Re: [aclug-L] Microsoft's plans to kill Linux

[John Goerzen <jgoerzen@xxxxxxxxxxxx>]

Well, as one person that's been a member of the free-software community for
some time, perhaps I can shed a bit of light on the matter.

On Wed, Nov 11, 1998 at 01:39:23PM -0600, Karl Juhnke wrote:

> I read the Halloween documents, and I hear there is no longer any debate
> about their authenticity.  What I don't understand is mixed responses I
> have read from the Linux community.  On the one hand, Microsoft is
> portrayed as not only evil, but also dangerous.  They are the Borg, the
> nightmare bogey man, and we should all tremble in fear.  On the other

Microsoft represents all that is wrong with proprietary software.  Their
operating systems, for instance, have a long history of being low-quality
and literally decades behind the competition in many areas.  Their current
dominance was accomplished via a serious of underhanded (to be nice)
dealings, slanted vendor contracts, and aggressive marketing that was often
untrue.

As a small highlight of the latter is the funny but true episode where
Microsoft's arrogant product managers tried to say their own "Unix-like
shell" was completely compatible with ksh.  (Included at the end; a good
read!)

These memos did something important.  They brought out in the open proof
that Microsoft is using these sorts of tactics to try to commandeer
everything good about the Internet.  A lot of people in the free software
commuity were responsible for the creation and evolution of large parts of
the Internet into what it is today (the mail system, for instance.)  To see
Microsoft try to take it over with proprietary protocols that are inferior
could literally be a problem for the world.

> hand, Linux advocates can't seem to bring themselves to concede that the
> Linux project is in any way vulnerable.

This is also correct.  Linux has never in any way been effected by
Microsoft.  The fact is that there really isn't much of anything Microsoft
can do to stop Linux.  Things that would stop a commercial company can never
stop Linux.

For instance, Microsoft likes to just buy out their competitors and then
kill their products.  Even if Microsoft somehow managed to hire Linus, this
doesn't kill Linux.  There are literally millions of copies of the Linux
source out there, and thousands of active developers that would make
competant replacements should the need arise.

The point is that Microsoft's tactics for combating Linux will not, and can
not work.  The reaons is simple: no matter what happens, the free software
movement has grown to such a point that they can never even come close to
having the same number of brain cells focused on Windows as free software
has focused on its OSs.  Their idea of taking over Internet protocols cannot
work either.  Most of the Internet servers are not running Microsoft OSs,
regardless of what they would like people to believe, and nobody wants to
run software that cuts them off from half the Internet.  And, in an
interesting recent blow, Microsoft was talking about an HTTP extension in
one of their documents that they said the free software community would
never have an implementation of because of its sheer complexity.  Less than
a week later, the first version of the appropriate extension to Apache was
announced.

> It seems ironic that Linux folks accuse Microsoft of being short-sighted
> and self-congratulatory at the same time that they reassure each other of
> the invincibility of Linux.

Well, I haven't seen the self-congratulatory elements much, at least not
among the people that are Linux developers.  

> In particular, I am astonished by the reaction to Valloppillil's assertion
> that "Linux is unlikely to be a threat on the desktop" because "Ease of
> use must be engineered from the ground up.  Linux's hacker orientation
> will never provide the ease-of-use requirements of the average desktop
> user."

He made a fundamental mistake here.  I know people are sick of me pointing
this out, but I'm going to continue doing so anyway.  "Ease of use" is *not*
the same as "ease of learning".  Linux has, and has had for some time,
better "ease of use" than Windows.  That is, once you know how, it's easier
to use Linux than it is to use Windows.  What he really means is "ease of
learning", which Linux probably is behind on for the "average desktop user".

> This assertion strikes me as very plausible.  I have never met a hacker
> who was at all sympathetic with ordinary users.  Ease of use in Linux

Now you have. :-)

<RANT MODE ON>
Part of the problem, I think, is a trend not limited to computers.  Here I
mean the idea that you should not have to learn anything about what it is
that you are doing or using.  For instance, I point to the popularity of the
"point 'n' shoot" 35mm cameras.  With only a few (1-3) hours spent learning,
you can take far better pictures with a SLR manual or semi-manual (automatic
assist for some thing) camera than you ever could with a point and shoot
type.  You save yourself time later on, when you discover that your "easy to
learn" point 'n' shoot camera got the faces of everyone in your family
reunion too dark because they were standing in front of a window, or got
everybody out of focus because it happened to focus on the water tower half
a mile away instead of your relatives.
<rant mode off>

Seriously though, here's how the hacker culture feels about this.  We *like*
new people to join.  Every hacker was a newbie once, too, and each new
person that uses an open source operating system is somebody that could be a
developer for that OS sometime down the road.

