Re: OT: nVidia driver [was: Wish list] -- nVidia doesn't own a lot of the IP
by Bryan Smith
From: Paul Iadonisi <pri.rhl4(a)iadonisi.to>
> If I saw nVidia and ATI maybe actively participating in the antitrust
> case against Microsoft in the EU
Oh don't even go there. Anti-trust is just another word for "everyone
ganging up on a competitor." It typically solves *0* of the actual
community issues, but only competitor ones. I think the results of the
US trial -- both during the Clinton and Bush administrations -- show
that it only addressed competitor issues, not standards-based ones.
And you can be sure the EU lawsuit is more about trade than standards
too. After all, HP and Red Hat donate massive amounts of IP and real
GPL software to Linux -- in far excess of EU corporations, yet they are
being "lumped in" with IBM as exploiting Linux in recent, so-called
"reports." That _really_ exposed with what the EU is doing.
I think the epitome of this can be seen in Boeing v. Airbus and the WTO.
In other words, I don't believe any federated organization of states has
any interest in standards and other community-focused endeavors -- at
least it's very small compared to fiscally-aligned interests of its
member corporations. Case-in-point: Who was the Senator that really
got the DOJ v. MS going and what state was he from? ;->
> and joining NoSoftwarePatents.com then I'd be willing to cut them some
> slack.
Patents aren't bad. Patents on _common_ideas_ are bad. What we need to
do is lobby for a massive reduction in patents, and 10x the scrutiny,
instead of this "organ grinder" system we have.
The US has a massive patent system that's out of control. At the same
time, the US continues to be the incubator of countless technologies --
especially in the medical field. Disrespect for IP is why medicine are
dis-proportionally expensive in this country, because no one else shares
the IP burden but Americans. But without that return, there would not
be the research.
I have no argument that Microsoft is the _least_ innovative and the
_most_ IP sucking company in the world. But just because of companies
like Microsoft doesn't mean all companies are "bad."
> And take a stand against the current patent system in the US.
I think nVidia and SGI has done a lot of good for OpenGL in the past.
You should read up on their donations.
> That goes for any other company that is in this kind of mess.
> Otherwise, they are just playing the victims.
I do agree with you on companies like Red Hat trying to form "grass
roots" efforts with other, _ethical_ companies to address this.
> Or perhaps it isn't them painting *themselves* as victims. But if they
> expended the resources necessary to *change* this mess more than in
> their cloistered little worlds (with the resulting limited effect) then
> maybe we'd get somewhere.
Unfortunately, if one company does that, they just get taken advantage
of by the other companies. Which is why we should support endeavors
like Red Hat's. It could change the face of the landscape.
Your points are noted. However, I don't believe in the current trend of
federated litigation and abolishment of all software patents. Anything
forced by a federated body smells like ... well, I don't want to say
it. ;->
Community-based efforts by _choice_, like Red Hat's current efforts, are
the best way.
--
Bryan J. Smith b.j.smith(a)ieee.org
---------------------------------------------------------------------
It is mathematically impossible for someone who makes more than you
to be anything but richer than you. Any tax rate that penalizes them
will also penalize you similarly (to those below you, and then below
them). Linear algebra, let alone differential calculus or even ele-
mentary concepts of limits, is mutually exclusive with US journalism.
So forget even attempting to explain how tax cuts work. ;->
18 years, 11 months
Fwd: [smart] multilib support (was: Smart package manager)
by Jeff Pitman
FYI: smart supports multilib.
---------- Forwarded Message ----------
Subject: [smart] multilib support (was: Smart package manager)
Date: Saturday 11 June 2005 20:22
From: Axel Thimm <Axel.Thimm(a)atrpms.net>
To: smart(a)mandrivalinux.org
On Sat, Jun 11, 2005 at 08:53:56AM +0800, Jeff Pitman wrote:
> Does smart do multi-lib yet?
It does. :)
--
Axel.Thimm at ATrpms.net
-------------------------------------------------------
--
-jeff
18 years, 11 months
Re: gnome-panel eats a lot of memory
by Damian Christey
> Some how it seems to be related to the text in the window title or
> something in that area. When i open Evince from the menu nothing happens
> to the memory usage of gnome-panel. When i now open a PDF file with that
> (empty) evince the memory usage of gnome-panel jumps up. If it open a
> PDF from nautilus evince opens with the PDF and the memory usage of
> gnome-panel jumps up.
>
> the same thing seems to be the case with archive manager, when opened
> from the menu nothing happens to gnome-panel memory usages, but when
> double clicking a tar file in nautilus or open a new tar file, the
> memory usage of gnome-panel jumps up.
>
> On the other hand, opening a gnome-terminal and cd'ing around (that also
> sets the window title) does nothing to the memory usage of gnome-panel.
>
Sounds to me like it has more to do with opening files with Gnome
programs than opening windows in general, which says to me
"gnome-recent", which is used by the panel. How much of a leak are we
talking about? More than would be expected for the menu item that it
adds for each opened file?
