Re: strsep sourc

> If you own the license in one language, you own it in all of them.
> Same thing applies for different platforms.

I am not a lawyer but have had some...experience in this area.  One
copyrites a written work.  One patents an idea.  This is the basis of
intellectual property law in the US.  One cannot patent a common idea like
addition.  One CAN copyright a written work containing addition, such as a
math textbook.  Copyrighting a math textbook in English does not
automatically grant me copyright status to all math textbooks in all
languages.

Computer programs are copyrighted works and are considered by US law to be
written works like novels.

Scott's comments about derivative works comes from the idea that you can't
copy pages from my copyrighted math textbook, publish it with a new cover
and call it your own.  Nor can you change the occasional word here and there
and call it your own.  Such uses are called derivative works, because you
basically derived your work from mine.  How much you are allowed to copy is
a very grey area, but if it's more than a movie review's worth, you're
probably liable.  If you can afford the court fees, feel free to copy away
and deal with the lawyers.

Dealing strictly with code, if you were to rewrite this C code as RPG IV,
you would probably not be subject to US copyright problems, despite the fact
that you derived your work from The Regents of the University of California.
That's because your work does not look like theirs.  You may be violating a
patent they hold on the algorithm (the idea of how they did it) which makes
the whole discussion even more complicated.

This 'translating' issue is a grey area, and the courts have decided both
ways on the issue.  The point is that it depends on how much of the
style/look moves from the original work into the derived work.  If it's
theoretically possible to automate such a translation, you're probably
liable.  Again, if you have the time, money and curiosity, feel free to test
the legal waters to your heart's content.

Scott's point that if you don't care for the BSD licence terms makes a great
deal of sense.  Design your own algorithm and implement it your own way.
How would you like some lawyer to take away ALL of your company's software
because you didn't abide by some piddly technicality?  That is what they get
paid to do.  And before I get pooh-poohed about it not being likely, think
of it like this: A disgruntled co-worker might turn in the company some time
in the future, and the lawyer will use posts from public archives to show a
repeated pattern of abuse/disrespect for copyright/patent law.  The Regents
don't have to come looking for you; they may never even care.  But a lawyer
can smell money a long way off, and cases like this are easy pickings.  It
matters not one whit how sensible this seems to you.  That's the way it is,
and the US Congress just amended the copyright laws to extend the copyright
terms, so they don't seem inclined to make changes that ease copyright
restrictions.  I have no idea how non-US copyright laws work; you'd have to
consult the Berne Convention for some details on that legal quagmire.

I served as an 'expert witness' in such a case.  The company is bankrupt.
By the way, an 'expert witness' doesn't have to be very expert to serve in
that capacity.  And the violation doesn't need to be severe for the legal
system to destroy you.

Take Scott's advice.  If you want to use their free code, follow their
rules.  If you don't want to follow their rules, design it yourself and
avoid the legal entanglements.  I'm sorry for the ramble, but this one is a
personal bugaboo.  I saw what something trivial did to a company first hand,
and would hate to hear that someone here was caught in a similar noose.
  --buck