Leading items and editorials

	file_uploads = Off

<linux/blk.h>
DEVICE_ON
DEVICE_OFF
DEVICE_REQUEST
kdev_t
ioctl
struct block_device
From:	 Brian Beesley <BJ.Beesley@ulster.ac.uk>
To:	 letters@lwn.net
Subject: BitKeeper controversy
Date:	 Thu, 25 Apr 2002 08:28:34 +0000

Hi,

Please note that I'm not an active kernel developer. However, like many in 
the linux community, I do have an ideological interest in this debate.

1. I do NOT think it is wrong to use a commercial product in association with 
development of the linux kernel (or any other open source software). As far 
as I'm concerned, if it helps those actively involved in the development 
and/or management of the development of the linux kernel, that's fine.

2. I think it IS wrong if for some reason developers who elect not to use a 
specific commercial product find obstacles in the path to getting their work 
implemented which would not be there if they were using that specific 
commercial product. (This situation is analagous to being unable to get work 
accepted by publishers unless you submit documents in MS Word format.)

3. The kernel source is big enough without inclusion of material in the 
nature of advertising for a commercial product. A small text comment to the 
effect that "Product X has been used to support development of this product" 
is more than sufficient.

4. I don't see any reason to accept the inclusion of "billboards" into the 
linux product (source or binary), even if commercial organizations were to 
offer real money to sponsor their inclusion. IMO "free software" means "free 
of intrusion by advertising" as well as "free as in beer" and "free as in 
spirit". The problem here is, if we accept advertisements in source code, 
where do we stop? Advertisements appearing during system startup? 
Advertisements during user login? Advertisements appearing at random times 
during normal operation? 

Regards
Brian Beesley

From:	 David.Kastrup@t-online.de (David Kastrup)
To:	 letters@lwn.net
Subject: Free Software / Bitkeeper
Date:	 25 Apr 2002 11:49:00 +0200

You wrote in your editorial:

    There seem to be two main camps in the free software realm. The
    first sees free software as something that is fun, useful, and
    preferable whenever possible. This group is far more interested in
    getting the job done than worrying about the pedigree of its
    tools. Linus Torvalds, a highly visible member of this group,
    expressed it this way:

        Quite frankly, I don't _want_ people using Linux for
        ideological reasons. I think ideology sucks. This world would
        be a much better place if people had less ideology, and a
        whole lot more "I do this because it's FUN and because others
        might find it useful, not because I got religion".

        Would I prefer to use a tool that didn't have any restrictions
        on it for kernel maintenance? Yes. But since no such tool
        exists, and since I'm personally not very interested in
        writing one, _and_ since I don't have any hangups about using
        the right tool for the job, I use BitKeeper.

    The other point of view sees proprietary software as an evil to be
    avoided at all costs. Even discussion of proprietary software is
    to be avoided; [...]

This _very_ clearly points out the validity of the arguments from the
"radical" Free Software proponents that try to avoid (and tell people
to avoid) non-free software at all costs.

Creating software needs an incentive: the non-availability is one
such incentive.  If one always turned to proprietary software
whenever Free Software of equal quality was not yet available, there
would be no incentive to develop or improve Free Software whenever
proprietary software was already available.  Free systems would be
non-existent, since it is hardly likely that _all_ components of such
a system were as much of a novelty as to never have existed in
proprietary form before.

Free Software needs users, developers, interest to thrive.  The stance
of Linus Torvalds with regard to Bitkeeper is not likely to foster
development of free alternatives.  But who knows?  Perhaps the obvious
lack of a free alternative will that way be pointed out much more
prominently, and will be an incentive to get something going.

When Stallman set out with the GNU project, avoiding proprietary
software was often not possible, and often painful.  If there had been
noone willing to subject himself to the insufficient situation,
progress would never have been made.  The continuous availability of
software, once free, makes it more resistant to bit rot, developer
focus changes and other mishaps often killing off or thwarting its
proprietary cousins.  That way, in the long run, Free Software stands
a good chance to prevail.  But only in those areas where stagnation
does not spell doom, and the short run has to get off the ground
somehow, too.

