See also: last week's Letters page.

Letters to the editor

From:	 <rs@rcsimpson.demon.co.uk>
To:	 <letters@lwn.net>
Subject: Audio editors
Date:	 Mon, 1 Oct 2001 23:15:23 -0400 (EDT)

Sir,

In last week's "On The Desktop" the opinion was expressed that Linux has
lots of sound file editors, but none that are realy useful.  Never was a
truer word spoken!!

I frequently need a sound editor to cut and mix sound effects for a local
amateur dramatic society, and obviously I try to do it in Linux.  Sadly, I
am often sorely tempted to give up and do it on Windows.  If you go to
Freshmeat you will find at least a dozen progams which claim to edit sound
files and let me tell you that I have tried all of them.  They either
won't install, or won't compile, or crash, or have incomprehensible user
interfaces, or zero documentation or don't provide even the simple
features which I require.  However, almost without exception they have a
web site which says "This is going to be the Linux CoolEdit2000 / Sound
GIMP but its still in the early stages of development"

If I were reading this letter, I would say at this point, "This is free
software.  If you don't like it get off your ass and write some code."
But that's not the problem.  There are plenty of people writing sound
editors.  The problem is that instead of writing 80% of two editors they
have written 15% of 12 editors.

As a user who just wants a simple reliable sound editor with a modern look
and feel can I send a plea to sound editor authors to get together and
concentrate their efforts on just one or two projects.  It will hurt some
egos, but what Linux needs most is quality applications.

Thank you,

	Richard Simpson

-- 
---
Richard Simpson @ home

From:	 Fred Mobach <fred@mobach.nl>
To:	 Linux Weekly News <lwn@lwn.net>
Subject: Linux DA
Date:	 Thu, 27 Sep 2001 10:37:20 +0200

Hello,

Regarding LinuxDA you wrote "The GNU GPL did not seem to be included."

I've also downloaded the source tarball and posted my comments on Linux
Today (see
http://linuxtoday.com/news_story.php3?ltsn=2001-09-25-012-20-NW-EM-LL).

One of my comments was 

"And before I forget, the contents of the file 
-rw-r--r-- 1 fred users 18458 Jul 13 1998 linux/COPYING 
seems quite usual."

That happens to be the GNU GPL with Linus' remark on top of it.

Regards,

Fred
-- 
Fred Mobach - fred@mobach.nl - postmaster@mobach.nl
Systemhouse Mobach bv - The Netherlands - since 1976

Fight terrorism, be it killers or software patents.

From:	 Leandro =?ISO-8859-15?Q?Guimar=E3es?= Faria Corsetti Dutra
	 <leandrod@mac.com>
To:	 letters@lwn.net
Subject: Gartner: dump IIS
Date:	 Thu, 27 Sep 2001 13:39:40 -0300

> Apache is the dominant web server platform; anybody wishing to attack

> large numbers of systems via a web server would look at Apache first.

> The "obscure and uninteresting" argument will not wash here.


	The argument is weak because it is hypothetical: no one know for sure how many 
crackers have tried their hands at Apache and failed.  So it's nearly 
impossible to prove it.

	But it is also impossible to disprove it.  While Apache is the most used web 
server, it is also uninteresting to attack, since the source code is 
available.  Not only that, IIS has a much bigger presence in the really 
interesting sites to attack, those of hated multinational companies or 
organisms making commerce or propaganda over the Net with little technical 
qualification in house.


	In other words, using Apache has some correlation to being "a nice guy" who 
won't be cracked, but a much bigger correlation to being a technically 
knowledgeable organization with whom perhaps a script kiddie shouldn't meddle.



-- 
  _
/ \ Leandro Guimarães Faria Corsetti Dutra           +55 (11) 246 96 07
\ / http://homepage.mac.com./leandrod/     BRASIL    +55 (43) 322 89 71
  X  http://tutoriald.sourceforge.net./     mailto:lgcdutra@terra.com.br
/ \ Campanha fita ASCII, contra correio HTML    mailto:leandrod@mac.com

From:	 Billy Tanksley <btanksley@hifn.com>
To:	 "'letters@lwn.net'" <letters@lwn.net>
Subject: Linux security debate
Date:	 Thu, 27 Sep 2001 13:33:50 -0700

I'm puzzled by the people claiming that the Linux security API should
include an automated check for free software.  My problem isn't a dislike of
free software; rather, I'm puzzled that people are proposing a modification
to a free software product which can easily be removed.

If a company wants to distribute a closed-source "security" plugin, they
simply have to modify the Linux kernel to allow closed-source.  Nothing
easier.

