Red Hat is no stranger to lawsuits, having grappled with the Firestar patent case in 2008 and dealt successfully with patent troll IP Innovations in
2010. But the company is now bringing a GPL compliance suit to court
for the first time in its history. Red Hat filed the complaint in a
countersuit against a patent infringement case that was launched in
early 2012. If it goes to trial, it could bring several
GPL-interpretation questions to the test.
The original litigant in the case is Twin Peaks Software (TPS), which
makes proprietary network backup software. TPS sued Red Hat and Red
Hat's recently-acquired subsidiary Gluster on February 23, 2012. TPS
charges that the GlusterFS software violates TPS's US patent
7,418,439, which covers TPS's "Mirror File System" (MFS). GlusterFS is a
network filesystem that aggregates multiple storage shares into a
single volume. TPS's products include TPS Replication Plus, which
automatically mirrors changes between two NFS filesystems over the
network, and TPS Clustering Plus, which extends a similar feature set
to larger clusters.
Red Hat initially responded to the patent infringement suit on August
2, both denying the infringement and asserting that the patent itself
because the alleged invention of the ’439 Patent lacks utility, is
taught by, suggested by, and/or, anticipated or obvious in view of the
prior art, is not enabled, and/or is unsupported by the written
description of the patented invention, and no claim of the ’439 Patent
can be validly construed to cover any Red Hat product.
Had things stopped there, the case might have proceeded as a standard
software patent infringement lawsuit. Red Hat's answer to the initial
claim invoked numerous other counterarguments, such as denying that
TPS has the right to ask for an injunction against the allegedly
infringing Red Hat products, but it stuck to denying the claims of the
initial suit. But Red Hat then followed up with a September
13 countersuit that charges TPS with a copyright violation claim, and
asks for an injunction against the violating products. The products in
question are TPS Replication Plus and TPS My Mirror, a freeware
edition of Replication Plus. Red Hat claims that both products
incorporate code from mount — specifically the 2.12a version
from the util-linux package for which Red Hat is the
registered copyright holder — and that TPS is in violation of
the terms of the license by not providing or offering the
corresponding source code.
At Groklaw, Mark Webbink argues
that this action ups the stakes considerably, because
even if TPS's suit against Red Hat were successful, Red Hat would
experience only a small impact on its bottom line, due to the
relatively minor role GlusterFS plays in Red Hat's core business. If Red
Hat's countersuit were successful, however, TPS would lose the sales
of 50% of its products — a hit few businesses could survive.
The countersuit is in most respects a standard GPL-violation charge,
much like those brought against other proprietary software vendors by
other enforcement entities. But it also brings to light some
peculiarities of the free software licensing realm. Red Hat alleges that
the mount code in question is under GPL version 2, specifically.
Failure to comply with GPLv2's source code provisions automatically
terminates the violator's rights to distribute the code (section 4).
The most common interpretation of this section of GPLv2 was that only the
copyright holder can reinstate the violator's right to distribute the
copied software. In that case, if TPS is found to have copied mount code, Red Hat could effectively force TPS to rewrite its
products by refusing to reinstate its rights under the GPLv2. But not
everyone agrees with that interpretation;
uncertainty over the meaning of that section was also one reason why GPLv3
added provisions for a violator to regain its right to distribute by
coming into compliance with the license.
Another wrinkle to the copyright-violation issue is the possibility
that there are portions of other GPL-licensed works inside TPS's
products. The countersuit does not address this possibility, but it
cannot rule it out, either. The difference between copying from one
GPL-licensed work and copying from several could be great. In the
event that there are multiple GPL violations of different copyrights,
even if Red Hat agreed to reinstate TPS's right to distribute
mount and all other Red Hat-copyrighted code,
it cannot reinstate TPS's right to distribute software written by
others. That problem is academic at the moment, but it may not remain
so: Eben Moglen wrote
on the Software Freedom Law Center (SFLC) blog that
he is investigating whether TPS's products contain software that has
been copied from SFLC clients.
