This weekend's witch-hunt

[Rich Felker]

On Mon, Sep 18, 2006 at 11:52:43PM -0400, Mark Richards wrote:
> If someone has an objection to a change in the license terms for code 
> that they wrote,

As you'll see from the forensic analysis, there is no significant code
which Bruce wrote.

> then my legally-uneducated view is to respect the 
> request and move on.

Bruce explicitly licensed the code he wrote (which was mostly removed
long ago) under GPL v2 or later, AT THE LICENSEE'S OPTION. That means
that if someone who received it makes a derivative work, they are free
to choose whether they want to use GPL v2 only, "v2 or later", or a
particular later version. They absolutely are not required to make the
derivative available under "v2 or later".

The whole point of free software is that you do not have to beg for
permissions and follow the changing whims of the copyright holder like
you might with old 'freeware' programs that said "contact the author
if you want permission to modify the program..." The permissions are
all permenant and irrevocable. Bruce can't come back 10 years down the
line and say "oh I meant 'v2 AND later' not 'v2 OR later'" like he's
trying to do now (well he can say it but it's irrelevant).

> Why is this simple request from a person who wrote some old code so 
> difficult?

The original request was made politely and I'd actually hoped Rob
would consider it. After that, all of Bruce's posts were full of
anti-GPLv2 FUD, pro-GPLv3 propaganda, threats, and ultimata. What he
succeeded in doing was convincing me that GPLv3 is a horrible idea.
The proper response to such childish behavior is to ignore the
person's "requests" (ultimata) and make them irrelevant. Rob may be
going a little over the top with the audit, but I think it proves the
point quite nicely and hopefully won't "waste" too much time.

Besides that, how would you propose to follow Bruce's request? If it
takes a whole weekend of auditing to determine what (little) code
Bruce actually holds copyright on, how is "The code written by Bruce
Perens may be used under later versions of GPL as well as GPLv2..."
feasible at all? Obviously it's become heavily intertwined with other
code (this is BB 1.x, a derived work, not the original) and it would
be impossible to license "just Bruce's code" under "v2 or later"
without also affecting large amounts of other code (written by Rob and
other authors) which he does NOT want to license under v3.

Rich