As an example, current proprietary compiler technologies from
processor manufacturers sometimes beat the performance of GNU gcc by
quite a margin.  If one followed the "pragmatical" approach of Linus
Torvalds to its conclusion, it would mean that distribution authors
and kernel developers should switch to proprietary compilers for
creating Linux kernels where available.  This would increase the
performance to be expected from Linux systems.  IT would also remove a
big incentive to further gcc development, not least of all by
processor vendors.  Currently, if they want to have their processor
supported and accepted by the Linux community, they need to help with
improving gcc.  If the nonchalant stance with regard to the use of
proprietary software for the creation of free systems spread further,
then there will come a time when Linux will be more efficient, at the
cost of not being able to recompile a kernel unless you shell out the
appropriate amount of money.

That's not in the interest of Free Software.  It is not even in the
interest of Open Source software which tries to deny the starting
labors of Free projects by claiming that the process itself is
guaranteed to lead to superior results from the setout.

For this reason, the obstinacy, pigheadedness, perseverance and
dedication of people like Stallman is not just an exercise in
futility.  Sometimes groundbreaking work is required, and the "there
is something proprietary available already, we don't need to do this"
stance will not set it in motion.

-- 
David Kastrup, Kriemhildstr. 15, 44793 Bochum
Email: David.Kastrup@t-online.de

From:	 Robert Hart <enxrah@nottingham.ac.uk>
To:	 letters@lwn.net
Subject: OpenCD
Date:	 Mon, 29 Apr 2002 17:12:23 +0100 (BST)

I have believed for a long time that the free-software is making a big
mistake by concentrating solely on *nix.

Any evangelist will tell you that you can't wait for somebody to come into
a church if you want to preach the gospel. 

I realise that windows and unix are very different beasts, but if people
are to be free then we must release the grasp of microsoft finger by
finger. Mozilla to replace IE, Gimp to replace Photoshop, Abiword to
replace word (ok wordpad), etc, and *then* Gnome/Kde/whatever to replace
the lot.

I hope the OpenCD project can help acheive this, however for it to work it
needs open source projects to support this vision. 

When this does happen (mozilla, freeamp, abiword) the results benefit us
all - more developers, more users, more bug reports, better software.

When this is ignored, we are left with 'flagship' applications (evolution,
gnumeric), that will never be used by the people who could benefit from
them the most.

Robert Hart

From:	 Rainer Weikusat <weikusat@students.uni-mainz.de>
To:	 letters@lwn.net
Subject: twisted
Date:	 Fri, 26 Apr 2002 14:04:09 +0200

Twisted is written in Python, a high-level language, rendering it immune i
to the most common class of security flaw in network software

	'incompetent programmers'. 

From:	 "Tom Cowell" <tcowell@anitesystems.de>
To:	 <letters@lwn.net>
Subject: Best Sot Linux
Date:	 Thu, 25 Apr 2002 11:13:52 +0200

So,

Best Linux is Sot Linux again, because it no longer matters that Sot means
"disease" or "soot" in Swedish.

Does it matter that "sot" means "drunkard" in English?

Tom

From:	 Michael Concannon <cekim@ix.netcom.com>
To:	 letters@lwn.net
Subject: Regarding RMS's interview (a litte belated)
Date:	 Sun, 28 Apr 2002 22:29:55 -0400

I have long been an advocate of free software in my personal and
professional life.  However, I am by nature a pragmatist and
economically conservative excluding all but the most mundane forms of
activism...

That said, after using Linux since the 1.x.x kernel days, flailing
through the early days of Alpha/AXP integration and now moving on to
integrating Linux into my professional life (ASIC Design and
verification), I am troubled by the extremism of some of those who are
out there rightfully and thankfully fighting to ensure that free
software can:

1. exist at all in the face of increasing regulation
2. coexist (I would hope that is a goal) with non-free software.

Like many Linux power users I write software for a living.  Whether it
is in C/C++, Perl, VHDL, Verilog or anything else, what I do day-to-day
is write, compile, debug and sell my services in developing software.

I greatly appreciate the ability to do so and I feel that it is my right
to sell my services in this form as I see fit at a price the market will
bear.

In no way does this mean I oppose the distribution, creation, protection
and general proliferation of free software.  

In fact, as I mentioned above, I find myself advocating and facilitating
its use frequently where it offers an equivalent or superior solution. 
Always making sure to account for the real costs of using free software
(all software has bugs -- someone has to make sure they either do not
hurt you or are fixed in a timely manner).  

