") is */ /* permitted provided that the following conditions are met: */ /* Firmware is redistributed in object code only, specifically, only */ /* in two file formats: (a) .h header file; or (b) .rom binary image file; */ /* */ /* Any reproduction of Firmware must contain the above copyright notice, */ /* this list of conditions and the below disclaimer in the documentation */ /* and/or other materials provided with the distribution; and */ /* The name of Atmel Corporation may not be used to endorse or promote */ /* products derived from this Firmware without specific prior written consent.*/ /******************************************************************************/
/******************************************************************************/ /* DISCLAIMER: ATMEL PROVIDES THIS FIRMWARE "AS IS" WITH NO WARRANTIES */ /* OR INDEMNITIES WHATSOEVER. ATMEL EXPRESSLY DISCLAIMS ANY EXPRESS, */ /* STATUTORY OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, */ /* THE IMPLIED WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR */ /* PURPOSE AND NON-INFRINGEMENT. IN NO EVENT SHALL ATMEL BE LIABLE FOR */ /* ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL */ /* DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS */ /* OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) */ /* HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, */ /* STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING */ /* IN ANY WAY OUT OF THE USE OF THIS FIRMWARE, EVEN IF ADVISED OF THE */ /* POSSIBILITY OF SUCH DAMAGE. */ /* */ /* USER ACKNOWLEDGES AND AGREES THAT THE PURCHASE OR USE OF THE FIRMWARE */ /* WILL NOT CREATE OR GIVE GROUNDS FOR A LICENSE BY IMPLICATION, ESTOPPEL, */ /* OR OTHERWISE IN ANY INTELLECTUAL PROPERTY RIGHTS */ /* (PATENT, COPYRIGHT, TRADE SECRET, MASK WORK, OR OTHER PROPRIETARY RIGHT) */ /* EMBODIED IN ANY OTHER ATMEL HARDWARE OR FIRMWARE EITHER SOLELY */ /* OR IN COMBINATION WITH THE FIRMWARE. */ /******************************************************************************/
Is this a package for livna, core or extras?
Richard.
On Mon, 2005-02-28 at 17:41 +0000, Richard Hughes wrote:
On Mon, 28 Feb 2005 12:09:47 -0500, Eric Warnke wrote:
After reviewing the archives of the various lists I have found no discussion on the topic of wireless driver inclusion in FC.
Same with Atmel at76c50x wireless chips. Is this a package for livna, core or extras?
Livna.
Dan
On Mon, 28 Feb 2005 17:41:47 +0000, Richard Hughes ee21rh@surrey.ac.uk wrote:
Is this a package for livna, core or extras?
If its not foss its livna. Fedora Core and Extras are chartered to be solely foss material.
-jef
Then I assume it'a an oversight that other closed source firmware ships with Fedora?
The bluez-bluefw module ships with a GLP license even though the firmware that is included appears ( as in I'm email broadcom right now ) to be free as in beer not free as in speech. Without the source code it's only Free and not FOSS.
If they can ship free-redistibutution firmware for the broadcom why not intel's free-redistribution license?
Cheers, Eric
Jeff Spaleta wrote:
On Mon, 28 Feb 2005 17:41:47 +0000, Richard Hughes ee21rh@surrey.ac.uk wrote:
Is this a package for livna, core or extras?
If its not foss its livna. Fedora Core and Extras are chartered to be solely foss material.
-jef
On Mon, 2005-02-28 at 14:50 -0500, Eric Warnke wrote:
Then I assume it'a an oversight that other closed source firmware ships with Fedora?
The bluez-bluefw module ships with a GLP license even though the firmware that is included appears ( as in I'm email broadcom right now ) to be free as in beer not free as in speech. Without the source code it's only Free and not FOSS.
If they can ship free-redistibutution firmware for the broadcom why not intel's free-redistribution license?
That would appear to be an oversight. File a bug against the package.
Dan
On Mon, 2005-02-28 at 14:58 -0500, Dan Williams wrote:
The bluez-bluefw module ships with a GLP license even though the firmware that is included appears ( as in I'm email broadcom right now ) to be free as in beer not free as in speech. Without the source code it's only Free and not FOSS.
If they can ship free-redistibutution firmware for the broadcom why not intel's free-redistribution license?
