See also: last week's Letters page.

Letters to the editor

Date: Wed, 14 Mar 2001 19:02:56 -0800
From: bryanh@giraffe-data.com (Bryan Henderson)
To: letters@lwn.net
Subject: What is Linux?

You bring up the issue of just what is a Linux distribution.  You 
conclude that it's fuzzy, but seem sure of one thing:  It must have a 
Linux kernel.  I can't see a classification of operating systems that
include the Linux kernel as being very useful.  In fact, 90% of the
times I read "Linux," the statement is not at all dependent on the 
system running a Linux kernel.

Replacing the Linux kernel with Hurd or the Solaris kernel makes about
as much difference in the overall system as replacing the Apache web
server or KDE desktop (or X Window System).

I don't know why Linus allowed his name "Linux" to be used to refer to
entire operating systems, while at the same time also being the name
of the kernel he distributes, but by far the most widespread use of
the name now is for the class of systems, not the kernel.  If Linus
decides to limit the use of the name to systems that include a Linux
kernel, we should respect that (and in many cases will be legally
forced to), but then we should get a new term for the general class of
systems that we know today as "Linux."  I think Stallman would donate
something with "GNU" in the name.

-- 
Bryan Henderson                                    Phone 415-505-3367
San Jose, California

Date: Tue, 20 Mar 2001 10:21:40 -0600
From: John Palkovic <palkovic@pobox.com>
To: lwn@lwn.net
Subject: re: XFree86 4.0.3 - time to dump version 3.x ?

I run Debian GNU/Linux and xfree86 on powerpc hardware. Non i386
hardware is not well-supported. Does lwn.net assume that everyone
running linux is on intel hardware? We who are not have a different
perspective.

I tried upgrading to X 4.02 last week. I think I had to power-cycle my
machine 4 or 5 times. I lost count. It was locking up. I ended up
restoring /, /var, and /usr from a backup to get back to a stable
configuration. My linux box is sitting behind a firewall on a home
lan. So I'll stay with xf 3.3.6 for now, thanks. It works and my box
is nice and stable.

Sincerely,

-John Palkovic

-- 
"The whole problem with the world is that fools and fanatics are always so
certain of themselves, but wiser people so full of doubts."
-- Bertrand Russell

Date: Sat, 17 Mar 2001 12:59:18 -0500 (EST)
From: Joe Klemmer <klemmerj@webtrek.com>
To: <letters@lwn.net>
Subject: "Harlan Ellison vs. the right to code" and "Perl Literacy"


	I read this article with a bit of amusement.  A number of years
ago the Sci-Fi cable channel had a show "Sci-Fi Buzz" on which Harlan had
a 5 minute rant segment on which he would spew venom on whatever topic he
felt like, though usually with some kind of SF bent.  Half the time I
agreed with him and the other half I thought he was a total brick-head.
I remember one where he blasted every 'Net user as being completely devoid
of any intelligence or even habits of personal hygiene.  But I digress...

	The point of his statement "This presents interesting issues
regarding the responsibility for the release of software which effectively
pollutes the intellectual property environment" is one that we all should
take a good, hard look at.  Just the idea that software is in some way
different than any other tool for information distribution is one that
seems to becoming more prevalent of late.  Napster being the highest
profile case of this but there's more issues, like the DMCA and
IP/Copyrights, that need to be fought off.  The phone company is not
legally responsible for illegal acts done by users of their system (i.e.
threatening calls, drug sales, etc.).  Software developers should have no
less protection.

	On a smaller side issue; the little blurb about 99.99% of high
school seniors not being able to read perl was, IMO, not something worth
publishing.  Perl is a good thing and and all but not being fluent in it
is in no way going to make US high school seniors "painfully unprepared
for life after graduation."  In the grand scheme of things (and even in
the world of software development) perl is a very little blip on the
radar.  When it comes to software development C/C++ and COBOL are more
important programming languages to be fluent in for working on any new
software development project (I won't even mention the maintenance side of
things where COBOL has more code in existing systems than all other
languages combined).  Better that high school seniors be fluent in English
and software development fundamentals than any specific language.
Expecting high school seniors to be fluent in perl is on par with
expecting them to be fluent in Japanizes.

---
If I actually _could_ spell I'd have spelled it right in the first place.

From: mschwarz@alienmystery.planetmercury.net
Date: Thu, 15 Mar 2001 12:37:31 -0600 (CST)
To: letters@lwn.net
Subject: Ol' Uncle Harlan

I am a deep admirer of Harlan Ellison.  Anyone who can create as much art
(and good art too boot) and still find time to stir every hornet's nest he
can find (always in the name of the integrity of the creator's rights,
mind) is someone far more worthy of admiration than any pop star or sports
hero.

What LWN finds ominous, I find familiar.  We have been through this before
with copiers and VCRs.  Harlan is not to be denigrated for attacking
Gnutella.  He is availing himself of his legal rights to bring suit in
court.  The question to be settled is not the right to code, but the right
to steal creative works.  There can be no doubt that the copier and the
VCR can be used to facilitate theft of creative works.  The question that
was settled in those cases was that there were substantial legal uses of
those devices that outweighed the potential for criminal use.  That's why
you can still buy copiers and VCRs.  That doesn't make it legal to pirate
books or movies with them.