Further I have written a great deal of code to facilitate its use (which
is very much not free and strictly speaking the property of my employer
-- to whom I am really selling it for the sum of my salary).

To quote Richard Stallman (taken from the lwn interview):
"At a deeper level, though, the biggest threat to the future of free
software is the idea that non-free software is acceptable."

Recognizing that historically, extremists have driven incremental
change, this seems like a flawed goal for the following reasons:

1. Software development has a high cost in the form of human time taken
or given from a skilled and specialized developer.  Any time you are
asking someone to give of their time, you are asking them to give up a
portion of their existence.  Some times, for the greater good, it is
worth it to do so without direct compensation, other times, it is only
valuable to a small group of people and should be traded for something
of value (i.e. money!)

2. Non-Free software need not fear free software and vice versa.  People
trying to make money from solving a problem which is already solved need
to fear it and move on to more interesting problems.  People solving a
given problem poorly and coercing you to use their flawed solution need
to fear even more.

3. The more pressing concern is not that copyright and IP ownership
exist but that at present anti-trust violations, inane interpretations
of the laws and laughable execution by the US Patent Office allow them
to be used to prohibit people from using free software to get what they
want...

True Type Font rendering and DVD players demonstrate this at present... 
I am willing to trade my time and frustration for my hard earned money,
but I cannot easily do so at present because of the current litigious
environment with regard to "digital rights".

In short, the more useful goal would be to facilitate the free
integration of free and non-free software.

Non-free software will ultimately benefit from standards created by free
software.  Users will benefit from the ability to select from a range of
solutions and pick what fits them and their needs.  Free software will
benefit in that it will be used and maintained as it is the foundation
on which non-free and free software rely to execute.

It is always nice to find tarball on the net that solves a pressing
problem.  It is rewarding to circulate your own "solution" in return... 
Further, there are tangible benefits to releasing your source to the
public:

a. If it becomes the standard, you can ensure that things you build upon
it will work in the future
b. Developers can spend more time solving more higher level problems
which build upon your solution.
c. Someone else may return the favor and release a solution to a future
problem of yours...

However, you must recognize that __everything__ you do in life comes
with a cost (time, money or longevity).  You have to keep this in mind
what you talk about software being "free".  

The best answer is almost always exactly half way between the two
extremes and I do not see how this is an exception...

/mike

Sections: Main page Security Kernel Distributions Development Commerce Linux in the news Announcements Letters See also: last week's Development page.	Development projects News and Editorials The future of omniORB. omniORB is a CORBA 2.1-compliant object request broker which was developed at the AT&T Cambridge Laboratory. It has a reputation as being one of the better free ORBs available, and it has a significant user base. AT&T, of course, has decided to close down the Cambridge Laboratory; this closure has cause some understandable curiosity about the future of free software projects that were run out of that lab. In the case of omniORB, the news looks reasonably good. omniORB hacker Duncan Grisby is taking the project independent, with a site on SourceForge. Interestingly, this project will become much more open to community contributions than it previously was: In the past, we always discouraged contributions of anything except bug fixes and ports to new platforms. That was largely to make sure AT&T kept the copyright to all the code, so we had the flexibility to relicense it if we wanted to, and to protect AT&T from copyright claims. Obviously, that reason is no longer an issue Mr. Grisby tells us that the previous policy did not impair omniORB development; indeed, he claims, a robust, high-performance system is best developed by a small, tight group. Even so, AT&T's firm grip on the code shows how many companies, while they are increasingly supportive of free software, are still reluctant to really turn the process loose. On the other hand, omniORB development could yet become a little too loose, given that AT&T's support for its development has been terminated. Mr. Grisby has plans for both the short term (omniORB 4.0 and omnORBpy 2.0 releases) and the long (asynchronous methods, passing objects by value), but the financial support for that work is no longer. So Mr. Grisby is seeking support for continued omniORB work; it looks like time for those who benefit from omniORB - or who would benefit from future enhancements - to step up and help ensure that development continues. For those who are curious, a web site dedicated to tracking ex-Cambridge projects and people has been set up at xorl.org. Audio Projects AlsaPlayer 0.99.60 has been released. This release consists mostly of bug fixes and infrastructure work; see the changelog for details. WaveSurfer 1.3.1 has also been released. The changes appear to be mostly related to documentation; details in the changelog. Education SEUL/edu Linux in Education Report. The April 29 Linux in Education Report is available from SEUL/edu. It looks at difficulties with Linux gradebook development, Microsoft's guide on donating computers to schools, and numerous other topics. Thai Tales: Taking Computers to Schools (Linux Journal). Linux Journal looks into the use of Linux in Thailand's SchoolNet project. "'Initially we used Windows NT on a straightforward PC. Then we developed the Linux schools internet server. We now have our own software, running GNU/Linux, which is managed via the Web, using the Thai language. That means, to run it the user hardly need to know anything of UNIX. This runs on just a PC. Compared to it, we could not afford a Sun Microsystem box and router for each school, for example,' says Koanantakool." Embedded Systems SnapGear Ports uClinux to Motorola ColdFire MCF5249. Here's a press release from SnapGear. "SnapGear engineer, Greg Ungerer, has recently completed uClinux support for the new Motorola MCF5249 ColdFire(R) CPU, specifically for M5249C3 development board. Patches have been made available for 2.0.x and 2.4.x versions of uClinux kernels." Building a secure kiosk with Embedded Linux. LinuxDevices features an article on building a Linux based information kiosk. "In this informative and entertaining technical article, embedded developer Patrick Glennon relates his experiences in creating a small Linux-based system for a client that required robust, easy-to-use, low-cost kiosks for conducting surveys at hotels." BusyBox 0.60.3. BusyBox 0.60.3 has been released; see the changelog for details. BusyBox may be designed to fit into limited space, but the developers have still somehow found room, alas, for color "ls". BusyBox may be downloaded from the project web site. Network Management OpenNMS 1.0 has been released. OpenNMS has announced the release of version 1.0 of its enterprise grade network management platform. Printing Software GNU Ghostscript 7.05 released. The release of GNU Ghostscript 7.05 has been announced. This is the first GPL release of Ghostscript 7.x - though this code has been available under the AFPL for almost a year. New features include better Asian language support, improved PDF handling, and more. Also available is ESP Ghostscript 7.05.1, which adds fancier configuration, the CUPS raster drivers, and a number of other things. Web-site Development Puffin 0.8.8. Puffin is a regression testing framework for web-based applications; it is written in Python. The announcement for this release claims a number of advancements, including greater ease of use, better response analyzers, etc. The Python Web Frameworks Overview has been announced by Paul Boddie. The Overview is a set of documents describing the (numerous) web development frameworks available for development in Python, and the ups and downs of each. It is still much in development (there are not yet entries for all frameworks), but the initial contents look promising. Nemein releases NemeinAuthentication library. Nemein has announced the release of its NemeinAuthentication library, session-based authentication library for Midgard. The library has been released under the LGPL license. (Thanks to Henri Bergius.) Standards LPI News April 2002. The April, 2002 publication of the LPI Newsletter is available. Table of Contents: LPI-Europe and LinuxTag - Karlsruhe, Germany Discounted Exams in California Linux World Japan - http://www.idg.co.jp/expo/lw/ May 29th - 31st LPI alumni conference in Toronto, Canada Volunteer of the month - Brian Beck of Bradford Learning.com Over 10,000 LPI exams taken ! Waiting for your LPI certificate to arrive ? New sponsors; http://www.lpi.org/a-sponsors.html LPI T-shirts. New LPI supporters - Gavin Buckley and Chloe York The LPI also has a press release out on the giving of its 10,000th test. Documentation LDP Weekly News, April 30th. This week's edition of the Linux Documentation Project's news it available. Besides several new and updated documents, most of this week's newsletter is devoted to the announcement of Lampadas, their brand new documentation management system.	May 2, 2002 Application Links GIMP Mozilla Galeon High Availability ht://Dig mnoGoSearch MagicPoint Wine Worldforge Zope Open Source Code Collections Berlios Freshmeat OpenSourceDirectory Savannah Le Serveur Libre SourceForge Sweetcode