That would appear to be an oversight. File a bug against the package.
Which is the oversight. That the bluez-bluefw module is shipped, or that the intel centrino stuff isn't?
Rodd
Eric Warnke wrote:
After reviewing the archives of the various lists I have found no discussion on the topic of wireless driver inclusion in FC.
Along these same lines is the Ralink 2500/2400 chip sets. They power a number of useful 802.11 b/g devices (most notably linksys PCI revision 4 here in the US which is a large number of devices sold). They gpl'd their source code back in December IIRC. As wireless NICs are becoming more popular and network install approach "default" (with FC5 anaconda and newly discussed capabilities) it will become increasingly more important to get these "popular" cards under "Just Works" functionality during installs. What is required to get ootb functionality for a device like this? Here are some relevant links:
Ralink site with source: http://www.ralinktech.com.tw/supp-1.htm
rt2x00 Open Source Project http://rt2x00.serialmonkey.com/wiki/index.php/Main_Page
-mf
tir, 01.03.2005 kl. 03.12 skrev Michael Favia:
Eric Warnke wrote:
After reviewing the archives of the various lists I have found no discussion on the topic of wireless driver inclusion in FC.
Along these same lines is the Ralink 2500/2400 chip sets. They power a number of useful 802.11 b/g devices (most notably linksys PCI revision 4 here in the US which is a large number of devices sold). They gpl'd their source code back in December IIRC. As wireless NICs are becoming more popular and network install approach "default" (with FC5 anaconda and newly discussed capabilities) it will become increasingly more important to get these "popular" cards under "Just Works" functionality during installs. What is required to get ootb functionality for a device like this? Here are some relevant links:
Ralink site with source: http://www.ralinktech.com.tw/supp-1.htm
rt2x00 Open Source Project http://rt2x00.serialmonkey.com/wiki/index.php/Main_Page
-mf
Same aplies to the adm8211 driver - its all GPL - and heading for the mainline kernel. It just needs to get all fixed up and "perfect" first :)
On Mon, 2005-02-28 at 20:12 -0600, Michael Favia wrote:
Eric Warnke wrote:
After reviewing the archives of the various lists I have found no discussion on the topic of wireless driver inclusion in FC.
Along these same lines is the Ralink 2500/2400 chip sets. They power a number of useful 802.11 b/g devices (most notably linksys PCI revision 4 here in the US which is a large number of devices sold). They gpl'd their source code back in December IIRC. As wireless NICs are becoming more popular and network install approach "default" (with FC5 anaconda and newly discussed capabilities) it will become increasingly more important to get these "popular" cards under "Just Works" functionality during installs. What is required to get ootb functionality for a device like this? Here are some relevant links:
Ralink site with source: http://www.ralinktech.com.tw/supp-1.htm
rt2x00 Open Source Project http://rt2x00.serialmonkey.com/wiki/index.php/Main_Page
Having followed the rt2x00 development I know the guys are keen to get this driver in the kernel eventually. The rt2x00 cards are avaiable cheaply in Australia in both PCI and PCMCIA formats and work well, so there's quite a few people using them
The intention is the unify their rt2400 and rt2500 drivers in a completely new base currently under development, the rt2x00. This driver compiles (sometimes) but isn't close to use, however they are working hard too make sure that it's fit for inclusion in the kernel at such a time that it is ready (whatever that means).
The rt2x00 is intended to share wireless infrastructure in the kernel. It will also move from using ra[x] as it's device name (something that comes from the RaLink driver) to using a more appropriate device name in line with current kernel thinking (I think this means using eth[x], but there seems to be another identifier used for some wireless cards, so this needs to be thought through.)
One think I'm sure of is that they would welcome anyone willing to get in and contribute the the development of the new (or exisiting) drivers.
Hope that gives a little heads up.
Rodd
On Mon, 2005-02-28 at 20:12 -0600, Michael Favia wrote:
As wireless NICs are becoming more popular and network install approach "default" (with FC5 anaconda and newly discussed capabilities) it will become increasingly more important to get these "popular" cards under "Just Works" functionality during installs.
Mmmm, this makes me wonder how Fedora will handle cards like the ipw2200 (Intel PRO/Wireless cards).
The driver for these cards is included in the kernel because it's under the GPL. However, the firmware won't be included because of it's licensing requirements.
