[LWN Logo]
[LWN.net]

Sections:
 Main page
 Linux in the news
 Security
 Kernel
 Distributions
 Ports
 Development
 Commerce
 Announcements
 Back page
All in one big page

See also: last week's Back page page.

Linux links of the week


The Linux Internet Server Administration Guideis a cooperative effort which is developing a comprehensive manual for the administration of a Linux machine on the network. They are still a bit short on actual content, though they have a nice table of contents. This looks like a good project.

Alan Cox's linux.org.uk site serves as his soapbox in front of the Linux community. It includes some basic useful resources, including his y2k page, and also has an editorial by Alan on whatever's on his mind (currently an Atlanta Linux Showcase summary and the Halloween memo). Plus, of course, a link to his ever-interesting diary page.


November 5, 1998

   

 

Letters to the editor


Letters to the editor should be sent to editor@lwn.net. Preference will be given to letters which are short, to the point, and well written. If you want your email address "anti-spammed" in some way please be sure to let us know. We do not have a policy against anonymous letters, but we will be reluctant to include them.

Of course, we should have known better than to solicit more letters to the editor; this week we have been buried in them. There is no way that we can include even a significant fraction of them. Our apologies to those whose letters were left out. Please write us again when the urge hits.

 
   
Date: Fri, 30 Oct 1998 09:40:07 -0500 (EST)
From: Jonathan C Day  <j.c.day@larc.nasa.gov>
To: editor@lwn.net
Subject: Some thoughts on the Mexican project

Dear LWN,

  I was thinking a bit about the Mexican project to install Linux in
140,000 centers. If they installed and configured some of the distributed
processing toolkits (eg: PVM, MPI), I know there are programs for
administrating groups of machines in one go, using such toolkits. This
would make life a lot simpler for whoever will be administrating these
machines.
  Purely incidently, (of course! :), by doing so, they would have set up
the world's largest Beowulf and (depending on how the speed calculations
are done) possibly the worlds fastest supercomputer. Whilst, in practical
terms, these would offer no direct advantages, it might help if they
wanted or needed corporate involvement or sponsorship. I don't know what
morale in Mexican schools is like, but I know in many countries including
the US and UK they aren't known for their self-esteem or self-image.
Getting into the world records might help a little there.

Jonathan Day


   
Date: Thu, 29 Oct 1998 22:09:48 +0200
From: Amos Shapira <amos@gezernet.co.il>
To: editor@lwn.net
Subject: Mexican announcement is a call for arms!

Hello,

I too was very glad to hear about the new Mexican schools project,
but there is a risk here too - the eyes of many other educational
and corporate bodies are now watching how this will develop, any
negative aspect found (or percieved) during this "experiment" might
make Linux look very bad and bloated by its opponents to discredit
it.

I'm not saying this to discourage the project, on the contrary, I'm
saying this in order to *encourage* every person who cares about
Linux to do whatever they can to help make this project succeed.

You should not look at this announement as a conclusion of whatever
been done so far, but rather as an announcement about a certain goal
to be achieved - namely to equip Mexican schools with Linux machines,
make them love it, and above all - show the world it's possible.

This is a call for arms much more than a pat on the back.

Cheers,

--Amos

-- 
--Amos Shapira                  | "Of course Australia was marked for
                                |  glory, for its people had been chosen
amos@gezernet.co.il             |  by the finest judges in England."
                                |                         -- Anonymous
   
Date: Thu, 29 Oct 1998 17:21:24 -0500 (EST)
From: Clemmitt Sigler <siglercm@alphamb2.phys.vt.edu>
To: editor@lwn.net
Subject: You asked for it... ;^)

Well, I thought I'd chip in my $0.02 worth on the current state of
Linux since you asked for letters to the editor.  You'll probably be
*swamped* now....

It's my opinion that, since Linus has worked so hard and diligently
and greatly improved SMP support, we need to turn our attention to three
other critical hardware subsystems.  My bet is that these will help make
or break wider scale acceptance of Linux in the 2.2/2.3 phase.  They are:

1.) USB support;
2.) FireWire (IEEE 1394) support, and;
3.) I2O support.

