Subject: Re: Why Open Source Sucks for the Consumer
From: "Stephen J. Turnbull" <turnbull@sk.tsukuba.ac.jp>
Date: Fri, 1 Sep 2000 15:53:53 +0900 (JST)

>>>>> "Ben" == Ben Tilly <Ben_Tilly@trepp.com> writes:

    Ben>   - inline skates - frisbee - transistor radios -
    Ben> pagers/cellular telephones - cars

    Ben> Excepting pagers/cellular telephones, most of these show
    Ben> little potential for becoming basic technologies that are
    Ben> used as a foundation for future waves of technological
    Ben> progress.

    Ben> By contrast consumer software such as spreadsheets, word
    Ben> processors and browsers, routinely do.

The fact that you don't have to strip the skin off software is an
artifact of it being soft.

The point is that not very far under the skin of all the hardware
consumer goods (but the frisbee, which has no "inside") you mention
are techniques that were innovations and have become universal in
their respective industries and foundations for future development.
In the case of the automobile, most of the U.S. economy is "technical
progress" built on the automobile, hood and all.  (You may choose to
call the shopping mall and the multilevel cloverleaf interchange
"regress"; they still are inventions.)

    Ben> There is a reason that I said that either *companies* will do
    Ben> it or it will not happen.  The reason I specifically did not
    Ben> say OSS is that I had in mind the same observation about OSS
    Ben> developers that you just made.

Ah, OK.  But then the companies will have control.  They will not take
decisions with the best interest of OSS uppermost.  Rather, they will
go where the profits are.  And that's extremely problematic.  (See
below on the definition of FSB, in particular.)

So you're just saying you think that a certain forward-looking
behavior is more profitable?

    Ben> But some (notably Red Hat) funded a proactive response.
    Ben> Whether or not Gnome or KDE ultimately "wins", the mere fact
    Ben> of free competition demonstrably put enormous pressure on
    Ben> Troll and changed their licensing.

"Proactive?"  IIRC, GNOME was "oh, yeah, you _can_ do desktop free;
let's run and catch up to the semi-free company."  GNOME was a
grassroots thing, Red Hat funding came later.  (Desktop leaves me
cold, like frozen, so my memory is very weak on this strand of
history.)

The KDE project was _pro_active, start a project and put resources
into so that we'll have a desktop and prevent future clashes of DnD
protocols and object brokering etc---and you see where that led.  (And
even there, it was reacting to the success of Windows.)

If GNOME is what you class as proactive, I have zero objection to it.
But I don't think it's a big change in the "100 drummers" OSS
development and marketing methodology, nor much of a preventative.
Just a natural extension of what was already happening.

    Ben> But at a given moment in time it can make sense [to label a
    Ben> company as "doing X"].

Sure.  My point was just that getting rid of all the companies with
that label leaves a vacuum which very likely will generate extra
profit for the next entrant to put on that hat.  We have to think
dynamically here.

    Ben> Proprietary firms are ones that think they can get more by
    Ben> keeping tight control.  FSBs are companies which think they
    Ben> can do better by nominally giving it away up front.  The two
    Ben> differ on tactics, not on the end goal.  (Being nice can also
    Ben> help you get ahead.  Being perceived as ruthless may be a bad
    Ben> idea.)  If the FSB is not as good at figuring out how to get
    Ben> honey, then that is their problem.

And what about tomorrow?

To be an FSB is not a strategic decision that can be changed, albeit
at some expense in reputation: today we'll GPL v1.0, tomorrow we'll
sell an exclusive license to v2.0 to Microsoft, the day after that
we'll release v3.0 under MPL.  On the third day, they do not rise
again.

Nor, at least in some people's eyes, can you pick and choose which
products to release free: OK, Cygnus works on GCC and other GNU
programs, and makes great effort to ensure that the code gets out
under GPL, and thus is a FSB.  But they distributed Code Fusion under
a proprietary license, and so they really weren't.

By the way, I'm quite sympathetic to your statement.  It's actually
the way I think of FSBs myself.  I certainly include Cygnus, and
Aladdin/Artifex, as part of some kind of "FSB community", and am
honored to be associated with them through this ML among other venues.
But they aren't FSBs by the common definition.  At best they're
wannabes.  No?  (In the case of Cygnus, most people would argue that
the Code Fusion stuff is a small enough part of their business to be
negligible, I suppose---which just proves my point.)

Furthermore, your description of fairweather FSBs (they don't desert
in a storm, they drown heroically of course) is not very inspiring for
non-FSB OSS projects.  Why should an OSS project pay any attention to
the FSBs, then, if they're at best unreliable code contributors which
may go under tomorrow, and at worst pure free riders?

-- 
University of Tsukuba                Tennodai 1-1-1 Tsukuba 305-8573 JAPAN
Institute of Policy and Planning Sciences       Tel/fax: +81 (298) 53-5091
_________________  _________________  _________________  _________________
What are those straight lines for?  "XEmacs rules."