Re: Interesting Licensing Issue -- Crafty

[Paul Serice <ugs@laci.net>]

John Hasler wrote:
> 
> Have you pointed out to Prof. Hyatt that the sort of people hs is
> worried about are unlikely to even read his license, let alone obey
> it?

I think as it stands now, if a hidden Crafty clone is competing in a
tournament, both Prof. Hyatt and tournament sponsers cannot do much
about it.  If a clause that requires the clone to expose itself exists
and is violated, then Prof. Hyatt will have grounds for demanding that
the tournament stop running the program.  (Again, I'm not sure this is
actually the situation.)

It's not so much about an honor system amongst thieves; rather, its
about an honor system amongst those who fear the legitimate threat of a
lawsuit.



> > Of special importance here I believe would be paragraph 4 which
> > states in part, "The license may require derived works to carry a
> > different name or version number from the original software."
> 
> But it does not say "The license may require derived works to carry a
> specific name."

I believe you are saying the author can require a different name to be
used but cannot specify that name?  Of course, the paragraph in question
doesn't literally say what you want it to say either.

Furthermore, suppose Prof. Hyatt wanted to require all clones to use a
different version number.  Could Prof. Hyatt, as per paragraph 4,
specify the versioning scheme to be used?  If so, instead of using
version numbers, a reasonable reading of the paragraph is that
Prof. Hyatt could use version tags.  So, instead of "Crafty v.16.7c"
(the "c" is for "clone"), couldn't Prof. Hyatt force it to be "Crafty
v.16.7_clone" or "Crafty Clone v.16.7"?



> Assume that I take crafty and improve it beyond recognition, replacing
> almost all of Prof. Hyatt's code with my own work.  You would require
> me to identify the result as a crafty clone?

Good question.  I could go with the practical argument that you would
never be detected, but I think I'd be a cop out for doing so.

I really don't know exact where the lines of demarcation are for what is
and is not a derived work.  No one does.  Admittedly, some people's
guesses are better than others.  My guess is that for the situation
above most reasonable people, including Prof. Hyatt, would be happy to
see the program in question with its own name.

As I understand it, the domain of problem cases to date consists of
people modifying two or three lines and claiming the entire work as
their own.  Another good question that has no bright-line rule is when
does the original work become a derived work.  I doubt changing two or
three inconsequential lines does the trick, and to defend calling such a
work anything other than "Crafty" or "a Crafty clone" is hard for me to
understand.


Thanks
Paul Serice