Bug number?
--
Damian Christey <christey(a)csee.wvu.edu>
18 years, 11 months
Fw: "I plan to work on..." for FC5
by Dimi Paun
From: "David Woodhouse" <dwmw2(a)infradead.org>
> I don't want it to lock whenever I walk away from it; I want it to lock
> only at {boot,resume} time if my phone isn't present _then_.
That's cool, but another very interesting usecase would be to
lock/unlock automagically when you walk away/to the box.
Talking for myself, I'd certainly use such a feature where I
currently work, simply because there have been problems in
the past with unlocked computers. And I admit, I forget
sometimes to lock mine when I go for a coffee. Such a feature
would be more then a simple convenience for me.
--
Dimi Paun <dimi(a)lattica.com>
Lattica, Inc.
18 years, 11 months
[PATCH] mkinitrd rescue mode
by Jeff Layton
I've opened a BZ ticket with a patch to add a rescue mode to the current
devel version of mkinitrd, but wanted to post a message about it here
since I know this is where most of the discussion occurs. The BZ ticket
with the patch is here:
https://bugzilla.redhat.com/bugzilla/show_bug.cgi?id=159287
Essentially this patch adds some functionality to mkinitrd and nash.
This adds some tools to the initrd images (busybox and fsck, in
particular), and adds a function to nash to allow running a shell before
the switchroot occurs.
This gives you some ability to troubleshoot booting problems, and gives
the ability to do some rescue-type work without having to boot to the
CD. This is a big plus for people that run boxes remotely and don't have
easy physical access to them.
The patch needs a little work,but a backported version basically worked
for me on RHEL4. I have not yet tested it on Fedora but it should
basically work there.
Essentially, I'm sending this to solicit some feedback on it. Anyone
have comments or concerns with this idea?
--
Jeffrey Layton <jlayton(a)redhat.com>
18 years, 11 months
SEP bit disabled in FC
by Jeffrey Buell
In arch/i386/kernel/cpu/common.c:
/* hack: disable SEP for non-NX cpus; SEP breaks Execshield. */
#ifdef CONFIG_HIGHMEM64G
if (!test_bit(X86_FEATURE_NX, c->x86_capability))
#endif
clear_bit(X86_FEATURE_SEP, c->x86_capability);
So, in order to enable Execshield, the SEP cpu bit (sysenter/sysexit) has to
be turned off. But this costs a lot of performance: as much as 2.5X in
syscall-heavy benchmarks (e.g., process tests in lmbench).
How permanent is this hack? Will Execshield be fixed (or removed) by FC5?
Ever? Note that SEP is enabled in SuSE 9.3, for instance.
Jeff
18 years, 11 months
Re: "I plan to work on..." for FC5
by David Woodhouse
On Fri, 2005-06-10 at 12:26 -0400, David Zeuthen wrote:
> Well, it needs to do that every time you want to unlock the screen
> saver so I don't really see a need to special case it?
The idea is that if I boot or resume the machine while my phone is
nearby, the screen shouldn't become locked in the first place.
I don't want it to lock whenever I walk away from it; I want it to lock
only at {boot,resume} time if my phone isn't present _then_.
--
dwmw2
18 years, 11 months
Re: OT: nVidia driver [was: Wish list]
by Horst H. von Brand
Denis Leroy <denis(a)poolshark.org> wrote:
> Arjan van de Ven wrote:
> > There probably are patents involved, but those don't in general prevent
> > open sourcing something unless you're on the wrong end of the stick.
> Yup that's likely, and if they opened the code, we'd still have our
> hands tied like with the MP3 code.
Maybe.
> OTOH, there's a trend lately of companies "pledging" their patents to
> the Open Source community (Nokia, CA, IBM...). Would it help if Nvidia
> did the same ?
What if nVidia just /licensed/ the patents for use with their cards, not
for /distributing source/? They may very well not even be their own. And
there might also be "trade secrets" in there.
OTOH, a friend of mine got to write a driver under NDA for the specs of the
device and protocols to talk to it... which he claims aren't all that
earth-shaking. But they came in form of photocopies of hastily printed
pages with handwritten amendments and additions. His take is that they'd be
ashamed to show the specs ;-)
--
Dr. Horst H. von Brand User #22616 counter.li.org
Departamento de Informatica Fono: +56 32 654431
Universidad Tecnica Federico Santa Maria +56 32 654239
Casilla 110-V, Valparaiso, Chile Fax: +56 32 797513
18 years, 11 months
Re: "I plan to work on..." for FC5
by David Woodhouse
On Fri, 2005-06-10 at 10:44 -0400, David Zeuthen wrote:
> Sounds good, however, we should just lock the screen when the machine
> is suspended so it is locked when we resume.
It still needs to check for the presence of my mobile phone at _resume_
time, and unlock if it's found, surely?
--
dwmw2
18 years, 11 months
