Re: Generation of permutations (copyright)

[dewar@gnat.com (Robert Dewar)]

"Steven Deller" <deller@smsail.com> wrote in message news:<mailman.1021324262.10270.comp.lang.ada@ada.eu.org>...

> While I remember writing this for a course on Ada that I prepared ada
> gave at UofMD on my own time, 

You have to be very careful about the notion of "on my own time". Many
employment contracts specify that your employer has IPR rights in everything
you do, wherever and whenever it is done. At NYU for example, specific paper
work has to be executed to recognize that the university has no IPR rights
in something a faculty member creates, since the default assumption is that
NYU does have such rights. Steve would have to carefully check the faculty
handbook etc at UofMD to know if he has clear copyright interest in this.

> I forgot that it was later included into a
> Verdix library unit "ordering" (along with a heapsort, quicksort and
> insertion sort) and even later included with all Apex releases.

As above, the CW status on this may still be unclear. Copyright issues are
indeed complex, and part of the point of a thread like this is to make people
more aware of the issues.
  
> As such, it now bears a Copyright from Rational, all rights reserved (it
> is slightly different than my original).

> If you have a Rational license you are able to use the code

Use is a very vague word. You would have to read the Rational license clearly.
And the mere fact that it "bears a Copyright from Rational" is still not
proof of anything. Of course part of the reason for having a formal contract
with a company like Rational is that you are getting a guarantee from Rational
that what they say is theirs is indeed theirs, and you have recourse if that
is not the case. 

For this particular code, there might be two details lurking

1. As above, UofMD might have a CW interest in the code

2. Steve may never have assigned the CW to Rational. Since this apparently
was not created while he was at Rational, it is not automatically covered
by the implicit or explicit work for hire agreement that Steve has with
Rational.

Now of course this code is trivial, and no one is going to sue over it, it
might not even be protectable. 

But it is an interesting illustration of the importance of knowing the CW
status of things you use. If you have a license agreement from a reputable
company, you are in reasonable shape, because that company takes 
responsibility for IPR issues. For example, if you purchase a copy of
GNAT Professional from ACT, you get a specific license agreement providing
assurance that everything you get is properly licensed.

But when you download stuff from the web, you have to take responsibility for
this yourself. If you are just an individual doing things for your self, you
don't have to worry in practice. A lot of such use is covered by fair use,
and in any case, it is unlikely that anyone comes after individuals.

But if you publish resulting work, or incorporate it into commercial products,
you have to be careful.

Another example. If you build a commercial product based on a version of
GNU/Linux that you have downloaded from somewhere, you have to worry about
whether all the copyright assignemnts etc are in place, and that's hard to
do yourself.

But if you acquire your copy of GNU/Linux from a company like Redhat, IBM,
Suse, etc then that company takes responsibility for the IPR integrity of
their products.

If this sounds like a bit of a complex mess, that's because it is!