tiemann@AI.MIT.EDU (Micheal Tiemann) (01/18/90)
Many people send me little pieces of code here and there to fix
problems with installation, configuration, and sometimes operation of
the compiler. For all that, I am grateful.
The Free Software Foundation has very strict requirements on the code
that they can distribute: it must be free software. If somebody sends
me a whizzy feature for GNU C++, but they fail to provide the
neccessary for FSF to distribute it, it will never be officially
distributed, and that work will be lost. Here is the general policy:
Return-Path: <rms@ai.mit.edu>
Date: Wed, 17 Jan 90 11:47:58 EST
From: rms@ai.mit.edu
To: tiemann@ai.mit.edu
In-Reply-To: Micheal Tiemann's message of Wed, 17 Jan 90 07:20:11 EST <9001171220.AA12407@apple-gunkies.ai.mit.edu>
Subject: GNU Papers
If <Mr X>'s changes are just a few lines here and there, then we don't
need papers from him. If they are between 10 and a few hundred lines,
then a disclaimer is sufficient. Otherwise, we prefer an assignment,
but a disclaimer will do.
In addition to the disclaimer or assignment from the author, we need a
disclaimer from the employer if appropriate. We don't need this if
the changes are so small that we don't need papers at all.
You can ask for the papers yourself. Send the person a copy of either
assign.changes or disclaim.changes from /a2/rms/gnuorg/. If you send
assign.changes, go to the second page and fill in the name of the
person and the program name, such as "the GNU C compiler". These
files contain full directions. Later on, ask me whether the papers
have arrived.
Therefore, remember that when you send me changes, I may turn around
and immediately ask for papers. THIS IS NOT MEANT TO HASSLE YOU! It
will probably save you time in the long run to have your changes
merged, legally, then to have to patch the release every time there is
a new one. It takes more time to get the signature at first, but I
have been doing that for 3 years, and it gets real easy after the
first time.
Michael