[LWN Logo]

 Main page
 On the Desktop
 Linux in the news
 Linux History
All in one big page

See also: last week's Letters page.

Letters to the editor

Letters to the editor should be sent to letters@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.

March 15, 2001

From: "Kapil H. Paranjape" (user: kapil host: imsc.ernet.in)
Date: Mon, 12 Mar 2001 12:12:44 +0530
To: letters@lwn.net, greve@gnu.org
Subject: Does free mean "free from" or "free to"?


While there is a lot of discussion on the "free beer" versus "free
speech" distinction along with the emphasis on free as in "freedom",
there is perhaps not enough said about the *duties* that go with
freedom. In High School civics (in India) the fundamental rights of
people were always tied to the fundamental duties---there's that
uncomfortable "d" word again.

One of the duties that goes with free (mukta) software is that of climbing
the learning curve. Users should be continually encouraged to step beyond
the boundaries of their current knowledge---not sit tight in beautifully
designed boxes of their favourite desktop/window manager/user interface. I
have yet to see an interface that encourages such exploration the way
(for example) adventure and nethack do at the level of games.

When we emphasise the development of software that makes computers easy
to use we should also beware that this should not lead to the "computer
is a toaster" analogy that many proprietary vendors like to promote.
Let us not bring more users into the free software fold by telling them
that life is easier here--rather that there are more opportunities to
use ones' abilities to the creative fullest.

Unfortunately, as long there will be people willing to "take the easy
way", there will also be a MicroSoft that leads them to a beautifully
decorated creative dead-end.


Date: Thu, 8 Mar 2001 12:18:29 +0000
To: letters@lwn.net
Subject: Mozilla Matters
From: home@alexhudson.com

RE: Luke Subert, 'Will Mozilla 1.1.1 be released in 2001?', 2/3/2001.

How Mr Subert is able to talk of the 'slow pace of Mozilla development'
defies belief. For anyone who has been using the software, the rapid
advancement (particularly noticable recently with the 0.8 release) is
obvious, and to further state that Mozilla 'will [not be ready] until
[release] 1.1.1' is bordering on ridiculous. Can I ask if Mr. Subert has
used Mozilla? It is now my primary browser, and I can say I am very happy
with it.

To further compare Mozilla to other browsers shows Mr. Subert's lack of
actual usages of these products. Mozilla compares very well to Galeon (it
should do, Galeon is based on Mozilla!), and already is more feature-rich
than either Konq or Opera (no disrespect to those browsers). Further,
Mozilla is being used in more and more projects (Galeon, as mention,
SkipStone, as well as a host of Gnome projects using Mozembed, and
countless others). Mozilla is in far wider use than Mr. Subert obviously
recognises, and is not only complying with standards, but setting them also
(witness XBL 1.0, for example).

I suggest Mr Subert downloads the current 0.8 build of Mozilla (or perhaps
wait for 0.8.1, or 0.9) and actually try it. Try building Galeon with it. I
think he would be surprised. Let's get off the Mozilla teams' back, shall
we? Try the software, it's actually really rather good :-)

Kind regards,

Alex Hudson.
From: "Matt.Wilkie" <Matt.Wilkie@gov.yk.ca>
To: editor@lwn.net
Subject: RE: Free Software has forgotten release often?
Date: Thu, 8 Mar 2001 17:47:13 -0800 

> While free software has many advantages, many projects 
> seem to have forgotten the "release often" rule. Long 
> cycles for projects like Mozilla (4 years to the 1.0 
> release) and GIMP (2 years between 1.0 and the next major 
> release of 1.2) leave ordinary users wondering just where 
> the added value of open source really lives. 


There is a new release of Mozilla every day via the nightly 
tarballs if you like the leading edge, and if you prefer a 
little more stability, more or less quarterly milestones (now
minor dot releases). I don't follow GIMP development as 
closely; I've only upgraded my installation four or five times 
in the last twelve months.