	Desktop Development Desktop Environments KDE Application Of The Month: KTouch (dot.kde.org). Here is this month's KDE favorite. "As part of the May 2002 issue of "Application of the month" series on KDE.de, Klaus Stärk has interviewed Håvard Frøiland, author of KTouch. KTouch is part of the KDE Edutainment Project and provides a quick and fun way to learn the useful and impressive skill of touch typing." Desktop Elegance (mosfet.org). An editorial at Mosfet.org defends KDE against Eazel cofounder Bart Decrem's attacks. KDE Stats: KDE Is Brought To You Today By.... KDE.News looks at the berliOS project's KDE CVS statistics. "Have you ever wondered who contributes what to KDE? The berliOS project attempts to answer this question with KDE CVS statistics, a site tallying every developer's contributions" GNOME 2.0 Desktop Beta 4: 'Thank You' (Gnotices). The Gnotices site is carrying an announcement for the new GNOME 2.0 beta 4 release, code named "Thank You". Games Civil 0.70 released. Civil is "a cross-platform, turn-based, networked strategy game" based on the U.S. Civil War. Version 0.70 has been released; it includes no end of improvements and new features. GUI Packages GNUstep Weekly Editorial. The GNUstep Weekly Editorial for April 26 is available with the latest from the GNUstep development community. Interoperability TheOpenCD: Free Software on proprietary operating systems. Here's the announcement of a new project that wants to make programs like AbiWord, the Gimp, and OpenOffice available on Windows and MacOS. "The key, as I see it, is to encourage people to use the high-quality Free Software now becoming available in the OS they are already using." Kernel Cousin Wine #120. The 120th issue of Kernel Cousin Wine is available, with coverage of events through April 18, including the "ReWind" fork. Also available is Issue 121, with coverage through April 25. Covered topics include incorporating the ALSA sound system and many others. Office Applications Gnumeric 1.1.3 available. The development version of gnumeric 1.1.3 is now available. "This is a DEVELOPMENT RELEASE it is not supposed to be stable..." AbiWord Weekly News published. The AbiWord Weekly News issue #89 is now available. "This week there have been some really long threads on the future development of AbiWord, primarily about the backend enhancements necessary for tables and improved rendering/i18n. Plans for the near term seem to be in place, and I suspect people have already started hacking." OpenOffice Native Language additions. Italian and Dutch have been added to the OpenOffice Native Language Development project, bringing the project's tally to four languages. German and French are the other supported languages. KOffice 1.2beta1 Ready for Testing, More Developers. The KDE Project announced the release of KOffice 1.2beta1. Open For Business has this report on the new beta. "The most notable improvement in KOffice 1.2 Beta 1 is the new WYSIWYG (What You See Is What You Get) display in KWord and KPresenter. This insures that these applications will finally show documents on screen like they will appear once printed." Kernel Cousin GNUe. Kernel Cousin GNUe #26 includes a summary of the status of the GNU Enterprise project, the beginnings of the GNUe Application Server 2, and more.	Desktop Environments GNOME GNUstep KDE XFce XFree86 Window Managers Afterstep Enlightenment FVMW2 IceWM Sawfish WindowMaker Widget Sets GTK+ Qt

