[geeks] Doorbells

[Phil Stracchino]

Lionel Peterson wrote:
>> From: "Jonathan C. Patschke" <jp at celestrion.net>
>> On Fri, 21 Mar 2008, der Mouse wrote:
>>> I believe that last paragraph is false, that it is actually a
>>> violation to practice the invention even privately.)
>> To the best of my knowledge, that is actually the case in the US, but,
>> it's a law I cannot keep on good conscience.  I do not give the
>> government my consent to regulate what I build and create for my own
>> purposes (provided I'm not harming others with those creations).
>>
>> I have to wonder if there's some sort of "fair use" sort of exemption
>> for education, though.  One of my assignments my first or second year of
>> college was to implement an LZW codec, and the patent on LZW was still
>> valid back then.
> 
> More likely an instructor that is ignorant of the law. I don't see how an instructor can (essentially) wave a valid patent in the classroom and ask students to violate it without impunity.
> 
> I don't think the law is on the teacher's side, I think personal/private use is all but impossible to detect, and while your instructor may never get caught, that doesn't make it right.
> 
> If the instructor asked you to implement a compression algorithm and you "invented" LZW out of ignorance of the existng patent, you would be fine (IMHO, IANAL, etc.) - ignorance is a valid defense (AFAIK, I think it is in US Tax law, anyway, since even the IRS can't give consistent advice ;^).

On the other hand, my understanding is that LZW is only an incremental 
development of Huffman coding, which is pretty much the definitive 
archetypal basic compression algorithm for arbitrary data.  When I was 
in college, the assignment was a Huffman coding implementation.  I don't 
see it's too unreasonable now to be telling students to implement LZW.

On the other hand, in my numerical analysis class, the professor (who 
was an ivory-tower pure mathematician) wasted half the quarter focusing 
on one specific algorithm for ......... I don't even remember any more 
*what* ......  had us code a programming assignment to implement it .... 
then took in the assignments, stacked them on his desk, never to my 
knowledge ever even opened them (we never got them back, and never got 
graded on them), and casually - even dismissively - tossed off the 
following verbatim remark:

	"Of course, you'd never use this algorithm in the real
	 world, because it propagates errors too badly."

Which is the point at which all of us computer scientists in the class 
looked at each other in complete disbelief, and demanded to know why the 
fuck he'd just wasted half the quarter in a numerical analysis class for 
computer scientists on an algorithim that was totally worthless in the 
real world.


Thought experiment:  How much could you NOT teach if you were not 
allowed to discuss or set assignments on anything covered by a patent? 
Not just in computer science, but in any subject?  How badly would 
pharmacology classes, for example, be impacted?


-- 
   Phil Stracchino, CDK#2         ICBM: 43.5607, -71.355
   Renaissance Man, Unix ronin, Perl hacker, Free Stater
   alaric at caerllewys.net            alaric at metrocast.net
           It's not the years, it's the mileage.