Yes I wish Mozilla was "done". Yes I wish for a couple of
new GIMP features (not that can think of many). However I'm
quite willing to trundle along with not quite perfect 
versions while the coders sweating in the smithy bang away,
folding the metal over a hundred times to build the best 
possible edge. Not that I'm above the occasional whiny "are 
we there yet?" ;-)

The proprietary softwares I make my livelihood with have
fairly regular "releases" about every 18 months or so. 
For minor programs it is standard policy to wait an extra 
6 months to a year before forking out the dough ("never 
buy a .0 version"). For our major program (ArcInfo) we 
pay the multi-thousand dollar yearly support fees and get
the latest and greatest. However we've been using the last
major version for about a year, and we -still- haven't
rolled it out into production use because it's too slow 
and crash prone. In my grumpy opinion it's still a beta
version (feature complete but not stable or optimised) and 
should be labeled as such.

So, version numbers can be a useful tool when wielded 
appropriately. Unfortunately too many major players in the
industry play fast and loose with them so they don't really
mean much anymore, if they ever did.


These are my own opinions and do not necessarily represent
my employer.
From: Anton Ertl <anton@a0.complang.tuwien.ac.at>
Subject: What is a Linux Distribution?
To: letters@lwn.net
Date: Sun, 11 Mar 2001 10:34:57 +0100 (MET)

To me a "Linux" system feels quite different from a plain NetBSD
system (as well as out-of-the-box proprietary Unices), mainly because
a typical "Linux" distribution is a GNU distribution and thus offers
many practical commands and command features that non-GNU systems

And I use these features quite frequently, as I noticed when I worked
on a NetBSD system: I had problems on most commands I entered, because
I used some GNUicism (starting with bash features).

OTOH, a proprietary Unix with lots of GNU packages installed feels
much more familiar.

Strangely, even tomsrtbt, which replaces many of the GNU tools with
smaller programs, feels much more GNU/Linux-ish than other Unices do

- anton
To: letters@lwn.net
Subject: Debian GNU/Hurd is a ``Linux distribution?''
From: Gordon Matzigkeit <gord@fig.org>
Date: 08 Mar 2001 10:37:56 -0600


   However, we would not call NetBSD a "Linux distribution". Why not?
   Well, to start, the NetBSD folks might very well get offended,
   since they've been around a lot longer than Linux.

You might also find some GNU folks who would be offended at having GNU
called a ``Linux distribution,'' since they've been around a lot longer
than Linux.

Debian GNU/Hurd is free applications (many of which are officially
GNU), running on the GNU C Library, on the GNU Hurd kernel servers, on
a GNU variant of the Mach microkernel.

That makes GNU at the very least two out of three of the core system
components you described.  People once argued that you could call
something a ``Linux system'' no matter what its makeup, so long as it
ran the Linux kernel.  Now, it seems, the argument is that you can
call it a ``Linux system'' if it shares any applications at all in
common with systems that run Linux kernels (so long as it doesn't have
BSD heritage).  Rationale?

To me, it's a problem of credit, and the unfortunate thing is that
it's probably impossible to reconcile ``Linux camp'' and ``GNU camp''
notions of where credit should lie.  This issue doesn't come up in
proprietary systems, because there's a single vendor who owns the
system, and says what the system is called.

I see the GNU/Linux, and GNU/Hurd designations as a compromise to give
both idealists and kernel hackers credit.  Of course, there could be
other groups who feel underrepresented, but they haven't been as vocal
as the kernelists.  Until I see a better naming proposal, that's the
one I choose to use.

BTW, thanks for your excellent coverage of Debian GNU/Hurd... it's a
high compliment when what I read in LWN makes the issues clearer than
being subscribed to the mailing lists.

