[ox] Meeting at WOS - some thoughts and comments
- From: Bettina Berendt <Bettina.Berendt educat.hu-berlin.de>
- Date: Tue, 3 Aug 1999 10:05:37 +0200 (MET DST)
Hello again,
in addition to the protocol, Holger and I also noted some thoughts and
comments.
So what do YOU think?
Cheers,
Bettina
-----------------------------------------------------------------------------
Bettina Berendt berendt educat.hu-berlin.de
Humboldt-Universitaet zu Berlin http://www.educat.hu-berlin.de/~berendt/
Abteilung Paedagogik und Informatik Fax [PHONE NUMBER REMOVED]
Geschwister-Scholl-Str. 7
10099 Berlin
-----------------------------------------------------------------------------
*Afterthoughts* (Holger Blasum, holger nerdbank.org)
===============
The discussion has revolved around Rishab Ghosh's assumption
that at leas part
of the open source economy is caused by authors who trade
their work for the
economic good reputation. This trade-off is covered well by
the economic model.
Deviating from this paradigm, it has been shown that there are:
- - authors who can directly reuse the (copyrighted) publications to
mailing lists for other paid publications (market analysts)
[* Berendt:
This refers to what I called "advertisement" above *], so
they do not
even treat their input as open content
- - authors who are tied by non-reputation-related motives,
e.g., "love for
freedom". It was not settled in the discussion whether these less
quantifiable motivations should have been included in the
economic model.
[* Berendt: This is related to my remarks on a different
definition of
the word "economic". There are many economic models who do
just this -
include things like "love for freedom", and even "altruism". *]
Although the discussion occasionally reverted to the unsolved
free rider
problem, it now appears to us (Blasum and Berendt) that its positivist
overtones have totally ignored limits of the open source model, e.g.,
software for groups that programmers are not so much interested in.
This includes software for beginners (WYSIWIG editors [* Berendt:
and generally programs
with a large emphasis on the user interface, a point mentioned in the
O'Reilly "Open Source" booklet *]), blind people, etc.
*Afterthoughts* (Bettina Berendt, berendt schulweb.de)
===============
Open Source software as a byproduct of commercial activities
- ------------------------------------------------------------
In terms of Economics,
"Open Source software is a byproduct of commercial activities"
seems to mean that Open Source is a (positive)
external effect of standard commercial activities in the
software market.
Just like air pollution is a negative external effect of
standard commercial
activities in the energy and transport markets. This would
explain why a lot
of the Open Source economy works.
This is also consistent with Ghosh's Economic model (if I
understand the model
correctly). Software developers earn a living somehow, and as
a byproduct of
what they do and enjoy doing, they develop things which may
be useful for others
too. Moreover, it doesn't cost them anything to provide these
things to others
(low marginal costs of distribution), and they may get
attention/comments/
reputation from it. If this is the case, I am not too
surprised that Open
Source works.
However, I doubt that large software products
(something like, say, Emacs)
can be the byproducts of specific, customised software. Large
projects to develop large products usually
develop their own dynamics, require a lot of dedication, and
cannot easily
be composed of a zillion little modules that happen to have
been byproducts
of a large number of diverse people's work (at least, this is
my observation).
In other words, I think that a
lot of high-quality Open Source software requires the
concentrated effort
(and large parts of the working time) of one or more
individuals. And with
respect to these, Bill Gates' 1976 question "and who is going
to pay the
programmers" has not been answered. I do not want to defend
Microsoft's policies, but I do think he has a point here.
Saying that "programmers do it, so they must have a rational
reason for it"
(and they are not exploiting themselves, etc. etc.)
seems to me to be a valid motivation to start an Economic
enquiry, but not
a valid reason to justify why the system is working.
This would neglect many interesting insights of Economic theory
- - that the neoclassical, rational "homo economicus"
is at best an idealisation to start with, that imperfect
information on the
one hand and institutional factors including altruism, family
or peer group bonds,
trust, etc., are fundamental for the understanding of much of
economic activity,
or human activity in general. [?Or am I
misunderstanding/misrepresenting you, Rishab?
But then I think we may have different definitions of
"economic" anyway: From Holger's
notes, I quote you "Economically, this is explainable by
reputation which
translates into value. In the Economic model, we cannot
include other motivations,
e.g., ideology ..." ?]
An interesting question in this context seems to be: Yes,
there is a fundamental
difference between Open Source programmers and other groups
of people who work
a lot for little or no money (let's say, unskilled manual workers).
Open Source programmers have skills for which there is a
lucrative market,
so it seems that they really have a choice (in contrast to
the unskilled
manual workers). Is that true?
Anyway, it seems to
me this is an empirical question: Who are the people who
publish Open Source
software, how much of their time does this take up, how do
they make a living,
what kinds of software are they writing (a tool to convert
file format xyz
into file format abc, or Emacs?), who organises the large
project their software
may be part of, etc.
I have read the little book "Open Source" published by
O'Reilly, which hints at the large diversity of existing
organisation forms. However, it has not answered all of
my questions. It probably can't, because it's
such a short book; and there probably is enough literature,
but I simply don't
know it. Does anyone have good recommendations?
Open Source and Science
- -----------------------
This topic was not very popular at
WOS. In the workshop, for example, it was cut short by a
remark to the effect
that "much of what appears in scientific journals is useless".
This may be a legitimate opinion, but it may also be
a wasted opportunity: The question
"are Open Source principles transferable to other areas"
comes up again and
again, and this area is simply excluded.
I had been looking forward to Friedrich Kittlers talk at WOS,
which seemed to promise a discussion of the commonalities and
common interests of Open Source and science.
But it seemd he was after something else (hardware?), or else I didn't
understand his point.
Therefore, I will simply list points that have been made in the
various discussions, and my own ideas, here:
- - Commonalities:
- scientific publications must be "open source" (either
include the "source"
in the publication itself - e.g., a mathematical proof,
or be prepared to
release it - e.g., experimental data in psychology,
program to evaluate
results in cognitive modelling). A publication saying
"this is what I found
out, but I am not going to tell you how it's done" is
simply not scientific.
- comments are the currency (scientists speak at
conferences, publish papers
to get comments) / economy of reputation:
a newsgroup article that is forwarded corresponds to an
article which is cited
- economy of attention
- high intrinsic motivation, enthusiasm of participants
- marginal costs of distribution very low
- no payment in exchange for a particular piece of writing
- sometimes, the act of filtering creates the value of a publication
(review articles)
- the Internet as a forum of publication / dissemination of
information
- - Differences:
- payment: Scientists are usually paid for the time they
work and indirectly
for the success they have had in the scientific
publications market
(Re-employment is, among other things, based on
publication lists).
- - Difference or commonality?:
- widespread exploitation and self-exploitation, unpaid overtime is
the rule in science.