However, it is a matter of priorities.  If you think about it, naturally an
open-source operating system will evolve the features that the hackers like
before it evolves the features that the "average desktop user" likes. 
Otherwise, the hackers would have no interest in it, development would
stagnate, and it would never get off the ground.  However, Linux is now
large enough that it's got all the features that hackers like, and so people
are adding features that others like, too.  Examples are KDE, Gnome,
Enlightenment, GIMP, xwave, WindowMaker, etc.   So you see, it is not that
hackers aren't sympathetic with "ordinary users"; rather it's that hackers
tend to write the tools that facilitate further development first.

> means what is easiest for hackers to use.  Why should we expect that to
> change?

Well, I think that the change has already taken place (see above examples)

> And if Microsoft continues to dominate the desktop, won't they always be a
> titanic force?  Won't desktop market share continue to be a club with they
> can beat up on Linux and all other competitors?

Microsoft has dominated the desktop since before the first line of Linux
code was written.  This didn't hurt Linux then and it's not going to hurt
Linux now.  I don't think Microsoft is going to go away overnight.  But
eventually, something Open Source -- probably Linux, but not necessarily --
is going to overtake them.  It will happen, and there's nothing they can do
to prevent it.  Except going open-source themselves.

Linux has never been about market share.  This is one beauty about
open-source software.  You aren't accountable to marketing.  You can do
what's *right* instead of something that makes it more popular (ie,
improving performance of the filesystem instead of adding animated
paperclips to word processors.)  The changes may not be readily noticeable
by people (except that their computer seems to run faster), but the people
that really understand what is going on will appreciate it.  And in the end,
even though these people are not necessarily the ones with the final say-so
in purchasing, they will (and are!) be heard..

> I think Linux fans should stop spraining their shoulders patting
> themselves on the back, and learn from two things which Microsoft does
> right.  1. Make stuff easy to use for average schmoes, and 2. Stay
> paranoid.

Well, I don't think either of those things have happened.  I think you'll
see patting on the back when kernel 2.2 comes out, and when Debian 2.1 comes
out.

But Linux has never consciously cared about what Microsoft is up to. 
Microsoft is, of course, viewed is an "evil" by most Linux developers.  But
most Linux developers concentrate on Linux, and don't really care about
Microsoft.  Linux is not aimed to compete with Microsoft; it's aim is to be
the best OS it can on its own right.  This generally means it is better than
Microsoft, sure.  And nobody is going to shed any tears when Microsoft
eventually self-destructs or something.  But it is not Linux's goal to go
after them.

Linux's (or rather, Open Source's) goal is to go after the entire
proprietary software system.  Not just Microsoft, but also the Netscapes and
Oracles of the world.  Of course, Netscape has wisely seen the light and
decided it's better to join the Open Source movement than to try to oppose
it, and they've already reaped the benefits of doing so.  If they hadn't,
one of the open source web browsers would have gotten to the point where it
would have been better than Netscape.  At that point, Netscape would have
been mostly killed.

Open Source OSs have already surpassed Microsoft's offerings, and you can
see that it's already started to hurt them by the concern evident in the
latest memos.  Microsoft will kick and scream for a lot longer than Netscape
would ever have been able to, but eventually they too will have to either
join the Open Source movement or by made obsolete by it.

Oracle is an interesting case.  They have a port of Oracle to Linux, but
without source code.  Of course, what this means is that for now, they make
a lot more money because a lot of Linux people are running Oracle.  But
eventually, somebody is going to write a database that's Open Source that
can surpass Oracle.  If you don't believe the community will ever come up
with something that complex, just consider this: we've already got the
entire operating system, complete with modern networkable GUI that's years
ahead of Windows, full-fledged development tools for dozens of languages, a
complete suite of Internet clients and servers, etc.  The most impressive,
of course, is the OS.  If the Open Source people can beat Microsoft at it's
most profitable game, then surely the same will happen for other programs
too.  Ideally, Oracle will eventually turn Open Source too.  What a way for
them to compete against MS SQL Server!  If they turn Open Source, they
instantly get a foothold in universities across the country, and more and
more people learn Oracle, and they'll get more and more money for support. 
Not only that, but the very people that use Oracle (and thus know the most
about its benefits and shortfalls) are ones that can help improve it
directly.  Netscape has already used this strategy, albeit after basically
being forced to, and it's already worked for them.

John

-- 
John Goerzen   Linux, Unix consulting & programming   jgoerzen@xxxxxxxxxxxx |
Developer, Debian GNU/Linux (Free powerful OS upgrade)       www.debian.org |
----------------------------------------------------------------------------+
Visit the Air Capital Linux Users Group on the web at http://www.aclug.org
---
This is the Air Capitol Linux Users Group discussion list.  If you
want to unsubscribe, send the word "unsubscribe" to
aclug-L-request@xxxxxxxxxxxx.  If you want to post to the list, send your
message to aclug-L@xxxxxxxxxxxx.