 Gordon Matzigkeit <gord@fig.org>  //\ I'm a FIG (http://fig.org/)
Committed to freedom and diversity \// I use GNU (http://fig.org/gnu/)
Subject: What about Domino?
Date: Thu, 8 Mar 2001 16:15:23 -0500
From: "Sightler, Tom" <tsightler@zeusinc.com>
To: <letters@lwn.net>

I'm writing to point out what I believe to be an inaccuracy in the
following statement on the Main Page of your March 8th, 2001 issue.  You
make the following statement:

'OpenMail is the only "enterprise ready," Exchange-compatible mail
server product which is available on Linux. Its demise leaves an
important corporate function with no Linux-based solution; all that's
left is windows-based, proprietary systems - and not very many of

Unless there's been an anouncement that I've missed, Lotus Domino still
fully supports running on Linux, as well as quite a few other platforms
as well.  They also support Microsoft Outlook 98/2000 with their iNotes
Access for Microsoft Outlook product.  I certainly think Domino
qualifies as "enterprise ready" and offers many features that Exchange
does not have.  Domino supports all major internet standards such as
SMTP, POP, IMAP, and LDAP so I don't really see how this makes them a
worse choice than OpenMail (maybe you just liked them because they used

Some have critisized Lotus for supporting their server platform on Linux
without providing a Linux client, but they do at least offer
step-by-step instuction on setting up the Win32 client to run under
Linux by using Wine, which is more than many companies do, and of course
you can use any standard POP or IMAP client for basic email services.

I did like OpenMail very much, but I think Domino on Linux still means
that Linux has can server this very critical corporate function.

Tom Sightler

To: letters@lwn.net
Subject: Source code
From: Ketil Malde <ketil@ii.uib.no>
Date: 09 Mar 2001 08:34:45 +0100


In Leon Brooks' letter concerning the DVD-CCA's threats against
Dr. Touretzky the issue of whether source code can be seen as a
"device" - more specifically, a "circumvention device" - is raised.

I think this issue may have bearing upon more than the DeCSS case, and
that some exploration of the concepts involved could be fruitful.

Clearly, source code in high level languages exists for the benefit of
humans, not computers.  Source code is only a means to express
algorithms in a manner close enough to natural language that humans
can understand them, yet precise enough to not be ambigous.

The compiled, binary (machine) code, on the other hand, exist only so that
together with the appropriate processor and other hardware, some task
can be performed - in other words, the hardware and software form a

Or, to analogize: Source code is the floor plan, machine code is the
building.   Given the former and the right tools, you can construct
the latter, which is actual useful.

The interesting thing is that if this view is accepted, it seems
source code might be a way out of a lot of problems for free
software.  For instance, I find it hard to accept that patents can
apply to anything but the compiled program.  The source code is just
an exact description of the (patented) process involved, you need the
compiled program to actually *perform* the process.

If distributing source code is a breach of the patent, then so must
distributing the actual patent text be, since it contains the same
information!  Surely I don't have to license a patent to read it?

If I haven't seen further, it is by standing in the footprints of giants
To: lwn@lwn.net
Subject: LWN: perltidy
From: Andrew Hilborne <andrew.hilborne@uk.concentric.com>
Date: 12 Mar 2001 18:24:33 +0000


Well, I like what perltidy did to dailystrips: dailystrips is a *real* mess and 
it _appears_ that perltidy didn't break anything...

..except: it wrote an output file by default. Happily it didn't overwrite the
input file, which would have been the greatest of sins, but even this isn't
good enough.

Unix programs should obey the principle of least-surprise, and they should
behave like filters by default. This makes them easier to stitch together in
ways unthought-of by their authors, with the minimum of extra flags. (Another
good rule is that an invocation with no flags should perform the "most common
task." Eg look at pr(1).)

While I'm at it, let me note that, to paraphrase someone from an old Bell Labs
Journal, "Unix is a mute slave: it assumes you know what you're asking for and, 
so long as there are no errors when it does it, doesn't report anything
back. This lets you, the user, get on with the next task as simply as
possible. Many programs break this rule: the BSD Mail program started the
trend, and GNU bc continues it.

I'm sure you know all this stuff, but I wish those who don't would read, for
example, "The Unix Programming Environment," by Kernighan and Pike.