Harlan (and his lawyers and co-plaintiffs) is asking a court to decide the
same question of Gnutella.  Now, if you want to defend gnutella, I suggest
that you stop accusing Harlan Ellison of trying to gag programming, and
start making legal use of Gnutella to copy files you have a right to
copy.  Start using to set up web server mirrors and so forth, so there is
a body of legal use to point to in court.

If, as I suspect, the primary use Gnutella is criminal copyright
infringement, then Harlan and company have every right to use the courts
to block its use.  I think too many users of Free Software think it means
the disappearence of author's ownership and rights.  Nothing could be
further from the truth.  Free Software is a redfinition of how the author
is compensated for his or her creative effort.  He gets paid in kind with
the free use of others source code.  Harlan is quote right to draw a
distinction between this and, for example, fiction.  A work of art is sui
generis, and quite different from an algorithm.  Certainly a program may
be artfully expressed, but it is not a purely aesthetic construct.  It is,
at leat in part, a practical contrivance.  And the value the GPL places on
the author's work is the practical value.  It says "I'm giving you this
practical thing, and in return, I expect any new practical thing you make
from it to be available to me, and to anyone else."

I must agree with Harlan that this concept doesn't extend to the signle
solitary work of art.  The *ability* copy does not confer the *right* to
copy; neither should Stallman's invention of the GPL be construed as an
implicit right to the same priviledges with a non-GPL'ed piece of code.

Both Ellison and Stallman are arguing that it is the creator's rights that
must be respected.  The copying of copyrighted works of art is an attack
on the same social order that upholds the GPL.  Only a child thinks
everything is "Mine!  Mine!  Mine!"

--
Michael A. Schwarz
mschwarz@sherbtel.net

Date: Sat, 17 Mar 2001 23:36:15 +0000
From: Pete Birkinshaw <pete@binary-ape.org>
To: letters@lwn.net
Subject: ESR on Socialism 

I respect ESR a great deal for the marvelous work he has done for Open
Source Software, but his claim that

"Under socialism, if you do not choose to "cooperate", you will be
oppressed, imprisoned, and quite possibly killed."

is bizarre. Most of the developed world is frequently governed by
"Socialists", yet I can't remember them being any more oppressive to
their citizens than the USA is. Is he really grouping the governments of
Canada, Europe, Australia and so on with Stalinist states? If he is,
how?

Totalitarian, oppressive states are nasty whatever their political
alignment. Mr Raymond should try to see the difference between liberal,
christian-democratic parties and communist dictatorships.

In all fairness, Thomas Hood and Andrew Pimlott made the same basic
error, but they aren't as high profile. ESR's views on Open Source
Software deserve to be heard. If he continues to make silly, extremist
statements like that, then most people outside the USA will think he's
crazy, and that will hurt the OSS movement he works so hard to promote.

Is Linux "socialist"?  No. It's based on a gift economy. ESR's right
about that.


Pete Birkinshaw

From: Christian Hellon <xian@lisardcage.fsnet.co.uk>
Subject: free software and politics
To: letters@lwn.net
Date: Thu, 15 Mar 2001 13:24:17 +0000 (GMT+00:00)

Oh, dear, what a tizz about one little word. Thomas Hood
writes a letter offering an opinion on the political
character of a particular social movement - admittedly,
not a commonly-shared opinion - and suddenly half your
letters page is devoted to a mass apoplexia.

The usual suspects figure, of course; Eric Raymond libels
a goodly number of European states with his broad
statement:

   Under socialism, if you do not choose to
   "cooperate", you will be oppressed, imprisoned, and
   quite possibly killed.

Hmm. The USA is a capitalist democracy, and has the death
penalty. Sweden is a pretty good example of a socialist
democracy, and doesn't. Where is one more likely to be
"imprisoned and ... killed", exactly?

P James writes:

   But when you publicly defame someone, as Mr. Hood did,
   he ought to at least provide some proof of his opinion.

This gets to the heart of the matter. Over in the UK (in
fact, in most of Europe) socialism isn't a dirty word,
and any attempt to claim that saying someone is a
socialist is somehow defamatory would be laughed out of
most serious arenas (except possibly John Smith House,
but that's a minor detail). Not to mention that socialism, 
communism and anarchism are all very distinct, with quite
separate historical roots and very little sympathy for
each other.

But this confusion over where in the political spectrum
the whole free software movement lies would appear to
indicate that we seem to have something genuinely
revolutionary on our hands, which indeed we do - as Eric
Raymond points out in CatB, we've rectified the tragedy of 
the commons; a commonwealth of software now exists, and
rather than being taken from, every time it is used it is
added to. More than this, we've created an entirely free
market; it's capitalism without the capital. Every player
can compete on equal entry terms, for there is no scope
for monopolisation of any kind. The only determinant to
how well you do is how good you are. Isn't this what Ayn
Rand was on about for pretty much the entirety of "Atlas
Shrugged"?