We now know that with Intel coming on board with Red Hat in a modest
way, I2O specs may come to Linux through, for lack of better terms,
"diplomatic" or "political" channels.  And, since it's a closed
standard there's not much an open development-model project like Linux
can do about it any way.

However, USB is starting to appear (I have a Kodak DVC300 web cam that's
USB that I bought in December of 1997 -- I wish it would work under
Linux :^), and with the spectacular performace Fire Wire can offer
its implementation is only a matter of time.  To date, there's no
non-alpha, non-beta support for either of these peripheral busses in
Linux.

I can hear proponents of NT/Windows 2000 now, dismissing Linux out of
hand because it has no USB or FireWire support.  We've already seen this
type of press starting to appear just in the last week.  We can argue
about how important these will be to current Linux users, that's for
sure, but to potential future users?  If we want Linux to break into the
desktop and home-user markets in a larger way (eventually), items like
this *have* to be addressed, IMHO.

The USB and FireWire project pages are located at, respectively:

http://peloncho.fis.ucm.es/~inaky/USB/

http://www.edu.uni-klu.ac.at/~epirker/ieee1394.html

Please, if you can, give these projects some help, even if it's only
by alpha- or beta-testing for them.  And it would be really sweet if a
sponsor like Red Hat who has hard cash to spend on development (a la
hiring kernel developers and supporting GNOME) would get behind these
efforts.  They could probably get these systems solidly supported even
before Windows 2000 is publicly released if they start now!

Thanks for letting me get this off my mind :^)

					Clemmitt Sigler
					Linux/Unix Users' Group
						of Virginia Tech

   
Date: Mon, 02 Nov 1998 00:45:30 -0500
From: Aaron Sherman <ajs@ajs.com>
To: editor@lwn.net
Subject: Letter re: recent Anti-Linux flame on ZDNet

ZDNet recently published a rather rabid anti-Linux rant, from a company
called "The Butler Group". I've composed a public response, which is open
for others to copy or quote. It's at http://kr.com/~ajs/linuxresponse.html

-- 
Aaron Sherman			Safety Net Solutions
ajs@ajs.com			Senior Perl Snob (and bottle-washer)
www.ajs.com/~ajs		finger ajskey@lorien.ajs.com
   
Date: Mon, 02 Nov 1998 10:37:49 -0800
From: "Roy P. Ammeraal" <ramme@concentric.net>
To: editor@lwn.net
Subject: Screenshot of WP regarding Microsoft France Reaction

Folks,

That text within the screenshot of WP, running on Linux, regarding the
Microsoft France reaction, does not look very professional. I understand
that we're all very enthusiastic about Linux, but I don't see why we
should react with mud, if another (opposing) party is throwing with mud.
I think we all should be professional in our responses, even if the
other party, Microsoft in this case, is not being professional. I
suggest that this screenshot be either removed or replaced with a more
professional screenshot. Otherwise, I have been enjoying LWN very much.

Thanks,

Roy P. Ammeraal

   
Date: Wed, 4 Nov 1998 17:09:20 +0100 (MET)
From: Maurizio de Cecco <Maurizio.de.Cecco@ircam.fr>
To: editor@lwn.net
Subject: Why Microsoft forget about commercial Applications ?


In the Hallowen report there is no single word about commercial
application running on Linux.

This make the whole competitive analysis of Linux vs NT or vs Windows
98 useless, beacause based on the restrict view of the OSS reality.

On the server side this is already happening with the Data Base editors:
they realized that Linux can be a market for a more open competition, and
now the Linux choice is not a pure OSS choice anymore.

It is happening on the desktop; Linux already have the Applix suite,
the Corel Office Suite will follow, and may be Lotus stuff also.

Software editors realize that Linux can be a market for a commercial
competition, a clean start over the current monopole situation.

I wonder why this threat, a lot more real and incombent that a world
with only OSS solutions, have be completely ignored in the report.

