Subject: Re: Competition by internal expertise for F/OSS vendors
From: Thomas Lord <>
Date: Wed, 03 Sep 2008 12:55:19 -0700

Stephen J. Turnbull wrote:
> Tim O'Reilly writes:
>  > And also FWIW, I've been castigated as recently as last year by the  
>  > FSF for saying that.
> This is not a surprise. 

That isn't what Eben castigated Tim for.  Eben castigated Tim
for helping to invent and promote the concept of "open source"
and accused him of profiting from the problems that effort

>  In recent threads on emacs-devel RMS has
> revealed that he has never installed a GNU/Linux system himself, and
> has argued vehemently against use of infrastructure (bug trackers,
> version control) which use the web (as opposed to email) heavily.  As
> with going to a Catholic priest for marriage counseling, given the
> right person you may get good advice -- but it sure isn't founded in
> experience.  I dunno about Eben Moglen, but RMS clearly has badly
> fogged vision when it comes to the capabilities and working of the
> web, 1.0 or 2.0.

In the video in question Eben describes services like
Google as (paraphrased): "the largest private intelligence
service / secret police ever assembled."  Or words to that
effect.    I think that they have a fairly clear picture of the
capabilities of "web 2.0" and their implications.

To be sure: RMS is not well up to speed on web-related tech.
Don't go to him for help with your XQuery problems.
And I would agree, though you didn't quite say it, that that
muddles his tactical thinking on technical issues for the
nominal GNU project.   The *strategic* thinking, though,
isn't that bad and Eben lays it out in that video Tim linked

The strategy:  engage in diplomacy with those service
providers and push as hard as you can on all forms of
truly personal computing in software freedom.

Rationale:  In 10 years, the number of people with access
to personal computing hardware is going to skyrocket relative
even to today's numbers.  That is where the main action.
In 10 years, the ad-broker model will have broken when
the price-bubble on ads bursts and also when users get fed
up with the privatized secret police starting to emerge out
of the muck -- and given all of that available *personal*
computing, the community will have ample room to work
around those scary services.

>  > What is the right battle to fight for user freedom today?
> Well, you say you've seen this coming for a decade (which seems
> plausible to me).  What do *you* think?
> If it would help to have a more precise query: It looks to me like an
> Open Software Service Definition is something whose time has come,

No, it hasn't.

That needs to come from the engineering community and it
needs to be a norm of professional behavior.

Look, it's a norm of professional behavior in programming that
professional programmers do not casually installing untested code
on life-critical systems.    If you got known for doing that you'd
rightly be excluded from a lot of professional jobs, people would
not be rude to criticize your professional conduct, etc.   So, that
shows that the profession *can* have grass-roots norms.

What we need in a "[reasonable person] Service Definition" is
a code that operates as a code of conduct for the engineers that
actually build services.   Let me unpack that a little, with a
quick analogy:

For several years, the Usenix organization's system administrator
branch (SAGE) made a big deal about their admin code of
ethics.   This had a specific economic impact:  if a pointy hair
boss came to an admin and said "please print out all of
joe's email so I can read it", and the admin was a SAGE member
that took that stuff seriously, the admin could say "no," point
to the code, and count on the support of SAGE in defending
against any retribution.   That's an example of the technocracy
being the fourth column.

We need "stuff like that" at a higher level, for programmers:
generally agreed upon principles about what kinds of coding
to NOT do, even for money, even under threat from a boss.

I don't see the potential for that yet.   There's too much
namby-pamby "open source" pseudo-libertarian gibberish
going around at the moment.