Subject: Re: Open Source and Government agencies
From: Frank Hecker <>
Date: Sun, 27 Feb 2000 13:39:17 -0500

"Kevin S. Van Horn" wrote:
> 1. Public domain software *is* free software.  You can't get any freer than
>    public domain, which places no restrictions whatsoever on you.  And PD
>    software is guaranteed to remain free, forever (although derived works
>    might not remain free).

Agreed about the "no restrictions" aspect of public domain software, but
it appears that the "although..." part appears to be a factor in
Asquith's thinking: "I am very concerned about releasing under public
domain as I do not want the code to fork or to be sucked up by a
closed-source company as has been done in the past with government

> > if Mr Asquith wanted to further ensure that that [GPL'd]
> > version remained the most used and useful version, he could release his
> > future (public domain) changes and extensions to the public _only_ in
> > the form of patches against the free software version,

Actually, I wasn't thinking here only of the GPL, though that is
certainly a license that would discourage creation of proprietary
derived works, and I think that was part of Asquith's motivation in the
first place.

> There are very successful open-source software projects that are close to being
> public domain, given the liberality of their licenses.  Apache is one that
> comes to mind.  It hasn't needed a GPL-style license to remain "the most used
> and useful version".  The advantages of bazaar-style software development over
> cathedral-style development have ensured that no proprietary derivative of
> Apache has ever gained a widespread following.

Your point is well taken. That's why I think the most important thing
for Asquith is not worrying about the exact licensing per se but rather
working with others to build a healthy developer community around the
software, with that developer community motivated to maintain the
software as a non-proprietary code base available to all. I suspect that
the lack of such a community accounts for some of the prior instances in
which government-created code eventually became most commonly available
as proprietary products.

However I still don't think that licensing is a null issue in this case,
because I think one of the motivating factors behind creating and
sustaining such a community is often the political desire to preserve a
body of code as a nonproprietary resource, and that finds its expression
in choices of licensing. Hence I'm not going to tell Mr Asquith that he
should stop worrying about free software licenses vs. releasing code
into the public domain.

However I personally believe that given the fact that Asquith's work is
funded by U.S. taxpayers, if he does work with others to release the
software under a free software license then I think they should choose a
license with as few restrictions as possible, say, something like the
MIT X license as opposed to the GPL.

Frank Hecker            work:        home: