The above-quoted advice in maintain.texi
A big part of why this so-incredibly-annoying issue never goes away,
IMHO, is that that node in maintain.texi has accreted over decades of
different advice and scenarios, never clearly distinguishing between
them. Sigh.
Since gunzip is a visually perceptible copy of an individual work (in
the sense of U.S. copyright law), its copyright notice is supposed to
reflect the last copyrightable change to that individual work, not the
last change made to anything in the gzip package
Well, that was my own long-time understanding too. But a few years ago,
rms/lawyers came up with this different scheme (I was not involved in
its creation), which is this paragraph in maintain.texi, which I know
you are well aware of, and which directly contradicts your statement:
(Here we assume you're using a publicly accessible revision control
server, so that every revision installed is also immediately and
automatically published.) When you add the new year, it is not
required to keep track of which files have seen significant changes in
the new year and which have not. It is recommended and simpler to add
the new year to all files in the package, and be done with it for the
rest of the year.
To my mind, that clearly states that you are allowed (not forced) to
update any file in a package as soon as the package has changes in a
given year, regardless of whether the file had copyrightable changes on
its own or whether the file is an "independent work" on its own. Does
it make "logical" sense? Not really, not to me. But who am I to say?
Regardless, I don't wish to argue against it, and don't see any harm in
it, despite the illogic, because of yet another fact. I quote the SFLC
lawyer (Aaron Williamson) who replied to me most recently about this
issue:
More importantly, none of this much matters because *notice is not
required for copyright protection.*
As we all know, everything is copyrighted instantly upon publication, in
this day and age.
The reason why rms wants years in the copyright notices is so people can
know when code enters the public domain. Fine. But the only way anyone
is going to know what *specific* code has become pd (assuming copyright
terms aren't extended indefinitely, a very big assumption IMHO) is to
have a copy of the file from that olden-times year, retrieved from a
repo or an old distribution or whatever.
So there's really no use in worrying about this, as far as I can see.
# This program is distributed as part of gzip, which is
# Copyright (C) 2010 Free Software Foundation
If you want to make that change, go right ahead.
karl