Maurizio


-- 
Maurizio De Cecco                 Real Time System Team
				  IRCAM, Centre Georges Pompidou
                                  1, Place Stravinsky 75004 Paris, France

tel:   +33-(0)1-44784779  - fax:   +33-(0)1-44781540 - email: dececco@ircam.fr
   
Date: Wed, 04 Nov 1998 10:03:50 +1030
From: Stephen Donaldson <Stephen.Donaldson@adelaide.on.net>
Subject: Tied to the Monster
To: editor@lwn.net

Its not easy to finally bite the bullet and make my presence known to
the linux world at large, however, needs must as the saying goes.

I have watched the tide of linux swell up in the media like a wave of
titanic proportions and wondered why?  Why is it easy to set up and go
go go? Can I quickly install software and get on line with little or no
hassle? The answer from 'me' is not really, YET.

I am using my NT machine and it runs well, good software and easy
connections to the net mail etc. etc.  Now please don't go jumping on to
me and telling me all sorts of things I already know....yes I have put
redhat5.1 onto my other machine, curiosity is my weakness and I feed it
constantly.  However, I don't want to spend hours and hours...and hours
getting my machine up to speed and I mean by me learning the ropes of
linux.

But I do, YES, like lots of my friends and colleagues want to give
Microsoft the heave (colloquialism for the boot) I desperately want a
machine is not microsoft dependent, has the latest software and I can
use to develop business notes on, web pages with the latest software
(easy stuff to use).

Am I expecting just a tad too much?  I hope not.  I want all that is
good morally and at the leading edge of development. I want my cake and
I want to eat it. And I want it now.  I want Microsoft to feel the heat
from good honest people like me. (smile) because that's where linux will
win, when people who know how microsoft operate and don't like it one
little bit suddenly find they don't need a year of sleepless nights
coming to grips with linux and vi (shudder) and who can easily switch
with as little pain as possible and find the final transition to be
nothing more than a couple of days readjusting to a few key strokes.

Make it easy for me guys and I'll come over, give me the good software
and I'll be there tomorrow.

Thanks for reading my rant.

(Give me a mate who knows linux and I'll buy the beer)


Stephen Donaldson
An ordinary user

   
Date: Thu, 29 Oct 1998 15:25:48 +0100
From: David Kastrup <dak@neuroinformatik.ruhr-uni-bochum.de>
To: editor@lwn.net
Subject: Recent attention for Linux


The last year or so has featured a lot of surprises from the corporate
world.  The latest being the announcement of Corel for providing a
no-cost version of Wordperfect for Linux non-commercial users.

We have a no-cost version of Netscape (and an upcoming free version of
Mozilla, which might be more important).  We have no-cost versions
restricted to non-commercial use of Staroffice, several databases and
so on.  We have quite a few promises of commercial versions of several
other software.  We have a lot of media attention, mostly due to it.
Linux is getting employed a lot for internet solutions by now.  We
have a few players not traditionally Unix-centered investing in
RedHat.  We have even the recent announcement that Mexico will go
Linux in its educational system.

What's in it for Linux?  Mostly trouble.  But a lot of chance to get
vastly faster ahead from where one is currently than anticipated, too.

The big problem is that Linux is still very much evolving.  It is not
yet ready for high-performance serving tasks (work on consistently
supporting SMP as well as better networking performance are ongoing,
but much is scheduled for 2.3 which will probably turn into a stable
version about 2 years from now).  Fortunately, this gets somewhat
obscured since what mostly counts to people is just performance as
compared to NT, a sitting duck, and there is quite lot you can make
Linux do with even a single-processor Alpha workstation (and, of
course, despite of working suboptimally, a two-processor machine under
Linux can still deliver quite a bit more serving power than a single
processor).