There are other cards that are in a similar situation, AFAIK.
Will Fedora be set up so that the firmware can be provided as a third party during the install?
I guess the way I see this happening would be for the installer to note that the computer it's installing on has a wireless card (or wired card) that needs 'extra' information to work, and then prompts the installee to provide this on some sort of meda.
Will this be possible, or are users of these cards be out of luck?
Also, as a related issue, will fedora then install the firmware so that the user doesn't need to do this again on first boot. This would make sense, and given that the user has provided the firmware once, would presumably be appropriate.
Something to think about.
Rodd
I know this will open a large can of worms, so please take this message as mearly a way of opening a dialog on the subject.
Objecive 2 of the Fedora "contract" specifies that it should be built exclusivly from open source software.
But is firmware software? It is more like a document, image, or other fixed item. What about Creative Commons "by" and "by-nd" licensed documentation are those items locked out of Fedora as well?
I think we should re-evaluate the belief that locking out "free as in beer" firmware ( and all firmware is only "free as in beer" unless they provide the source ) is the proper action for the community when these files enable open source operating systems to interact with otherwise propratary hardware. I think that the first objective of the "contract" is to provide a general operating system platform equivilant to other competing systems must be evaluated before locking these files out. To say that these types of firmware are no longer welcome I think will hurt us long term as the hardware get smarter and more flexable. Communications hardware especially that are generally bound by one or more legal body to constrain it's abilities. Hardware manufactures who open their hardware to blanket hacking risk loosing the license to sell the hardware, so don't expect them to open the firmware as that is legally not an option for them.
Is this lock out based on ideology or reality. Including "free as in beer" firmware does not compromise the ability to redistribute Fedora freely and I think it moves us closer to the goal of a genreal-purpose operating system out of the box. It improves user feedback and reduces problamatic and frustrating installs for both users and the people that are drawn in to help new users get past these repetitive and common problems.
Cheers, Eric
On Wed, 2005-03-02 at 23:56 -0500, Eric Warnke wrote:
I know this will open a large can of worms, so please take this message as mearly a way of opening a dialog on the subject.
Objecive 2 of the Fedora "contract" specifies that it should be built exclusivly from open source software.
But is firmware software? It is more like a document, image, or other fixed item. What about Creative Commons "by" and "by-nd" licensed documentation are those items locked out of Fedora as well?
It's an interesting take on firmware.
It makes me wonder if there's a lot of difference between Fedora trademarks and firmware, in that both are free-as-in-beer, but if you wish to roll your own fedora (AFAIK) you have to remove all the Fedora trademarks.
Or am I totally confused?
Rodd
On Wed, Mar 02, 2005 at 11:56:19PM -0500, Eric Warnke wrote:
I think we should re-evaluate the belief that locking out "free as in beer" firmware
For ipw2100, this misses the point; ISTR we are shipping several other firmware images in Fedora without source.
The ipw2100 license doesn't allow free redistribution (the recipient must explicitly agree to the license); you can't just burn a CD including that firmware and give it to someone.
I don't know whether you condider that "free as in beer", but IMHO sacrificing redistributability of FC for the firmware is just not worth it, all other factors aside. Mirek
Miloslav Trmac wrote:
The ipw2100 license doesn't allow free redistribution (the recipient must explicitly agree to the license); you can't just burn a CD including that firmware and give it to someone.
No it does not. To redistribute you must only agreee to copy the license in it's entirety to the same location as the software. As computer users firmware has and will be a fact of life. It's firmware that takes propriataty x86 hardware and makes it into a commodity platform. Without the BIOS we would all be hacking individual registers in each componient chip to try and get the memory subsystems initalized and stable. Without firmware many parts of the system would be useless.
I just think that someone with the appropriate level of knowledge should review the license to see if will truly prevent redistribution of the software. Based on my reading of the license it does stop copying and it's license clauses are no more burdensom than the GPL in their requirements.
Cheers, Eric
I just think that someone with the appropriate level of knowledge should review the license to see if will truly prevent redistribution of the software. Based on my reading of the license it does stop copying and it's license clauses are no more burdensom than the GPL in their requirements.
last time I looked the license was different for end users and "os distributors". The end user one was as you describe, but the other one needed special permission.
