Re: Identify auto tick locations

[John W. Eaton]

On  1-Nov-2007, Peter A. Gustafson wrote:

| (Rambling)
| You can reimplement an algorithm after studying it in detail

I urge people working on Octave to NOT take this as advice that it is
OK to read Matlab .m files to learn how they work in order to
reimplement them.  I ask that anyone who works on Octave should NEVER
read any Matlab .m files that they may have available to them.

Think of it this way.  If I have some code that is not freely
available (default copyright conditions; no copying allowed) then I
can't distribute it.  So I modify it a little.  Then I clearly have a
derived work, which is not entirely my own, and which I can't legally
distribute.  Then I modify it some more.  It is still a derived work
and I can't distribute it.  Then I continue modifying it until it no
longer resembles the original.  Is it still a derived work?  I would
argue yes.  If you think no, then at what point does it stop being a
derived work?  Is it OK if you make all the modifications at once
instead of doing it incrementally?  You say it is a grey area, and
open to interpretation.  Perhaps, but I would not want to bet on the
interpretation going in your favor if you are the one reading and
reimplementing.  So I ask people to please not implement code for
Octave using any method like this.

jwe