Hi Mike,

On Fri, Feb 5, 2010 at 10:47 AM, Mike Miller
<mbmiller+l at gmail.com<mbmiller%2Bl at gmail.com>
> wrote:

> On Thu, 4 Feb 2010, Robert Nesius wrote:
>
> > Re: Collusion between Intel and Microsoft  to somehow harm Linux *
> > Having just finished a twelve year tour at Intel where I maintained a
> > distribution of Open Source tools across multiple Linux distros, and
> before
> > that, multiple flavors of vendor-supported *nix, I can tell you this
> isn't
> > true.  If you think Microsoft and Intel get together for group love fests
> > and anti-linux strategy sessions, think again.  Microsoft and Intel are
> > strategic partners.  They will work with each other to when necessary to
> fix
> > problems.  Strategic business partners do that in every industry.
> Beyond
> > that Intel doesn't give a damn how many PCs are running Windows, and how
> > many are running Linux.
>
> Beyond that?  Yes, they don't care about Linux and they do care about
> themselves.  To the extent that Linux is in the way of Microsoft,
> Microsoft cares about it and will push Intel to do things that help
> Microsoft.


I don't disagree with any of that.

But before going into more detail, with all due respect you're using
point-examples to draw broader conclusions that may or may not be true.


> It isn't so much about Linux as about how Microsoft and Intel
> work together -- two monopolies illegally colluding in their
> anti-competitive practices to promote their products.
>

Microsoft is a documented Monopoly.  Intel is an effective Monopoly (they
haven't been judged as such by a court yet, to my knowledge).


> Intel has to do what Microsoft wants


How do you mean that?  That Intel is at the mercy of Microsoft and has no
choice?  Or that Intel has to take Microsoft seriously?


> because if Intel produces a chip that doesn't run
> Windows, they are screwed.
>

/me points to the Itanium, Atom, and ARM (before they sold ARM off again).


Whether an Intel chip will run Windows is Microsoft's decision.
>

Not true.

The degree to which iterations of the x86 family of microprocessors are
backwards compatible is ultimately  Intel's decision.  But the decision is
made for them in essence by the market (the demand for backwards
compatibility).  This isn't just driven by windows, but by every x86 object
file compiled over the past 15 years or so.  By the way, this includes all
of the XCOFF's and ELF's compiled with GCC over Linux.


> There are many web pages about it:
>
> http://www.google.com/search?q=microsoft+intel+collusion
>
>
That's not proof of your argument.  I clicked through your link and scanned
the articles.  But if you'll accept google search results as proof of a
point, here's one for you. ;-)

http://www.google.com/search?q=proof+of+existence+of+god


> > Where Intel has to tread carefully is that they cannot trumpet the
> > virtues of Linux over Windows too loudly - to some degree their hands
> > are tied.
>
> Exactly.  Microsoft does favors for Intel:


What I meant is that Intel can't promote sales of chips by saying Linux is
better than Windows.  I don't believe they say the opposite either.  What
they do say is, famously during one add campaign, is that Intel processors
make your Internet faster. ;-)


>
> http://www.microsoft-watch.com/content/vista/microsoftintel_capable_collusion.html
>
>
You're using this as a proof point for a broader allegation, but I don't
think the way you paint the business relationship is accurate.  It's not
"buddy buddy".  It's more like the antagonist and protagonist in a John Woo
movie finding themselves at arms length with guns pointed at each others
heads.  Both companies have the ability to do tremendous harm to each other
- the degree to which they don't is simple game theory - the outcome would
be lose-lose.  You could argue "Proof of collusion"!  I'd say "strong
reasons to cooperate."  Weather they cross the line at some point into
collusion and criminal activity is not for me to say.  And even if they do,
I can assure you that relationship is not nearly as chummy as you allege.

Microsoft hates the x86 platform being dominated by one player.  That's why
they have supported some different micro-architectures over the years,
mainly as insurance and leverage against Intel.  None of them really took
off...  Didn't Windows 98 and perhaps XP supported a RISC architecture from
NEC (I can't remember the name of it and couldn't google it.  I do know
Intel purchased that FAB from NEC as part of a settlement with NEC, and I
think that FAB is where the chip was manufactured.)

You also specifically point to the events surrounding Vista's roll-out and
Vista-capable certification.  But I think think that scenario is a bit
misunderstood.   Microsoft wanted to set minimum standards for capability so
that people wouldn't run Vista + Aero on older hardware and say "Vista sucks
because it's slow" when really it is the hardware that is slow.   But you
can turn Aero off and run Vista just fine.  Microsoft's initial
certification based on running Aero suddenly pulled the rug out from under
one of Intel's chipsets that was in the middle of its life cycle.  Intel
said "Hey, what gives?" and Microsoft conceded the chipset can run Vista
without Aero, and can chug along with Aero passably.  Did they collude?  I
just truly don't know.  It could be a court rules they did collude, but even
if that happens I don't think that validates the broader conspiracy you seem
to be hinting at.  If anything, I think it shows Microsoft has to do what
Intel says too, which goes against your assertion earlier?


> > But in the market where Linux makes money - the enterprise computing
> > environment - Linux wins the day on its own merits.  Consider the fact
> > that Bing runs over Linux clusters before disputing that claim too
> > loudly. (Not to mention, check out the Halloween Documents.)
>
> I'll look for it.
>
>
Here is a good starting point.

http://en.wikipedia.org/wiki/Microsoft_Halloween_documents_leak

http://www.opensource-definition.org/halloween/halloween1.html

-Rob
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.mn-linux.org/pipermail/tclug-list/attachments/20100205/509c76d9/attachment.htm