Re: ncurses 4.1

From: Eric S. Raymond <>
Date: Mon, 2 Jun 1997 10:30:46 -0400 (EDT)

> If ncurses is free software, that means that Dickey (or anyone else)
> is entitled to make his own version and control it.
> However, you can continue to develop and release your version, and
> no one but you can control that.
> So if "it" refers to ncurses in general, who controls it? No one.
> No one can control a free program in this sense.

Under your GPL, this is technically true.

However, hacker-community custom forbids copying someone else's live
project and haring off with it in a different direction. There are
several good reasons for this, including avoiding duplication of work
and preserving authors' perceived right to control the representation
of work in which their reputation is invested.

Accordingly, the community tries hard to avoid having multiple parallel
developments of the same code going on. To prevent the kinds of unresolvable
clashes that would lead to project groups fissioning all the time, there
are informal but strong customs which amount to giving authors a kind of
homesteaders' property right over a particular corner of free software's
coonceptual space; a right they may pass, delegate or share with others
but which outsiders will feel social pressure not to infringe.

This right, in the case of old and well-established projects, can be
very strong. It's pretty hard to imagine anybody forking most of the
GNU utilities, for example, even though GPL technically permits this.
The community would blow them off; it would not accept the results.

Accordingly, there is a fairly well-defined notion of "primary maintainer"
for a project; the "primary maintainer" is the person who holds the
hacker community's trust to steer a project. There are responsibilities
associated with this position, and it can be compromised or lost (the
trust can be eroded or destroyed entirely) if the holder fails to
fulfil them.

What I am trying to show is that Thomas Dickey (a) never legitimately
became the ncurses's project primary maintainer under generally
recognized custom, and (b) has failed in the ethical responsibilities
of a project maintainer.

		Eric S. Raymond
Received on Mon Jun 02 1997 - 12:26:58 EDT

This archive was generated by hypermail 2.2.0 : Mon Dec 19 2011 - 06:24:16 EST