Moglen also says that if a violation is proven in the TPS case, it
would be "a particularly severe offense" because TPS has
chosen to sue a member of the free software community. Consequently,
it would be profiting from the work of free software developers while
simultaneously suing them. In contrast, most GPL violations are
reported to be unintentional; Bradley Kuhn estimated
in 2011 that 98% of the violation incidents he had worked on were
cases of negligence and not malice. "Malice" might be hard to pin
down, but the fact that TPS actively initiated this legal battle
certainly increases the chances that Red Hat will choose to fight it
out rather than settle.
If Red Hat does pursue the suit, this will also be the first GPL
violation case brought by a commercial Linux distribution. Many of
the high-profile GPL compliance cases in years past have been fought
by independent projects like BusyBox or non-commercial groups like gpl-violations.org. Fighting out the GPL violation charge also has a
different feel in this case because most other GPL enforcement actions
are taken in order to bring the offending party into compliance. That
is not the goal here: Red Hat is using the charge to wage an injunction-versus-injunction battle. The highest-grossing Linux distributor pursuing
a GPL violation charge may not have the David-versus-Goliath feel of the
other cases, but it could still be an important day in court —
both for Red Hat and for anyone else who builds a business on free software.
Comments (15 posted)
The Fedora carbon footprint is getting larger:-)
-- Chuck Forsberg
If by "intuitive" you mean "the same as the old interface" then I must
agree. Otherwise, I think you are just trying to hold on to what you
-- David Lehman
I start to think that Enterprise Software is an alias for broken packages and bad software installations.
Alpha does not mean 'finished'. Alpha does not even mean 'nearly finished'. Alpha means 'here's this thing that will probably eat your babies and burn down your house, we want you to tell us exactly where the teeth marks were on the babies and what the fire smelled like'.
(Thanks to Don Waugaman)
Comments (4 posted)
The alpha release of Fedora 18 "Spherical Cow" is available for testing. "Already mooo-tivated to give F18 Alpha a try? Great! We still hope that you'll read onwards; there are fabulous features in this release you may want to know about, as well as important information regarding specific, common F18 Alpha installation issues and bugs, all of which are detailed in this release announcement.
Full Story (comments: 63)
Mandriva Linux 2012 Alpha (Tenacious Underdog) has been released
The new alpha features Linaro's gcc 4.7 branch, installer improvements,
LXDE shipped by default, KDE 4.9.0 and more.
Comments (none posted)
Ubuntu 11.04 has been out for nearly 18 months, and will reach its
end-of-life on October 28, 2012. There will be no updates, including
security updates, for Natty after that date.
Full Story (comments: none)
CentOS has released a new installer for 6.3 x86_64 that will work on UEFI
enabled machines. "I'm trying to make sure that we do enough testing
and have enough
resources for UEFI testing to ensure that the next and subsequent
releases do not have a problem in this environment. In the mean time,
the installer buildsystem for CentOS-6 has been updated to also build
and test the UEFI requirements in sync with the rest of the installer
Full Story (comments: none)
Newsletters and articles of interest
Comments (none posted)
Miklos Vajna, founder of Frugalware, looks at the history
of this general purpose distribution. "Looking back, it was all quite lame. :-) I used a mail address called "mamajom" (English translation could be "momonkey"), tied to an ISP, with a lengthy signature at the end of every mail I sent and was using my IRC nick instead of my real one everywhere… OTOH, I made some decisions I’m happy about even today. The first four developers (Ádám Zlehovszky, Krisztián Vasas, Zsolt Szalai and me) were all Hungarian and despite of this, I forced every code, test and documentation to be in English, to possibly turn the project into an international one in the future. And that proved to very, very useful.
Comments (none posted)
The H takes
at the first alpha for Mandriva Linux 2012. "Nearly two months later than originally planned, the first alpha for Mandriva Linux 2012, code-named "Tenacious Underdog", has been released for testing. The new development release upgrades the KDE desktop to version 4.9.0 from August and brings improvements to the distribution's installer, which is now said to be smaller and faster; the installer's text mode is also noted to be working again. Other changes include the complete removal of the HAL (hardware abstraction layer) and the switch to Linaro's GCC 4.7 branch, as well as various package updates and bug fixes.
Comments (none posted)
Page editor: Rebecca Sobol
Next page: Development>>