<div dir="ltr">Hi Folks,<div><br></div><div style>In a recent thread in the openstack legal-discuss mailing list [1], we determined that copyright headers in source files are not legally necessary to preserve the original copyright. In addition, copyright headers in OpenStack projects right now are kind of a mess, resulting in confusing, inaccurate, and at times unwanted copyright assignments.</div>
<div style><br></div><div style>My main interest in starting this conversation is for developers and reviewers to no longer have to spend any time maintaining or otherwise concerning themselves with copyright headers. To that end, I offer the following proposals for cultural and procedural changes.</div>
<div style><br></div><div style>CULTURAL: Let's encourage our fellow contributors to stop adding copyright headers to new files, mostly by example, but possibly also by a few gentle -1s during review. If a contributor insists that their employer's legal counsel won't allow them to submit code without a copyright notice, suggest that they instead put it in a per-project NOTICE file (after suggesting they consider hiring a different lawyer :-) Note, no consensus is required for individuals to just stop adding their own headers!</div>
<div style><br></div><div style>PROCEDURAL (two independent parts):</div><div style>A) Start gradually removing existing copyright headers as time and permission from the copyright holder allows. This is a 90/10 problem; we can make a huge amount of progress by talking first to the OpenStack Foundation, then to a handful of other companies. But past that, progress will be slow and probably not worth the effort. [2]</div>
<div style>B) Add a hacking check to prevent new copyright headers from being added. In order to accomodate any existing copyright headers not already cleaned up, we probably have to add some sort of pylint-like "disable" comments to short-circuit the hacking check.</div>
<div style><br></div><div style>So, what do you think? Is this a bad idea? If so, why do you hate me? Did copyright headers once save your kitty from a burning building? Are you worried about consistency? If so, please explain why you believe consistency is useful in this specific case to avoid a choice Emerson quote.</div>
<div style><br></div><div style>Curmudgeonly,</div><div style>markwash</div><div style><br></div><div style><br></div><div style>[1] - <a href="http://lists.openstack.org/pipermail/legal-discuss/2013-May/thread.html">http://lists.openstack.org/pipermail/legal-discuss/2013-May/thread.html</a></div>
<div style>[2] - Try this tortured one-liner to see the distribution of copyrights in your favorite project. Some projects are going to be easier to clean up than others.</div><div style><br></div><div style>git grep -i copyright | perl -pe 's/copyright//i' | sed 's/.*:#//' | sed 's/([cC])//' | sed 's/[0-9][0-9][0-9][0-9]\(-[0-9][0-9][0-9][0-9]\)*,*//' | sed 's/^ *//' | sed 's/[. ]*$//' | sort | uniq -c | sort -gr | head -20</div>
</div>