>> There's no contamination from looking at code.  Look all you want.

True but, in most jurisdictions, so dangerously misleading that you're
probably better off treating it as false.  The contamination actually
comes from then using what you looked at, something that's sometimes
very hard not to do.  (But it's not _always_ contamination even if you
do - see below.)

> "Derived" is derived.  If you look at code that explains how to set
> bits to enable certain settings, any code you write that applies what
> you have learned is "derived" from the code you looked at.

Yes, but that does not necessarily make it a derivative work within the
meaning of copyright law.  It may be a derivative work, but it also may
not.  For example, consider all the works of fiction (most recently in
my own reading experience, the Midkemia ficton by Raymond E. Feist)
which include elves as long-lived humanoids with pointed ears and sharp
eyes.  Derived from Tolkein's writings, undoubtedly, but not thereby a
derivative work thereof.

> You can argue [...], but unless you have the legal means to defend
> yourself in court, you lose.

Yes, but that's basically irrelevant - it's true more or less
regardless of the issues.  (In the USA.  While most legal systems have
tendencies in that direction, few have taken it as far as the USA.)
What it amounts to is "don't tick off anyone who has way more money
than you do", regardless of actual legalities.

/~\ The ASCII				der Mouse
\ / Ribbon Campaign
 X  Against HTML	       mouse@rodents.montreal.qc.ca
/ \ Email!	     7D C8 61 52 5D E7 2D 39  4E F1 31 3E E8 B3 27 4B