[4216] in linux-announce channel archive
Linux-Announce Digest #508
daemon@ATHENA.MIT.EDU (Digestifier)
Sat Aug 16 09:13:06 2003
From: Digestifier <Linux-Announce-Request@senator-bedfellow.mit.edu>
To: Linux-Announce@senator-bedfellow.mit.edu
Reply-To: Linux-Announce@senator-bedfellow.mit.edu
Date: Sat, 16 Aug 2003 09:13:02 EDT
Linux-Announce Digest #508, Volume #4 Sat, 16 Aug 2003 09:13:02 EDT
Contents:
Six limitations to the current Open Source development methodology ("Frederick Noronha (FN)")
Bogofilter-0.14.5 - New Current Release (David Relson)
Happy birthday Debian... ("Frederick Noronha (FN)")
----------------------------------------------------------------------------
From: "Frederick Noronha (FN)" <fred@bytesforall.org>
Subject: Six limitations to the current Open Source development methodology
Date: 15 Aug 2003 18:30:01 GMT
========== Forwarded message ==========
==============================
Date: Thu, 14 Aug 2003 15:07:05 +0200
From: Felix Stalder <felix@openflows.org>
Subject: <nettime> Six Limitations to the Current Open Source Development Methodology
Six Limitations to the Current Open Source Development Methodology
The "Open Source Approach" to develop informational goods has been
spectacularly successful, particularly in the area for which it was
developed, software. Also beyond software, there are important, successfull
Open Source projects such as the free Encyclopedia, Wikipedia; collaborative
sites writing/publishing projects such as koro5hin.org; and the Distributed
Proofreading Project, attached to the Gutenberg Project.
However, particularly outside the software domain, the Open Source projects
remain relatively marginal. Why? Some of it can be explained by the relative
newness of the approach. It takes time for new ideas to take hold and to be
transferred successfully from one context to another. But this is only part
of the story. The other part is that the current development model is based
on a number of specific, yet unacknowledged conditions that limit its
applicability to more diverse contexts, say the music distribution or drug
research.
The boundaries to the open production model as it has been established in the
last decade are set by six conditions characterizing virtually all of the
success stories of what Benkler called "commons-based peer production." The
following list is a conceptual abstraction, a kind of ideal-type. The actual
configuration and relative importance of each condition varies from project
to project, but taken together they indicate the boundaries of the current
model. In this elaboration, I draw from examples of free and open source
software, but it would be simple to illustrate these limitation based on open
content projects.
1) Producers are not sellers
The majority professional, i.e. highly-skilled, programmers do not draw their
economic livelihood from directly selling the code they write. Many work for
organizations that use software but do not sell it, for example as system
administrators. For them the efficient solution of particular problems is of
interest, and if that solution can be found and maintained by collaborating
with others, the sharing of code is not an issue. For others employed in
private sector companies, for example at IBM, the development of free
software is the basis for selling services based on that code. The fact that
some people can use that code without purchasing the services is more than
off-set by being able to base the service on the collective creativity of the
developer community at large. From IBM's point of view, the costs of
participating in open software development can be regarded as 'capital
investment' necessary for the selling of the resulting product: services.
For members of academia (faculty and students) writing code, but not selling
(often explicitly prohibited), contributes to their professional goals, be it
as part of their education, be it as part of their professional
reputation-building. For them, sharing of code is not only part of their
professional advancement, but an integral part of the professional culture
that sustains them also economically,. in form of salaries for the faculty
and stipends for the (graduate) students.
Last but not least are all those who use their professional skills outside the
professional setting, for example at home on evenings and weekends. Having
already secured their financial stability, they can now pursue other
interests using the same skill set.
2) Limited capital investment
Particularly the last, and very important group of people, whose who work
outside the institutional framework on projects based on their own
idiosyncratic interests, can only exist due to the fact that the means of
production are extraordinarily inexpensive and accessible. Materially, all
that is needed is a standard computer (often even a substandard one would
already suffice) and a fast, reliable connection to the communication forums
of the community. Of course, the computer and the network rely on a level of
infrastructure that cannot be taken for granted in large parts of the world,
but for most people in the centers of development, they are within relatively
easy reach.
Once this access to be means of communication is secured, the skills necessary
to participate in the development of code can also be acquired
collaboratively, free of charge. The number of self-taught programmers is
significant. Since no expensive diplomas are necessary to become active, the
financial hurdle is, indeed, extraordinarily low.
3) High number of potential contributors
Programming knowledge is becoming relatively common knowledge, no longer
restricted to an engineering elite, but widely distributed throughout
society. Of course, truly great programmers are rare as truly great artists
are, but average professional knowledge is widely available. This has a
quantitative and a qualitative dimensions. Quantitatively, the number of able
programmers is in the millions, and rising. Qualitatively, the range of
people capable programmers is also unusually wide, not the least because the
material hurdles are so low and the learning can take place outside of
institutions with entry exams and tuition fees. This large and diversified
pool of talents makes it possible to create the critical mass of contributors
out of only a fraction of population.