	Programming Languages Caml Caml Weekly News. Here's the Caml Weekly News for April 30, with coverage of the "PhoX" proof assistant, interfacing with Matlab, and more. Perl Second Call for YAPC::NA Papers (use Perl). The second call for papers is out for YAPC::NA. Deadlines are 1 May for papers and 5 June for lightning talks. Lightweight Perl Blogging Tools (use Perl). UsePerl links to several web logging tools, Movable Type, Bloxsom, and Blagg. mod_perl Developer's Cookbook reviewed (Perl.com). Simon Cozens has reviewed the mod_perl Developer's Cookbook. Don't be afraid of Perl 6. Here's an article in System Administration Magazine on what is really changing in Perl 6. The conclusion is that most Perl code will still work, that there is no need to fear the changes. "The same dangerous misfocus occurs every time Larry releases another Perl 6 design document. Our brains instinctively skip over the majority of familiar, unchanged Perl landmarks and, instead, zero in on the comparatively few features of the language that are actually changing." PHP PHP Weekly Summary for April 30. The April 30 PHP Weekly Summary is available; it looks at the 4.2.0 release and several other topics. PHP.net: A Tourists's Guide. Those trying to find their way around the PHP.net site may well want to take a quick look at PHP.net: A Tourist's Guide. "Everyone knows the www.php.net site. All of us went there sooner or later, and will keep going back there. This is the central reference point for PHP users, and it has a wealth of informations there. All of it isn't that obvious. Come with me, I'll show you." 'Programming PHP' Released by O'Reilly. O'Reilly has released "Programming PHP" by Rasmus Lerdorf, creator and lead developer of PHP, and Kevin Tatroe. Python Stackless Python in Limbo? Christian Tismer, author of the stackless Python patch, has announced a bit of a change in direction. Stackless Python will start to look more like the Limbo language used in the Inferno operating system. Limbo defines a "tasklet" mechanism which makes it easy to write concurrent, multi-threaded applications; the tasklets communicate through "channels" which also handle synchronization issues. This week's Python-URL. Dr. Dobb's Python-URL for April 30 is available with the usual roundup of interesting happenings in the Python community. Ruby Ruby Weekly News. The April 29 Ruby Weekly News looks at Ruby/Google 0.3.0, the first release of the Practical Ruby IDE, and more. Review: Ruby Developer's Guide. Slashdot has posted a review of the Ruby Developer's Guide, written by Robert Feldt, Lyle Johnson, and Michael Neuman. "Directed towards programmers with a working knowledge of Ruby, the text is a quick read even with working through the examples. It effortlessly introduces the basic concepts of each package worked through and then gives locations where more in-depth information can be gathered." Debuggers GDB 5.2 has been released; details in the announcement. There are a few new commands and supported architectures, but seemingly no radical changes. Section Editor: Jonathan Corbet	Language Links Caml Caml Hump Tiny COBOL Erlang g95 Fortran Gnu Compiler Collection (GCC) Gnu Compiler for the Java Language (GCJ) Guile Haskell IBM Java Zone Jython Free the X3J Thirteen (Lisp) Use Perl O'Reilly's perl.com Dr. Dobbs' Perl PHP PHP Weekly Summary Daily Python-URL Python.org Python.faqts Python Eggs Ruby Ruby Garden MIT Scheme Schemers Squeak Smalltalk Why Smalltalk Tcl Developer Xchange Tcl-tk.net O'Reilly's XML.com Regular Expressions

Sections: Main page Security Kernel Distributions Development Commerce Linux in the news Announcements Letters 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.	May 2, 2002

From: Brian Beesley <BJ.Beesley@ulster.ac.uk> To: letters@lwn.net Subject: BitKeeper controversy Date: Thu, 25 Apr 2002 08:28:34 +0000 Hi, Please note that I'm not an active kernel developer. However, like many in the linux community, I do have an ideological interest in this debate. 1. I do NOT think it is wrong to use a commercial product in association with development of the linux kernel (or any other open source software). As far as I'm concerned, if it helps those actively involved in the development and/or management of the development of the linux kernel, that's fine. 2. I think it IS wrong if for some reason developers who elect not to use a specific commercial product find obstacles in the path to getting their work implemented which would not be there if they were using that specific commercial product. (This situation is analagous to being unable to get work accepted by publishers unless you submit documents in MS Word format.) 3. The kernel source is big enough without inclusion of material in the nature of advertising for a commercial product. A small text comment to the effect that "Product X has been used to support development of this product" is more than sufficient. 4. I don't see any reason to accept the inclusion of "billboards" into the linux product (source or binary), even if commercial organizations were to offer real money to sponsor their inclusion. IMO "free software" means "free of intrusion by advertising" as well as "free as in beer" and "free as in spirit". The problem here is, if we accept advertisements in source code, where do we stop? Advertisements appearing during system startup? Advertisements during user login? Advertisements appearing at random times during normal operation? Regards Brian Beesley