Andrew Hilborne
From: Eric <esr@golux.thyrsus.com>
To: letters@lwn.net
Subject: Thomas Hood's LTE
Date: Thu, 08 Mar 2001 10:46:46 -0800

Thomas Hood's letter describing the goals of the open-source and
free-software movements as "socialistic" is dangerously confused.
"Promoting a social goal" is not the same as "socialism"; the key
difference is whether cooperation is voluntary or not.

In our community, cooperation is voluntary -- we don't force anyone to
write or share software.  Under socialism, if you do not choose to
"cooperate", you will be oppressed, imprisoned, and quite possibly

The difference is important.
		<a href="http://www.tuxedo.org/~esr/">Eric S. Raymond</a>

A wise and frugal government, which shall restrain men from injuring
one another, which shall leave them otherwise free to regulate their
own pursuits of industry and improvement, and shall not take from the
mouth of labor the bread it has earned. This is the sum of good
government, and all that is necessary to close the circle of our
	-- Thomas Jefferson, in his 1801 inaugural address
From: andrew@pimlott.ne.mediaone.net (Andrew Pimlott)
Date: Thu, 8 Mar 2001 13:51:38 -0500
To: Thomas Hood <jdthoodREMOVETHIS@yahoo.co.uk>
Subject: Re: Stallman on Freedom and the American Way

Thomas Hood wrote:
> The GPL is socialistic in that it is designed to promote a social
> goal

You make with this statement both a marvelous insight and a terrific

Your insight is that Free Software is above all a social mission: it
seeks to create a community based on certain ideals (a utopia, if
you will), into which anyone may enter.  It does not (to a first
approximation) seek impose change on any individual or state, beyond
any natural outcome of voluntary choices.  Anyone who doesn't
appreciate this is encouraged to read the GNU Manifesto.

Your blunder is to confound a "social goal" with "socialism".  I am
not prepared to enter into a debate on socialism, but it is
generally considered a political movement, and so is inherently
different from a social movement.  You ascribe to socialism the
tenet, "To each according to his need; from each according to his
ability".  But this is not espoused by the Free Software movement.
Indeed, the Free Software movement is happy to give software to
anyone, independent of need; and does not pressure anyone into
contributing.  It argues that, if you write software, making it free
is the right thing to do; but this is very different.  I challenge
you to find a statement from RMS to the contrary.

The Free Software movement does have a political aspect, I will not
deny (the politial and the social can rarely be separated entirely);
but I believe it is clearly overshadowed by the social aspect (and
also by the philosophical and ethical aspects).

Date: Sat, 10 Mar 2001 18:38:00 -0500
From: Steve Waldman <swaldman@mchange.com>
To: letters@lwn.net
Subject: Thomas Hood on "Stallman on Freedom and the American Way"

In your March 8 2001 letters, Thomas Hood writes:

  I think we should be frank.  There is no point in fighting
  a war of propaganda.  There is no denying the accusation
  that one of the main aims of the free software movement is
  a socialistic one... The goal is to revolutionize the means 
  of production of software and to establish a new mode of 
  software distribution: To each according to his need; from 
  each according to his ability. 

As a software developer who releases code under the GPL, I too 
think we should be frank.

I, like may others in the free software community, am a
filthy capitalist bastard, and offer absolutely no apologies
for that. I hope and expect to derive from my work far more
wealth than I in any sense need, by virtue of fully exploiting
my own abilities.

I participate in the free software community and release code
under GPL out of a carefully calculated self-interest. Proprietary 
software licenses have created a distorted and inefficient market 
for the skills and services that I sell. In the proprietary software
world, success derives largely from incumbency and interoperability 
requirements; from the ability to define and the wherewithal to
enforce crafty licenses, contracts, and patents; from strategic
connections and marketing reach; and (as Mr. Allchin's comments 
underline) from political influence.