4) Modularized Production
A large software program consists of many smaller code segments (libraries,
plug-ins etc.)This makes it possible to break down the production process
into many small steps which can be carried out by distributed contributors.
If the act of integration is relatively straight forward, it allows to keep
the amount of work that each has to contribute highly flexible and also make
use of smaller contributions (bug reports, patches). Furthermore, the
modularity of the production process allows a high number of people to work
in parallel without creating significant interferences.
5) Producers Are Users
According to Eric S. Raymond, a good open source projects starts with a
programmer scratching his own itch and finding out in the process that there
are many others with the same problem. Wanting to use a program is a great
motivation of contributing to developing it. Often, it's much more efficient
that waiting, hoping that someone will write and sell a program that will
address one's particular need.
6) No Liability
Last, but not least, software has no product liability. Paragraph 11 of the
GPL states, similar to most other licenses, that "the copyright holders
and/or other parties provide the program 'as is' without warranty of any
kind, either expressed or implied, including, but not limited to, the implied
warranties of merchantability and fitness for a particular purpose" (GPL,
v2). The absence of liability makes it possible to produce a program without
having to assign clear ownership, or other markers allowing to determine
liability, to it.
The space delimited by these condition is large and still not fully explored.
We can expect that the current open production model will find additional
niches in which it can thrive. Few could have predicted the success of
Wikipedia only three years ago, even though Open Source Software had already
been very successful at the time. However, it is also clear that many
information goods fall outside of this space. Not always are the means of
production inexpensive and readily available or the production process
modular. Sometimes, the number of potential producers is small, more often
than not are the producers not the users of their own products, and, in many
cases, product liability is desirable.
This does not mean that the "open source model" cannot apply to, say, the
production of literary works, music, or medical drugs. What it means,
however, is that to make it viable, another round of social innovation is
required. This is slowly happening. The growth of "Open Access Journals" or
discussions around "compulsory licensing" are good, though very early
examples.
- ----+-------+---------+---
http://felix.openflows.org
# distributed via <nettime>: no commercial use without permission
# <nettime> is a moderated mailing list for net criticism,
# collaborative text filtering and cultural politics of the nets
# more info: majordomo@bbs.thing.net and "info nettime-l" in the msg body
# archive: http://www.nettime.org contact: nettime@bbs.thing.net
==============================
##########################################################################
# Send submissions for comp.os.linux.announce to: cola@stump.algebra.com #
# PLEASE remember a short description of the software and the LOCATION. #
# This group is archived at http://stump.algebra.com/~cola/ #
##########################################################################
------------------------------
From: David Relson <relson@osagesoftware.com>
Subject: Bogofilter-0.14.5 - New Current Release
Date: 15 Aug 2003 20:30:05 GMT
Bogofilter is a mail filter that classifies mail as spam or ham (non-spam)
by a statistical analysis of the message's header and content (body). The
program is able to learn from the user's classifications and corrections.
The statistical technique is known as the Bayesian technique and its use
for spam was described by Paul Graham in his article "A Plan For
Spam". Gary Robinson, in his weblog Rants, suggests some refinements for
improved discrimination between spam and ham. Bogofilter's primary
algorithm uses the f(w) parameter and the Fisher inverse chi-square
technique that he describes.
Bogofilter is run by an MDA script to classify an incoming message as spam
or ham (using wordlists stored by BerkeleyDB). Bogofilter provides
processing for plain text and html, supports multi-part mime message with
decoding of base64, quoted-printable, and uuencoded text and ignores
attachments, such as images.
Bogofilter is written in C. Supported platforms: Linux, FreeBSD, Solaris,
OS X, HP-UX, AIX, ...
******* ******* ******* ******* *******
Bogofilter version 0.14.5 has been released on SourceForge,
http://sourceforge.net/projects/bogofilter.
A couple of parsing problems have been corrected and a new switch added for
use by scripts.
See files NEWS-0.14 and RELEASE.NOTES-0.14 for the details.
=================
BOGOFILTER NEWS
=================
0.14.5 2003-08-15
* Corrected parsing error (in html code) that caused
bogofilter to miss message separators.
* Revised processing of From and empty lines so that parsing
works correctly with both flex-2.5.4 and flex-2.5.31.
* Added '-T' as terse mode (with fixed formatting).
* Updated bogominitrain.pl to version 1.3.
* Revised FAQ.
0.14.4 2003-08-10
* Revised database API so that there are 3 distinct layers
(program, datastore, and database) with a clean interface
between them.
* Correct exitcodes in bogoutil by using EX_ERROR.
* Updated FAQ.
* Fixed token registration bug in 0.14.x versions.
* Fixed seg fault caused by database lock contention.
