Karl-
  1. The GPL is _viral_. This was a major reason behind it's creation
and why an LGPL was also written. If you use, modify, distribute GPL
code you are under the requirements of the GPL. The LGPL allows for
linking of libraries and such with out triggering the viral effect. It
most definitely applies to derivative works. This would also very much
apply to what you claim RedHat is doing. I rather doubt that Redhat is
releasing kernels in violation of the GPL.  As I am using several
Redhat kernel's at work I will look in to your claims. If true, they
could cause some problems.

Reference:
 http://www.gnu.org/licenses/gpl.txt
 http://www.gnu.org/licenses/gpl-faq.html#DistributeExtendedBinary
 http://www.gnu.org/licenses/gpl-faq.html#ModifiedJustBinary
 http://www.gnu.org/licenses/gpl-faq.html#MereAggregation
 http://www.gnu.org/licenses/gpl-faq.html#RequiredToClaimCopyright
http://www.gnu.org/licenses/gpl-faq.html#WhyDoesTheGPLPermitUsersToPublishTheirModifiedVersions

  2. Licenses cover source code. Patents cover methodology and ideas.
You appeared to have mixed terms you email. Especially in the case of
the GPL, it is was written to make use of modern copyright law.

  3. The MIT license is a highly permissive non-viral license. It
gives you the rights to do about what  you want with the code as long
as you keep the MIT lic. on the software or any substantial portion of
it that is used. This means that you will have to split license your
code with the MIT lic but your enter work does not have to be MIT
licensed. The MIT offers far less protection and coverage of rights
than the GPL does.

Reference:
http://www.opensource.org/licenses/mit-license.php

  4. The goal of the GPL is to protect users rights not the rights of
the developers.

Reference:
 http://www.http://www.gnu.org/philosophy/free-sw.html

  5.  When providing GPL binaries you really do have to provide the
source yourself. Their are two exceptions. You may rely on an upstream
provider for your distribution of source code providing that 1) you
are merely redistributing the upstream providers unmodified binaries
and 2) you have an explicit agreement with the upstream provider
stating that they will act as your source code repository. The other
exception is in the case where you where given a written offer for the
source code. Again in this case you must be distributing the
unmodified binary and must also distribute the written offer. You must
also make the source code available for up to 3 years, this explicit
when using a written offer.

Reference:
 http://www.gnu.org/licenses/gpl.txt
 http://www.gnu.org/licenses/gpl-faq.html#DistributingSourceIsInconvenient
 http://www.gnu.org/licenses/gpl-faq.html#DistributeWithSourceOnInternet
 http://www.gnu.org/licenses/gpl-faq.html#UnchangedJustBinary
 http://www.gnu.org/licenses/gpl-faq.html#SourceAndBinaryOnDifferentSites
 http://software.newsforge.com/software/06/06/23/1728205.shtml

6. GPLv2 impose patent limitations. Namely that you can not distribute
patented encumbered code under the GPL unless the patent allows for
Royalty-free redistribution by third parties. If the program is
covered by the GPL you can not add patented code to it unless you are
wiling to give away the patents (Section 7).

Reference:
http://www.gnu.org/licenses/gpl.txt

7. Burden of prove is indeed on the accuser in a court case. But
burden of compliance is on the licensee. In this case, if someone
wants the the source for the distributed GPLed binaries then Asus must
comply. Failure of Asus to comply with that requirement then becomes
proof of failure to comply should it need to go all the way to court.
Personally I don't that would have to happen. One would only have to
point to Asus's previous screw up to remind them that they do have an
obligation to meet and everyone goes away happy.

8. Actually having the exact source code for the distributed binaries
is very helpful in get things to work. If they have in fact patched
code then you can see what is wrong in the patch  or port it to a
newer version. If in fact there is no modification to the code, then
that tells you something too, like for example it is using the parport
driver as per your example or some other user space method of driving
the LCD.