Yet it is already now that Linux has to encounter enormously rising
demands.  In the server market, I suppose it will grow to meet them
smoothly enough.  But the expectations with regard to userfriendliness
and application availability frankly cannot be met in a whiff.  A lot
of development will need to be done to satisfy even a fraction of the
people reacting to Linux hype.  The current calls for applications are
what has caused a few players (including the major database vendors)
to experiment with offers for Linux.  But in the long run they will
want to see whether Linux gives a successful path to revenue for them,
and that is when we might be experiencing some serious backlash, at
least with the availability of evaluation copies, if not with whole
product lines.

I'll skip commenting on the office sector which is largely
monopolized, even though Linux currently provides a small uncontested
niche for alternatives.

In the gaming sector, Linux is currently not an attractive platform.
Kernel sound support leaves much to be desired (even just compiling
the drivers in current stable kernels can be a nightmare, much less
getting them to run).  Hardware GL is supported only sketchingly, and
mostly by commercial Xservers not in wide circulation.  Good GL
implementations would be very important to have, not only because this
means support for high-performance graphics, but also because it
enables easy multi-platform programming and porting and thus lowers
the cost of side-entry into the Linux market.

So what's the gist?  In the serving section, Linux is pretty much up
to demand except under highest strain, and will probably prove mature
enough to reasonably comply with demands, or at least be able to react
to them satisfactorily and adapt properly in time.  For this reason I
believe that it will be able to provide a satisfactory platform also
for database servers, so it is reasonably probable that the current
database server excitement will not cause much trouble when abating.

To the average user, expectations will not be met soon.  It is only a
question of time until people trying it out will come to the
conclusion: "yes, it does not crash, but it really does nothing else
worth noting, too".  This could be changed if application and game
vendors were to join the quest.  They just might, if not the
environment was as dissimilar to Windows as it is, making the porting
quite a lot of work, and if it was not yet as immature (no established
desktop up to now, no serious GL support...).  Yet at least there have
been encouraging ports by large players, such as Corel and Netscape,
and lots of announcements.

Will Linux be able to ride the crest?  A lot of people are placing
their bets on Linux, and Linux is not yet ready to deliver.  It is
crucial for the long-term success of Linux that it *will* deliver
before people get turned off.  I would have said that we could stand
vendors currently trying out the Linux market eventually turning their
back again, and we could at least pick up the truly free scraps that
happen to fall from their tables sometimes (like the things RedHat
develops with an interesting set of sponsors, like the Mozilla
browser, like a set of developers getting familiar with Linux and
quite a few other things).  Even if the tables get cleared one day,
these things will remain with us.  But what we cannot afford is having
something like the project for equipping Mexican schools and high
schools fail.  Having a whole country's educational system hedge its
bets on Linux and failing is press Linux will not be able to put aside
anytime soon.  And you can bet on it that a lot of people are eagerly
waiting just for this to happen.  Currently Microsoft can still say
"oh, just try out Linux and see where this will lead you" for scaring
people.  Now that somebody indeed seriously is going to try just that
on a large scale, we better take care that they'll end up better than
anybody anticipated.  The pioneering is not yet finished.  Having this
succeed will set an example for free software systems that will
literally change the world.  It will also create markets and
infrastructures that will make Linux a major platform for developing
for.

But for this to happen, the usability aspect (including word
processors etc.) has to be tackled much more vigorously than before.
In addition, hosts of *new* developers need to be supported.  This
means that both online help for applications and programmer's
documentation for GUI programming must be forthcoming.  Right now,
both are sadly lacking almost anywhere.  Just try to press a "help"
button in some GNOME application.

Can we take the heat?  We better should.  And we'll need all help we
can muster, including what parts of the software industry one can
persuade to join the ranks.  And we'll need a lot of work in order to
catch up with free offerings eventually for what may be temporarily
available only proprietarily, if we want to finally reach that old
dream that started the GNU project decades ago: the availability of a
completely free system for the generic needs of computing.