From: David.Kastrup@t-online.de (David Kastrup) To: letters@lwn.net Subject: Free Software / Bitkeeper Date: 25 Apr 2002 11:49:00 +0200 You wrote in your editorial: There seem to be two main camps in the free software realm. The first sees free software as something that is fun, useful, and preferable whenever possible. This group is far more interested in getting the job done than worrying about the pedigree of its tools. Linus Torvalds, a highly visible member of this group, expressed it this way: Quite frankly, I don't _want_ people using Linux for ideological reasons. I think ideology sucks. This world would be a much better place if people had less ideology, and a whole lot more "I do this because it's FUN and because others might find it useful, not because I got religion". Would I prefer to use a tool that didn't have any restrictions on it for kernel maintenance? Yes. But since no such tool exists, and since I'm personally not very interested in writing one, _and_ since I don't have any hangups about using the right tool for the job, I use BitKeeper. The other point of view sees proprietary software as an evil to be avoided at all costs. Even discussion of proprietary software is to be avoided; [...] This _very_ clearly points out the validity of the arguments from the "radical" Free Software proponents that try to avoid (and tell people to avoid) non-free software at all costs. Creating software needs an incentive: the non-availability is one such incentive. If one always turned to proprietary software whenever Free Software of equal quality was not yet available, there would be no incentive to develop or improve Free Software whenever proprietary software was already available. Free systems would be non-existent, since it is hardly likely that _all_ components of such a system were as much of a novelty as to never have existed in proprietary form before. Free Software needs users, developers, interest to thrive. The stance of Linus Torvalds with regard to Bitkeeper is not likely to foster development of free alternatives. But who knows? Perhaps the obvious lack of a free alternative will that way be pointed out much more prominently, and will be an incentive to get something going. When Stallman set out with the GNU project, avoiding proprietary software was often not possible, and often painful. If there had been noone willing to subject himself to the insufficient situation, progress would never have been made. The continuous availability of software, once free, makes it more resistant to bit rot, developer focus changes and other mishaps often killing off or thwarting its proprietary cousins. That way, in the long run, Free Software stands a good chance to prevail. But only in those areas where stagnation does not spell doom, and the short run has to get off the ground somehow, too. As an example, current proprietary compiler technologies from processor manufacturers sometimes beat the performance of GNU gcc by quite a margin. If one followed the "pragmatical" approach of Linus Torvalds to its conclusion, it would mean that distribution authors and kernel developers should switch to proprietary compilers for creating Linux kernels where available. This would increase the performance to be expected from Linux systems. IT would also remove a big incentive to further gcc development, not least of all by processor vendors. Currently, if they want to have their processor supported and accepted by the Linux community, they need to help with improving gcc. If the nonchalant stance with regard to the use of proprietary software for the creation of free systems spread further, then there will come a time when Linux will be more efficient, at the cost of not being able to recompile a kernel unless you shell out the appropriate amount of money. That's not in the interest of Free Software. It is not even in the interest of Open Source software which tries to deny the starting labors of Free projects by claiming that the process itself is guaranteed to lead to superior results from the setout. For this reason, the obstinacy, pigheadedness, perseverance and dedication of people like Stallman is not just an exercise in futility. Sometimes groundbreaking work is required, and the "there is something proprietary available already, we don't need to do this" stance will not set it in motion. -- David Kastrup, Kriemhildstr. 15, 44793 Bochum Email: David.Kastrup@t-online.de

From: Robert Hart <enxrah@nottingham.ac.uk> To: letters@lwn.net Subject: OpenCD Date: Mon, 29 Apr 2002 17:12:23 +0100 (BST) I have believed for a long time that the free-software is making a big mistake by concentrating solely on nix. Any evangelist will tell you that you can't wait for somebody to come into a church if you want to preach the gospel. I realise that windows and unix are very different beasts, but if people are to be free then we must release the grasp of microsoft finger by finger. Mozilla to replace IE, Gimp to replace Photoshop, Abiword to replace word (ok wordpad), etc, and then* Gnome/Kde/whatever to replace the lot. I hope the OpenCD project can help acheive this, however for it to work it needs open source projects to support this vision. When this does happen (mozilla, freeamp, abiword) the results benefit us all - more developers, more users, more bug reports, better software. When this is ignored, we are left with 'flagship' applications (evolution, gnumeric), that will never be used by the people who could benefit from them the most. Robert Hart

From: Rainer Weikusat <weikusat@students.uni-mainz.de> To: letters@lwn.net Subject: twisted Date: Fri, 26 Apr 2002 14:04:09 +0200 Twisted is written in Python, a high-level language, rendering it immune i to the most common class of security flaw in network software 'incompetent programmers'.