9. I will be happy to continue this debate of list, but I believe we
are now totally off topic for the mailing list

Evan

On 10/14/06, Karl Lattimer <[EMAIL PROTECTED]> wrote:

> Sorry this is a very primitive understanding of the GPL and its
> implications. The GPL is intended to protect developers rights when they
> open source it, not to force derivative works to assume the exact same
> licensing, thats what the MIT license does! For derivative  works you
> are only required to provide the original code you derived yours from
> unless you use other bits of code inside of your code. Thats where it
> gets complicated. For instance if I take a methodology for one network
> card driver which IS GPL and insert that methodology in to my network
> driver then I'd either have to change that code or release it under GPL.
> This still doesn't mean I have to release ALL of the code, just the bits
> which put the methodology in context.
>
> Even in the most complicated GPL violations the burden of proof is on
> the FSF/EFF/GNU to prove it. This is exceptionally difficult, in most
> cases the accused with provide debugging symbols to a third party to
> prove innocence.
>
>
> > If they have written all new code under a different
> > lic. they can keep that. I just want to be able to get what I am
> > legally entitled too.
>
> Fact is, what you're entitled to is already available online! thats
> where they got it from.
>
> > I have to do the same thing when I am
> > distributing packages of GPLed code. That is why there is also a
> > source directory on my website for everything I distribute. As to RH
> > kernel patches those are binary modules
>
> Actually redhat made significant changes to the kernel memory handling
> in order to provide NX for shell code protected memory.
>
> They've also made significant changes to other parts of the core kernel,
> which are in fact patches against the kernel internals. They CAN'T be
> binary modules. The idea that redhat release drivers is absurd. They
> harden linux kernels, they have code patches to do this and don't
> release that source. Even though they are modifying GPL code they do not
> have to provide the modifications they are only required to provide the
> code which is already licensed to other developers.
>
> > ( like nvidia or ATI, Yes
> > there is a disubute about this but I fall on teh pragmatic side) not
> > actual source patches to the kernel. Any source patch to the kernel
> > would have to be GPLed.
>
> As explained above NO - IT - DOESN'T!
>
> GPL doesn't over ride patents or proprietary development, if it did you
> wouldn't be able to make money from it!
>
> > I suspect that the LCD driver is a binary
> > module or userland code, but till you have the source or a good distro
> > in this case,it is kind of hard to tell.
>
> It could be an echo script pushing chars at a standard parport driver.
>
> To summarise, GPL does not prohibit derivative works or restrict
> derivative works as long as credit is acknowledged where due. Code can
> be kept secret where ever the developers see fit, unless it violates an
> existing license on existing code (duplication/IP theft, eg. nvidia
> taking GPL code for memory management and putting it in their driver as
> happened recently), is not correctly credited or credit is stolen by
> another (eg. pearPC and cherryOS).
>
> The reason the posts were deleted is probably purely because asus don't
> want GPL upstarts having a go at them when they've already been burned
> once in the past. All they are required to do is respect the copyright
> holders, distribute the GPL or tell interested parties what license it
> is and where they can get a copy of it, and where the code they used can
> be obtained.
>
> K,
>
>
>
> -------------------------------------------------------------------------
> Using Tomcat but need to do more? Need to support web services, security?
> Get stuff done quickly with pre-integrated technology to make your job easier
> Download IBM WebSphere Application Server v.1.0.1 based on Apache Geronimo
> http://sel.as-us.falkag.net/sel?cmd=lnk&kid=120709&bid=263057&dat=121642
> _______________________________________________
> Freevo-users mailing list
> Freevo-users@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/freevo-users
>

-------------------------------------------------------------------------
Using Tomcat but need to do more? Need to support web services, security?
Get stuff done quickly with pre-integrated technology to make your job easier
Download IBM WebSphere Application Server v.1.0.1 based on Apache Geronimo
http://sel.as-us.falkag.net/sel?cmd=lnk&kid=120709&bid=263057&dat=121642
_______________________________________________
Freevo-users mailing list
Freevo-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/freevo-users

Reply via email to