And yet, there are some distinctly anti-capitalistic
overtones to the whole business. As capital (also known as 
property) has effectively ceased to exist, so the concept
of "ownership" has been undermined. Everyone knows Linus
is "the guy who wrote Linux", but which of us would be
brave enough to claim that he, or Alan Cox, or any of the
other developers "own" their code? Certainly in this
society, ownership dictates a certain level of rights, but 
adopting the GPL as a licence amounts to a voluntary
rescindment of many, if not all of those. The right to
future control over the product, for example - Linus can't 
withdraw his code from the GPL, only his continued
efforts. Nobody can. So we end up with a protected
commonwealth - a deeply left-wing ideal, whichever hue you 
prefer. And since we've managed it without state support
(one could even say "despite state opposition", given what 
copyright law was intended to do), it could be argued that 
it's closer to anarchism than to anything else.

Hence all the ideological arguments - it's a genuinely
confusing position. I come off well from it, because it
reflects my own confusion. :-) But at the end of the day,
it's just software; it makes computers work better, but it 
doesn't solve any pressing social concerns. Could someone
please figure out a way to apply it to growing wheat?
-- 
the desk lisard is at reply@lisardcage.fsnet.co.uk
"i don't know why i'm crying, am i suspended in gaffa?"



____________________________________________________________
Freeserve - get your free ISP service including web-mail at:
www.freeserve.co.uk

Date: Sat, 17 Mar 2001 17:14:59 +0000 (GMT)
From: Jonathan Riddell <jr050@jriddell.org>
To: letters@lwn.net
Subject: Eric Raymond's LTE

I would like to thank Thomas Hood for his recent letter on free software
and socialism.  He made some good and reasoned arguments.  What he didn't
do was accuse his targets of being murderers, as Eric Raymond did in his
reply - a more shallow and un-thought out argument would be hard to find.  

Clearly American schools are doing a poor job of teaching a balanced
criticism of all political ideologies.  If you can't "cooperate" under
pure capitalism then you'll struggle to survive. A person who is, say,
physically disabled in a socialist country will receive fair state help.

As for Eric Raymond, he has lost all my support for anything "open
source", from now on it's free software all the way for me.

Jonathan Riddell
Bridge of Allan, Scotland
jr@jriddell.org
http://www.jriddell.org


________________________________________________________
1 Allanvale Road   |   jr@jriddell.org
Bridge of Allan    |   http://www.jriddell.org
FK9 4NU Scotland   |   01786833048

Date: Thu, 15 Mar 2001 16:20:13 -0500
From: "Steve Mercer" <mercer@nortelnetworks.com>
To: metanews@metagroup.com
Subject: re: Commentary: Microsoft co-opts open source approach

>From http://news.cnet.com/news/0-1003-201-5067896-0.html
Some comments...

>However, the agreement does not allow customers to modify or customize the
>code, and Microsoft anticipates that problems or bugs that customers may
>find in Windows will be reported to Microsoft for resolution through normal
>support channels. 

[snip]
>The advantage of providing Windows source code is that Microsoft enlists tens

>of thousands of software professionals in 1,000 or more of its biggest
>and best customers to help it test its key operating systems in their unique

>environments. This will create a flood of bug fixes, improvements and

>extensions that will flow back to Microsoft to improve those products. 

 How can bug fixes be developed, when the code is legally read only?
Personally, I'd be hard pressed to accept any bug fixes provided to me
that were, legally, untestable by the writer.

 Further, there's more to bug fixing than simply perusing code. Whole
classes of bugs, such as store tramplers and race conditions, are almost
impossible to see via perusal only.

>In our opinion, the Windows source code will inevitably end up on 
>the Web--within six months or less--where thousands more hackers will start

>working on it, exposing weaknesses. This will help Microsoft improve its

>products further until they are bulletproof. 

  The motivations - and ability - to create these fixes goes down 
drastically when the source of the code is a black hole for fixes, 
and the code is illegal to modify. Sun has tried opening code 
through limited NDAs, and it seems to have fizzled into obscurity 
as a model upon which to cornerstone technology development.

  It's much like how openness works in crypto circles. No one will
bother with cryptanalysis of a cipher that is developed in secret,
because it can't be systematically analyzed for weakness. The benefits 
to cracking a particular cipher challenge contribute nothing in terms 
of verifying the cipher's strength, thus, to the cryptanalyst, there
is little worth in doing the work. Similarly, offering read only access
to code, and preventing programmers from making thier own improvements
to the code, essentially removes any incentive for a programmer to peruse
the software at all, as, like cryptanalysts analyze ciphers, programmers
want to write code. There's little benefit to a programmer to find a bug,
only to send a report to Microsoft. They could be doing other projects
that will allow them to write the fixes here and now.

  So that thousands might be scores, maybe hundreds, and that flow of
bug fixes might be a trickle, and none of which would really be 
worthwhile to Microsoft.

  It is my suspicion that Microsoft will, in future, see that this
model is not as benefical to the code quality as expected, and cite this 
as a failure of the open source concept, and not its own implementation 
of it. Which is sad, as they already have Sun's example to work from.


-- 
Stephen Mercer <mercer@nortelnetworks.com>, Optical Design (613) 765-3214