0.14.3 2003-08-05
* Fixed critical locking bug introduced into bogofilter 0.14.0 with
the combined-wordlist code: when working with separate wordlists,
bogofilter would lock only the first one opened, rather than all.
* Documentation updates.
* %g formatting is now supported by bogofilter's formatting functions.
* Merged trio 1.10 (http://ctrio.sourceforge.net/) to support
compilation on ancient systems (Solaris 2.5) that do not have
[v]snprintf functions.
Trio is Copyright (C) 1998-2000 Bjorn Reese and Daniel Stenberg.
* Various documentation updates, including the FAQ.
* The test suite was adjusted for older grep variants (Solaris 2.5)
that don't cope with long lines.
* Print database version in print_version().
* Postfix integration instructions have been upgraded.
* Debug output for wordlists and databases was enhanced.
0.14.2 2003-08-02
* Replaced use of memcpy() by memmove() in an input routine. The
overlapping copy migh cause data corruption on some systems.
* Fixed "make check" failures for bogoutil introduced with the
"combined wordlist" feature in 0.14.0. There has been a buffer
overflow. All users of bogofilter with combined wordlist prior to
0.14.2 are advised to upgrade.
* Fixed bogus "t.valgrind" test FAILures.
* Fixed uninitialized data in db_get_dbvalue(), for split word lists.
* New file, contrib/vm-bogofilter.el, provides an interface
between the VM mail reader and bogofilter."
* Revised lexer_v3.l for compatibility with flex-2.5.31
* Break up long line in regression test input for Solaris 2.5
compatibility.
0.14.1.1 2003-08-01
* Fixed check for adding spam_subject_tag to Subject: line.
* Correct problem with t.degen regression test.
* Updated French version of FAQ.
0.14.1 2003-07-31
* Implemented named exitcodes, with Unsure having its own value (2)
and changing the value for error from 2 to 3.
* Initial release of token degeneration code.
* Revised lexer pattern to better recognize encoded tokens.
* Updated English version of FAQ.
0.14.0.1 2003-07-23
* Fix problem with encoded text.
* Fix handling of absolute paths.
* Fix defect in base64 decoding that can cause segfaults.
* Bogoutil now complains before exiting when it can't open a
file.
* Updated bogominitrain.pl to work with combined wordlists.
0.14.0 2003-07-22
* Initial release of code allowing bogofilter to use a single,
combined BerkeleyDB database for storing both ham and spam tokens.
The file is named wordlist.db
* Default wordlist mode is single, combined wordlist.
File wordlist.db contains all spam and ham tokens.
* Bogofilter and bogoutil detect whether one or two wordlists are in
BOGOFILTER_DIR and use the appropriate wordlist mode (combined or
separate).
* Added tdb (trivial database) support.
* Decode encoded text in header lines.
* Updated contrib/bogominitrain.pl prints more info and can save
messages used in training.
* Bogofilter's -V output now includes algorithm and database info.
* Miscellaneous documentation updates.
##########################################################################
# Send submissions for comp.os.linux.announce to: cola@stump.algebra.com #
# PLEASE remember a short description of the software and the LOCATION. #
# This group is archived at http://stump.algebra.com/~cola/ #
##########################################################################
------------------------------
From: "Frederick Noronha (FN)" <fred@bytesforall.org>
Subject: Happy birthday Debian...
Date: 15 Aug 2003 21:55:08 GMT
***********************************************
Debian is 10 tomorrow
Saturday sees the tenth anniversary of the foundation of the Debian
project - over 1,100 registered free software developers, over 12,000
binaries and the distribution is still going strong. There are
parties taking place all over the world, with the UK event taking
place in Cambridge. It's all back to Steve McIntyre's place - but it
looks like accommodation on his floor may be limited, so it might be
a good idea to book a local bed and breakfast establishment...
http://the.earth.li/~huggie/cgi-bin/moin/Debian10thBirthday
_______________________________________________
_______________________________________________
LinuxUser mailing list
LinuxUser@www.linux.org.uk
http://www.linux.org.uk/mailman/listinfo/linuxuser
##########################################################################
# Send submissions for comp.os.linux.announce to: cola@stump.algebra.com #
# PLEASE remember a short description of the software and the LOCATION. #
# This group is archived at http://stump.algebra.com/~cola/ #
##########################################################################
------------------------------
** FOR YOUR REFERENCE **
The service address, to which questions about the list itself and requests
to be added to or deleted from it should be directed, is:
Internet: Linux-Announce-Request@NEWS-DIGESTS.MIT.EDU
You can submit announcements to be moderated via:
Internet: linux-announce@NEWS.ORNL.GOV
Linux may be obtained via one of these FTP sites:
ftp.funet.fi pub/Linux
tsx-11.mit.edu pub/linux
sunsite.unc.edu pub/Linux
End of Linux-Announce Digest
******************************