As a programmer, I am confident that I can compete successfully in 
fair contests with my peers based on skill. Thus, I am directly 
harmed by a situation where market power, rather than technical 
ability, determines how wealth is allocated in my industry. I am 
also harmed when, to use the tools prerequisite to my trade, I must 
cede rights over my own work to parties that may exact monopoly 
rent from me or my clients, or otherwise interfere with my full 
enjoyment of the fruits of my work.

GPLing code serves my interest in several ways. A vibrant,
copyleft free software community harms powerful incumbents who 
compete against me unfairly. Since what I sell is expertise at 
solving problems, not software that I have already been written, 
it harms me very little to make the code I write generally available. 
No matter how much free software there is, I am not concerned that
the world will run out of software problems to solve. There
would be some harm if my competitors had no-strings-attached 
access to my software tool-set while I was prevented from free
(in both senses) access to theirs. Here the GPL makes for a fair
bargain, a level playing field: you can use mine if I can use yours,
under the same terms. Lastly, but importantly, making code
widely available is excellent marketing, increasing demand for
my services and the rates I can charge.

There is no conflict between the RMS-style, copyleft free software
movement and a very deep capitalism. In fact, the free software
movement is just capitalism working well, destroying through
competition inefficiency and corruption in the marketplace. If 
Thomas Hood and Jim Allchin (strange bedfellow, I'm sure) wish 
to emphaisize analogies to socialism in the free software movement, 
they are entitled to. But the socialism they see is in their own 
eyes; it has little to do with free software itself.

        Steve Waldman

Date: 13 Mar 2001 12:16:08 -0800
To: letters@lwn.net
From: P Jones <pja@justice.com>
Subject: Stallman's Alleged Hidden Agenda

I wish to respond to Mr. Hood's false assertion that Richard Stallman has a
hidden agenda, namely socialism, behind his Free Software Foundation.

I have directly inquired of Mr. Stallman as to his position on this very
issue.  He told me pointblank that he believes in the free enterprise

I have seen Richard accused of being a socialist/communist many times on
various boards.  This is not true, and he has denied it publicly, on
Slashdot, to mention one forum.  It wasn't until I read Mr. Allchin's
remarks that I began to seriously wonder if there is an organized campaign
to misrepresent Stallman's views, as in FUD. While I cannot speak
authoritatively on this, obviously, my warning bells are going off and I
think everyone else should be on the alert as well.

I think it is important to respond and assert the truth.  No one should be
able to casually slander a man just for fun.  Not for profit either.  A
public forum isn't the same as a private dinner party conversation, where
any number of foolish things can be said, and often are, without serious
consequence. But when you publicly defame someone, as Mr. Hood did, he
ought to at least provide some proof of his opinion. He certainly did not
do so. Nor can he, because what he wrote is not founded on truth, though he
may not know it, having been himself influenced by what I now suspect is a
campaign of vilification.  Richard Stallman isn't a socialist. Period.
There is no hidden agenda.

If there is one thing that *can* be authoritatively stated it is that
Richard Stallman has *nothing* hidden. His views are well known.  He isn't
a politician.  He writes software.  He encourages the use of and openly
promotes free software, free as in speech.  Selling the software is fine
with him, and he does so himself.  What Stallman wrote in response to
Mr. Allchin's remarks on the American Way came from his heart.  That is who
he is and what he believes.  To accuse him of lying, which is in essence
what Mr. Hood wrote, is calumny, and you really ought not to have printed
Mr. Hood's letter.  It was irresponsible, in my opinion.  Yours is not an
open forum, like Slashdot's, where all kinds of idiotic comments can be
posted (and are).  Because you have editorial oversight of what is posted
on your site, you are responsible for what is posted. I hope you will find
a way to correct the error and that you will be more careful in the future.
A man's reputation is a very precious thing.

P. Jones <pja@nospamjustice.com> 

FindLaw - Free Case Law, Jobs, Library, Community
Get your FREE @JUSTICE.COM email!


Eklektix, Inc. Linux powered! Copyright © 2001 Eklektix, Inc., all rights reserved
Linux ® is a registered trademark of Linus Torvalds