Therefore, this debate shouldn't be about automated mechanisms; it should be
about whether a security module is the same thing as a kernel module, or
whether security modules should get the same exclusion from the GPL which
the kernel modules enjoy.

And I won't even go there.  :-)

OTOH, it occurs to me that perhaps the error was in the story (or my
understanding of the story), rather than in the arguments.  This issue
certainly deserves better coverage -- I hope you'll discuss it in greater
length.

-Billy

From:	 <anonymous>
To:	 <lwn@lwn.net>
Subject: SSSCA
Date:	 Sat, 29 Sep 2001 22:18:39 -0700

Honorable Gentlemen:

I read the following on the internet
(http://www.linuxnews.com/stories.php?story=01/09/17/9591923):

A controversial law was recently drafted, although not yet introduced, by
U.S. Sen. Fritz Hollings (D-South Carolina), chairman of the Senate Commerce
committee, and Sen. Ted Stevens (R-Alaska), titled "The Security Systems
Standards and Certification Act" (SSSCA). If made into law, the proposed bill
would make it illegal to "manufacture, import, offer to the public, provide
or otherwise traffic in" digital devices that do not use "certified security
technologies." The SSSCA, if enacted, will have serious negative affects on
free software, most specifically to open-source operating systems such as
Linux. 

I find this extremely disturbing. I work at Intel and this bill is coming on
the heels of some related requests recently made by Microsoft of Intel (I
don't feel comfortable going into details).

Frankly, it appears that not only has Microsoft gotten away with alleged
anti-competitive behavior with only a token "slap on the wrist", but now
wants to make it illegal to even compete against them. If this bill is
passed, I will seriously start to wonder whether Microsoft is the only
software lobby in Washington. Please consider this letter the voice of
another lobby -- the open source community.

If you feel you must pass a law to protect copyright material on computing
platforms, please develop a law that allows all software writers and hardware
developers the ability to comply without the threat of financial ruin by an
alleged monopolist such as Microsoft. My definition of "ability to comply"
would be a law that follows a standard software file format and standards for
encryption/decryption algorithms/hardware mechanisms. 

Either the military and/or the university system should develop these
standards such that no licenses would have to be obtained from or fees paid
to private companies such as Microsoft. If ideas are taken from private
industry for incorporation into this standard, then it should be up to the
government to compensate contributing private companies for the intellectual
property incorporated into the government standard. 

An individual or company should only be charged a fine if they are found to
be out of compliance. It should not be that a company should have to wait for
government certification before issuing software. This avoids a situation in
which a large company such as Microsoft has a dedicated inspector that can
make timely certifications verses a small developer who may have to wait
months to get a non-dedicated inspector to certify the software. Developers
that are found to be in compliance with the standards should be immune from
lawsuits that may arise from damages caused by hackers.

From:	 Charles Cazabon <wwwpatentpolicy@discworld.dyndns.org>
To:	 www-patentpolicy-comment@w3.org
Subject: "Non-discriminatory" patents will kill the web
Date:	 Tue, 2 Oct 2001 08:06:57 -0600
Cc:	 letters@lwn.net

Dear sirs,

The W3C is currenty bandying about the idea that incorporating patented
technologies in proposed W3C standards would be acceptable, providing that the
patent holder agreed to license the patents to all comers for a reasonable and
"non-discriminatory" fee.

What leap of logic led to this great fallacy?  How did the W3C come to propose
this absurdity which could lead to the destruction of not only itself, but of
the web in toto?

Patents serve one purpose -- they allow the patent holder to exclude others
from manufacturing or using an "invention" of the patent holder.  The patent
holder can then use this government-granted permit to either monopolize a
market legally, or to extort money from all other parties interested in
participating in a market.

The W3C, on the other hand, has always tried to promote technical
interoperability and sane standards.  It has done an admirable job of this.
Without royalty-free standards such as those that underpin the web today,
where would we be?  How advanced or useful would the web be if HTML or other
standard were encumbered by patents?  How diverse would the content of the web
be if the only organizations that could publish content were the ones that
could afford to purchase patent licenses?  The actual dollar amounts are
immaterial; any patent license fees at all would have completely eliminated
all but the commercial players, and the commercial players are not what have
made the web a success today.

The W3C must reject this RAND proposal; it must refuse to endorse any proposed
standard that uses patented technologies or methods unless a royalty-free
license is granted to all interested parties.

Charles Cazabon
-- 
-----------------------------------------------------------------------
Charles Cazabon                  <wwwpatentpolicy@discworld.dyndns.org>
GPL'ed software available at:  http://www.qcc.sk.ca/~charlesc/software/
-----------------------------------------------------------------------