--On Thursday, March 03, 2005 7:27 AM -0500 Eric Warnke eric@snowmoon.com wrote:
It's firmware that takes propriataty x86 hardware and makes it into a commodity platform. Without the BIOS we would all be hacking individual registers in each componient chip to try and get the memory subsystems initalized and stable.
USB host controllers seem to have succeeded with a standardized register layout and no proprietary firmware.
Hi
Is this lock out based on ideology or reality. Including "free as in beer" firmware does not compromise the ability to redistribute Fedora freely and I think it moves us closer to the goal of a genreal-purpose operating system out of the box.
but only if the firmware is actually freely redistributable. many vendors do not agree to this. OpenBSD had a users campaign and they did get some success with it but intel has refused to cooperate
===== Regards Rahul Sundaram
__________________________________ Celebrate Yahoo!'s 10th Birthday! Yahoo! Netrospective: 100 Moments of the Web http://birthday.yahoo.com/netrospective/
tor, 03.03.2005 kl. 05.56 skrev Eric Warnke:
I know this will open a large can of worms, so please take this message as mearly a way of opening a dialog on the subject.
Objecive 2 of the Fedora "contract" specifies that it should be built exclusivly from open source software.
But is firmware software? It is more like a document, image, or other fixed item. What about Creative Commons "by" and "by-nd" licensed documentation are those items locked out of Fedora as well?
I think we should re-evaluate the belief that locking out "free as in beer" firmware ( and all firmware is only "free as in beer" unless they provide the source ) is the proper action for the community when these files enable open source operating systems to interact with otherwise propratary hardware. I think that the first objective of the "contract" is to provide a general operating system platform equivilant to other competing systems must be evaluated before locking these files out. To say that these types of firmware are no longer welcome I think will hurt us long term as the hardware get smarter and more flexable. Communications hardware especially that are generally bound by one or more legal body to constrain it's abilities. Hardware manufactures who open their hardware to blanket hacking risk loosing the license to sell the hardware, so don't expect them to open the firmware as that is legally not an option for them.
Is this lock out based on ideology or reality. Including "free as in beer" firmware does not compromise the ability to redistribute Fedora freely and I think it moves us closer to the goal of a genreal-purpose operating system out of the box. It improves user feedback and reduces problamatic and frustrating installs for both users and the people that are drawn in to help new users get past these repetitive and common problems.
Cheers, Eric
IMO firmware is just a part of the hardware. It's not something running on the main CPU etc.
Besides, the chances to get it opened, are probably really small
On Mon, Feb 28, 2005 at 12:09:47PM -0500, Eric Warnke wrote:
After reviewing the archives of the various lists I have found no discussion on the topic of wireless driver inclusion in FC.
I have checked with LICENSE and FAQ for the intel driver and firmware and was wondering if it could be included in core. Without those driver it makes most centrino based laptops useless without special handling.
Specifically since I know people will ask, the problem must lie in the firmware since the driver is GPL. The firmware is under a commercial licence that allows redistribution as long as a small number of conditions are met. Distributed by rpm and installed the only condition that needs to be met is that the LICENCE file needs to be installed with the firmware.
http://ipw2100.sourceforge.net/firmware.php?fid=1
"*Your rights to redistribute the Software shall be contingent upon your installation of this Agreement in its entirety in the same directory as the Software.*"
So, is this a non-starter, or can this be accomodated by FC? I know this would be a bog help to the laptop users out there. At least if we can get the driver support in core so that each kernel release does not break laptop support I would be most grateful.
Try the packages at
http://ATrpms.net/name/ipw2200/ http://ATrpms.net/name/ipw2100/ http://ATrpms.net/name/ipw2200-firmware/ http://ATrpms.net/name/ipw2100-firmware/
and, if you are adventurous
http://ATrpms.net/name/ipw2200-testing/ http://ATrpms.net/name/ipw2100-testing/
devel@lists.stg.fedoraproject.org
-
Arjan van de Ven -
Axel Thimm -
Dan Williams -
Dave Jones -
David Woodhouse -
Eric Warnke -
Jef Spaleta -
Kenneth Porter -
Kyrre Ness Sjobak -
Michael Favia -
Miloslav Trmač -
Owen Taylor -
Rahul Sundaram -
Richard Hughes -
Rodd Clarkson -
Tim Fenn -
Tomislav Vujec