Re: [Phys-L] spreadsheet modeling

[John Denker <jsd@av8n.com>]

On 12/28/2013 08:45 AM, Richard Tarara wrote:
> The final spreadsheet developed can then be adjusted ....

This is a reeeeally good idea.

> The point of all this is to try and really get at how calculus can
> give us very precise solutions to problems where the variables are
> interdependent.

That's true.  Indeed, things are even better than that.

Advantages of this approach include:

1) Relatively speaking, it is easier for most students to get 
 started with a spreadsheet ... compared to an imperative 
 programming language such as C++.  There are tangible practical 
 reasons for this, in addition to an intangible "intimidation 
 factor".

    Beware that there are /some/ people who despise functional
    programming languages such as spreadsheets and labview,
    and prefer something like C++.

2) It is relatively easy to draw graphs using the spreadsheet.
 If you change the data, the graph changes instantly.  The 
 point is that /visualization/ is an important part of physics.
 There is no point in modeling if you can't visualize what
 the model is telling you.

3) This also makes the point that you can integrate a differential 
 equation numerically, whether *or not* you can integrate it 
 analytically.

 You definitely want to do some of each:  Some where the numerical
 solution can be compared to the analytic solution, and some where
 there is no analytic solution.

4) Any skill students develop using spreadsheets is portable to
 other disciplines, so that lots of them see the work as relevant.
 This includes business-oriented and nursing-oriented students, 
 not just physics-oriented and engineering-oriented students.

5) This is partially contrary to item 1, but there are some
 ways in which clever tricks can be used to make the spreadsheet 
 easier to use and more powerful.  That is to say, I would rather
 not spend hours going click-click-click-click if a little bit 
 of cleverness allows me to achieve the same result with 100-fold 
 less clicking.
     http://www.av8n.com/physics/spreadsheet-tips.htm

6) This underlines something I said previously:  The calculator
 is not the enemy.  The computer is not the enemy.  The goal is
 to achieve understanding of the math and the physics.  There
 are of course situations where people use the machine instead
 of thinking, but it doesn't have to be that way.  Any tool
 can be abused.  It is a poor workman who blames his tools.  The 
 machine should be used to /enhance/ thinking and understanding.

 Do not confuse the presence of one thing with the absence of
 another.  If there is an absence of thinking and understanding,
 do not blame the computer.  Instead, teach people to use the
 computer more wisely.

7) As a corollary of the above:  Teach 'em to *document* their
 spreadsheets.  Make it part of the grade.  Insist on it 
 starting at the very beginning, so they have zero opportunity
 to develop bad habits.  /Checking the work/ is the foundation
 of critical thinking, and lack of documentation makes it hard
 for the author to check the work, and even harder for anybody
 else.
  -- At the very least, if there is a number in cell d3, write 
   some text in e3 specifying the units.
  -- If there is a long column of numbers, put some text at the 
   head of the column that explains the meaning, perhaps something
   like "KE".  Sometimes a single header will do for multiple
   columns, e.g. "momentum x,y,z".
  -- Use frames and colors to delineate logical groups of stuff.
  -- Use the get.formula() function to make it that much easier
   to see what's going on.  This is particularly valuable when
   explaining the code to somebody else.
  -- etc. etc. etc.
      http://www.av8n.com/computer/htm/good-software.htm