[Phys-l] credit + priority - spoilers

[John Denker <jsd@av8n.com>]

On 07/13/2007 12:14 PM, Joseph Bellina wrote:
> Why does everyone have to give away the answer so that other lose the  
> opportunity to figure it out for themselves...hum is that a metaphor  
> for what happens in some classrooms.

I know it's bad luck to answer a rhetorical question ... but I 
think Joe raises a very worthwhile issue.

Answer:  Yes, classrooms train people to make sure they get credit
for solving a puzzle.  (It's not even a metaphor.)

In my opinion:
 a) If taken to the extreme, this leads to "spoilers" and other problems.
 b) But in its less-extreme forms, it is not entirely bad.  There 
  are lots of situations (in the classroom and in the workplace)
  where establishing credit, priority, et cetera is necessary for
  survival.

So the trick is to train people to do this responsibly and in moderation.

A widely used method for publishing an answer without spoiling the 
riddle is to encrypt the answer.  

1) For low-stakes riddles rot13 is sufficient.
  http://en.wikipedia.org/wiki/ROT13

On any unix system, rot13 can be implemented by the command
  tr a-zA-Z n-za-mN-ZA-M

You don't need to memorize the command;  all you need is the manpage for
tr and the fact that the middle of the alphabet falls between m and n.

  There is a plugin called "leetkey" that offers to encode/decode rot13
  and various other weak codes, right in your thunderbird mailreader
  window ... but I do *not* recommend it.  It's full of bugs and weird
  side-effects.  It's hard to use.

If you are unix-deprived I suggest something like this:
  http://personal.inet.fi/cool/jjlammi/rot13.html

2) If serious prize money were at stake, stronger crypto would be appropriate.
This allows you to demonstrate that as of a certain date, you could solve
the problem /and nobody else could/.  (In contrast, if you just publish
the solution, every jerk in the world can claim an almost-contemporaneous
solution.)

The formal name for this is "zero-knowledge proof".  Some of the results in
this area are powerful and elegant.
  http://en.wikipedia.org/wiki/Zero-knowledge_proof