Re: [Phys-l] real-world mechanics problem

[John Denker <jsd@av8n.com>]

On 01/31/2011 04:42 PM, William Robertson wrote:
> To clarify why this is a good transfer problem, although I'm repeating  
> much of what JD said: You can't solve the problem by plugging numbers  
> into an equation. You can't solve the problem using a memorized  
> procedure from a homework or in-class problem. Yet, if you understand  
> the underlying physics principles, you can solve the problem. A good  
> transfer problem generally evokes responses, verbal or unspoken, from  
> students such as, "But we've never done a problem like this before!  
> We've never dealt with more than one pulley system. This isn't fair!"   
> I recall in my HS physics class (worst class and worst teacher ever)  
> that we solved pulley problems by counting the number of ropes between  
> pulleys and plugging into a formula. Absolutely no understanding  
> involved.

We are definitely on the same page.

As $.02 of further clarification, I see nothing bad about counting
ropes.  
 -- If you understand the physics, counting ropes won't hurt.
 -- If you don't understand the physics, counting ropes won't help
  with real-world problems.

FWIW, when I solve the problem, I count ropes.  I /interpret/ the
rope-counts in a very particular way, but at some level there is
rope-counting involved.

I see rope-counting as the fur on the end of the tail of the dog.
I hope we all agree that the emphasis should be elsewhere.  The
fur should not wag the tail, and the tail should not wag the dog.

To say the same thing in more constructive terms:  
The _principle of virtual work_ is tremendously valuable, here and 
in innumerable other real-world applications.  I don't spend any 
effort trying to "remember" the rope-counting rule, because I know 
I can rederive as a corollary of PVW.... I can rederive it in a 
millisecond if/whenever I need it.  (So once again this blurs the
distinction between thinking and memory.)

If you want another real-world problem where rope-counting is
nowhere near sufficient, see next message.