Copyright years in source code

[Richard Laager]

On 5/8/20 1:10 PM, Gary E. Miller via devel wrote:
> I think the year of first publication still has some use as it disambiguates
> which version of copyright law applies.

The "year of first publication" applies per copyrightable thing, so if
the file has multiple changes, you'd need multiple years. For example,
imagine you had a book written in 1900 with changes made in 2000. If you
say "Copyright 1900", it looks like it is now public domain, but it's
not. This is more-or-less the reason for years in copyright statements.

Essentially all computer code is in the modern era of copyright. The
copyright terms are all so long now that you can basically assume that
code is copyrighted for as long as it remains relevant to use. And for
an actively maintained project, something is changing every year, so the
project _as a whole_ would always be under the current copyright regime
anyway even if the terms were shorter. So the years have essentially no
practical value; you can just assume it is "copyright this year".

If terms were shorter _and_ someone wanted to pick out portions of the
code for reuse elsewhere once the term ended _and_ the copyright
statements were perfectly maintained, they might in theory be able to do
so. But even if the copyright statements were perfectly maintained,
you'd still have the issue that routine refactoring, churn, and cleanup
work would mean that most files would have a recent copyright (if not
the current year), so you'd still have to hit the VCS to determine the
history on the code you want to reuse (e.g. some core algorithm). So
even in that scenario, the years don't seem to help.

-- 
Richard

-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 833 bytes
Desc: OpenPGP digital signature
URL: <https://lists.ntpsec.org/pipermail/devel/attachments/20200508/2a779c52/attachment.bin>