Re: [Groff] License for files with »ideal« parts

[Ingo Schwarze]

Hi Bernd,

Bernd Warken wrote on Mon, Sep 15, 2014 at 09:24:59PM +0200:

> This is about MIT licenses for `ideal' files.

These files almost certainly do not belong to the MIT.

>From http://www.users.drew.edu/~cvanwyk/cvwvita.htm ,
my impression is that the author wasn't ever associated
with the MIT in any way.

> In `contrib/gideal/files', I stored an `ideal' document
> `opaque_MIT.ideal.ms', which is available as
> <http://web.mit.edu/usrdoc/ditroff/ideal.man/opaque>
> (so I changed the file name).

Gah.  You shouldn't republish random files you find on the
Internet without first figuring out whether you are allowed
to do so.

> Other documents in <http://web.mit.edu/usrdoc/ditroff/ideal.man>
> are also useful for `ideal'.

Frankly, I doubt that the MIT has any rights to redistribute these
particular files.

How did they end up there?  My impression is that the files below
/usrdoc/ are ancient documentation provided for users of the MIT
Project Athena more than two decades ago, then forgotten, and soemhow
found their way to the Internet.  It looks like the people who
originally put them up weren't quite aware what Copyright is all
about.

Obviously, these files have been assembled from various sources.
I see MIT stuff in there, AT&T, IBM, BSD...  Under various licenses,
quite a bit of it free, many without any licenses whatsoever, some
parts explicitly non-free.

The ideal files you are inquiring about have been put into a
subdirectory of Kernighan's device independant troff, which makes
some sense because Kernighan and Van Wyk both worked at Bell Labs
before that time.  From heresay, device independent troff was
included in the Caldera release of Ancient Unix; but that doesn't
mean everything somebody cobbled together with some version of
device independent troff is covered by the Caldera license, too.

> Does anyone know which license is valid for these documents
> or where I can get this information?

The task is to find some distribution of device independent troff
that has an unambiguous license and includes these files.
I tried and, so far, failed.

The following distributions of device independent troff do *NOT*
include any part of ideal:

 * Alcatel-Lucent Bell Labs Plan 9:
   http://plan9.bell-labs.com/sources/plan9/
   http://plan9.bell-labs.com/sources/plan9/sys/lib/tmac/
   http://plan9.bell-labs.com/sources/plan9/sys/src/cmd/troff/

 * Oracle OpenSolaris Heirloom DocTools:
   http://heirloom.sourceforge.net/doctools.html

 * AT&T Documenter's Workbench
   
http://www2.research.att.com/~astopen/cgi-bin/download.cgi?action=list&name=dwb

 * Fork of AT&T Documenter's Workbench
   https://github.com/n-t-roff/DWB3.3/tree/master/doc

In conclusion, i see no indication that you might be entitled
to redistribute these files.  If you have put any of them into
the groff git repo, that may already be a violation of Copyright
and hard to undo (it would be relatively easy to undo with CVS,
but it's quite problematic with git).

Your best bet may be to write a mail to Professor Chris Van Wyk
(the Copyright holder in the sense of the Berne Convention) and
kindly ask him for permission.  His address is available on his
homepage cited above.  He may be willing, and he may or may not be
*able* to grant it; if not, he is likely to know, or at least get
you started trying to figure out, who the Copyright holder in the
sense of U.S. law (because that's applicable for him) is; it could
possibly be Alcatel-Lucent, AT&T, or the SCO Group.

Yes, this does look like a mess, most definitely.  :-(

Yours,
  Ingo