From: "Tom Cowell" <tcowell@anitesystems.de> To: <letters@lwn.net> Subject: Best Sot Linux Date: Thu, 25 Apr 2002 11:13:52 +0200 So, Best Linux is Sot Linux again, because it no longer matters that Sot means "disease" or "soot" in Swedish. Does it matter that "sot" means "drunkard" in English? Tom

From: Michael Concannon <cekim@ix.netcom.com> To: letters@lwn.net Subject: Regarding RMS's interview (a litte belated) Date: Sun, 28 Apr 2002 22:29:55 -0400 I have long been an advocate of free software in my personal and professional life. However, I am by nature a pragmatist and economically conservative excluding all but the most mundane forms of activism... That said, after using Linux since the 1.x.x kernel days, flailing through the early days of Alpha/AXP integration and now moving on to integrating Linux into my professional life (ASIC Design and verification), I am troubled by the extremism of some of those who are out there rightfully and thankfully fighting to ensure that free software can: 1. exist at all in the face of increasing regulation 2. coexist (I would hope that is a goal) with non-free software. Like many Linux power users I write software for a living. Whether it is in C/C++, Perl, VHDL, Verilog or anything else, what I do day-to-day is write, compile, debug and sell my services in developing software. I greatly appreciate the ability to do so and I feel that it is my right to sell my services in this form as I see fit at a price the market will bear. In no way does this mean I oppose the distribution, creation, protection and general proliferation of free software. In fact, as I mentioned above, I find myself advocating and facilitating its use frequently where it offers an equivalent or superior solution. Always making sure to account for the real costs of using free software (all software has bugs -- someone has to make sure they either do not hurt you or are fixed in a timely manner). Further I have written a great deal of code to facilitate its use (which is very much not free and strictly speaking the property of my employer -- to whom I am really selling it for the sum of my salary). To quote Richard Stallman (taken from the lwn interview): "At a deeper level, though, the biggest threat to the future of free software is the idea that non-free software is acceptable." Recognizing that historically, extremists have driven incremental change, this seems like a flawed goal for the following reasons: 1. Software development has a high cost in the form of human time taken or given from a skilled and specialized developer. Any time you are asking someone to give of their time, you are asking them to give up a portion of their existence. Some times, for the greater good, it is worth it to do so without direct compensation, other times, it is only valuable to a small group of people and should be traded for something of value (i.e. money!) 2. Non-Free software need not fear free software and vice versa. People trying to make money from solving a problem which is already solved need to fear it and move on to more interesting problems. People solving a given problem poorly and coercing you to use their flawed solution need to fear even more. 3. The more pressing concern is not that copyright and IP ownership exist but that at present anti-trust violations, inane interpretations of the laws and laughable execution by the US Patent Office allow them to be used to prohibit people from using free software to get what they want... True Type Font rendering and DVD players demonstrate this at present... I am willing to trade my time and frustration for my hard earned money, but I cannot easily do so at present because of the current litigious environment with regard to "digital rights". In short, the more useful goal would be to facilitate the free integration of free and non-free software. Non-free software will ultimately benefit from standards created by free software. Users will benefit from the ability to select from a range of solutions and pick what fits them and their needs. Free software will benefit in that it will be used and maintained as it is the foundation on which non-free and free software rely to execute. It is always nice to find tarball on the net that solves a pressing problem. It is rewarding to circulate your own "solution" in return... Further, there are tangible benefits to releasing your source to the public: a. If it becomes the standard, you can ensure that things you build upon it will work in the future b. Developers can spend more time solving more higher level problems which build upon your solution. c. Someone else may return the favor and release a solution to a future problem of yours... However, you must recognize that __everything__ you do in life comes with a cost (time, money or longevity). You have to keep this in mind what you talk about software being "free". The best answer is almost always exactly half way between the two extremes and I do not see how this is an exception... /mike

News and Editorials

Security Reports

Updates

Resources

Events

News and Editorials

New Distributions

Distribution News

Minor Distribution updates

Distribution Reviews

News and Editorials

Audio Projects

Education

Embedded Systems

Network Management

Printing Software

Web-site Development

Standards

Documentation

Desktop Environments

Games

GUI Packages

Interoperability

Office Applications

Caml

Perl

PHP

Python

Ruby

Debuggers

Press Releases:

Open Source Products

Proprietary Products for Linux

Linux Hardware

Embedded Linux Products

Products and Services Using Linux

Products With Linux Versions

Java Products

Training and Certification

Partnerships

Financial Results

Personnel & New Offices

Recommended Reading

Companies

Business

Reviews

Resources

Interviews

Miscellaneous

Resources

Events