David Kastrup                                     Phone: +49-234-700-5570
Email: dak@neuroinformatik.ruhr-uni-bochum.de       Fax: +49-234-709-4209
Institut für Neuroinformatik, Universitätsstr. 150, 44780 Bochum, Germany

   
From: "maskatron" <maskatron@curvedspace.org>
To: <editor@lwn.net>
Subject: article submission
Date: Wed, 4 Nov 1998 20:05:01 -0500

Hello,
I'd like to submit this editorial about Linux for possible inclusion in your
magazine (which I enjoy very much; thanks for doing it).  It is called
Politics, Money and Linux, as it discusses the recent media success of Linux
& where I see things going in the next couple of years. I hope you enjoy it.

thanks
Shayne White
http://curvedspace.org/html/shayne.html


Politics, Money and Linux

Let me just preface this article by explaining that these views are entirely
my own.  They are based on my distorted view of the world. If you disagree
with me, take a deep breath, send me an email and we can discuss it.  I'm
not out to start a flame war with anyone. What do I see when I look into my
crystal ball and utter the word "linux".  Well, it looks something like
this...

I see some very chaotic times ahead.  There are many forces at work in the
rapidly growing linux community. I see more and more developers working
together, sharing ideas and sharing code. I think this is great.  It never
fails to amaze me what can be accomplished when people get excited about
something and work together to make it happen.  Linux is the poster child
for open source, and it should be.  It keeps getting more and more press.
It seems to be the newest media darling.  The up and coming OS.  The
greatest threat to Microsoft's dominance.  This is all well and good, but
what people that are new to Linux don't understand is that it is much more
than a piece of software.  It seems to have a character all it's own.  And
everything is just great.  The plan of world domination is right on
schedule.  Or is it?  Personally, I don't think things are quite that rosy.

Because of it's success, Linux as we know it is headed for trouble. Now hold
on...before I get labeled as a non-believer, listen to my reasoning.  Linux
will become wildly successful, but during this trip to the top, it will
change.  Thus, it won't be the Linux we know and love. The Linux community
is currently made up of developers who do this as a hobby in their spare
time. They do it because they love computing.  They do it because they want
to give back to the community of developers that has given them so much
enjoyment. Most don't get paid for it, and that is fine with them. For some,
they actually prefer it that way.

But along comes politics and money. In the current model, Linux is truly
free.  From both an economic and political standpoint. Anyone who is
reasonably technically proficient can download and install Linux. For free.
It is written for developers by developers.  Anyone is free to make changes
and generally these changes get thrown back into the pot.  This system works
and works quite well. It has resulted in perhaps the best OS in the world.
So what the fuck am I talking about?  Well, many people are excited about
some of the big companies jumping on the Linux bandwagon.  They see this as
a good thing for the Linux community. It makes it legitimate.  They see it
as big companies who are changing their ways and are getting into open
source and free
software.  Well, I think these companies are merely latching onto something
that can enable them to compete with Microsoft.  These companys exist to
make money for themselves and their shareholders.  They have a very
different agenda than the average Linux developer.  Today they are embracing
the Linux community, but it is a community that they don't really
understand, nor will they ever fit into. Today it is a political move.
Tommorrow it will be an economical one.  That's when the problems will
start.  Personally, I don't think developers will be as excited about coding
up feature X for Linux so that XYZ Corp can make money off 'supporting' it.
And no, I'm not talking about Redhat. They understand the Linux community.
Thus, XYZ will support their own version.  What will happen?  My guess is
that Linux will split into different groups.  XYZ will lag behind in
features once the politics of a big company get in the way and hamper
progress. The mediocre masses will use the XYZ version, while technical
users will stick with a more grassroots version, like today's Debian.  Like
so many things, the whole thing will come full circle.

This is not an article against commercial development for Linux.  I believe
that people
should be compensated for their work.  I just think that Linux does not fit
the mold of a typical business software product, and when push comes to
shove, the 'real' Linux will go back 'underground'.  Things have already
started in this direction.  Just look at the UDI project by Intel....

-Shayne White



 

 

 
Eklektix, Inc. Linux powered! Copyright © 1998 Eklektix, Inc., all rights reserved
Linux ® is a registered trademark of Linus Torvalds