Hi Giles,
"Copy and Paste" is not something only unexperienced
programmers do. Let us admit it, programmers do it
all the time. It is a basic tool which is applied
on many scales, from the highest level of the application
to the lowest level of the individual function.
Similar or duplicated code is not completely
bad, it can even be faster than entirely unique
code (for instance if you code a loop directly
in Assembly language, the branching command
can be eliminated). If repetitive code is good or
bad depends on the judgement of the programmer -
whether he wants performance or elegance.
There is a Harvard Business Review article
named "When Should a Process Be Art" which
says that some creative processes naturally
resist definition and standardization.
Programming is such a creative process.
The programmer adjusts the raw material until
it matches the desired requiremewnts. It is an
artistic process where the quality can not be
measured by counting the lines of code, see
http://4loc.wordpress.com/2010/11/22/progress-in-the-software-world/
I doubt that refactoring can be automated.
The book from Martin Fowler is useful:
"Refactoring: Improving the Design of Existing Code"
A list of patterns from this book can be found here
http://www.refactoring.com/catalog/
I am not sure if it makes sense to define a
fractal dimension for text or code. The box
counting method is a common method to
determine the fractal dimension of images
and graphics. We divide the space up into a grid
of boxes of size x, and count the number of boxes
of that scale that would contain a part of the
attractor.
If we divide the code into lines, and count the
number of lines which repeat themselves, we would get
a number between 0% and 100% (only repetition:
100%, no repetition: 0%). This is not a fractal
dimension. And if we are honest, we must consider the
whole stack, which begins on the deepest level of
machine code. Ruby for example is written in C, and
C in Assembly language. Many programs which look
elegant just hide the messy part under the hood.
They are based on large frameworks which contain
all the messy and ugly code.
-J.
============================================================
FRIAM Applied Complexity Group listserv
Meets Fridays 9a-11:30 at cafe at St. John's College
lectures, archives, unsubscribe, maps at http://www.friam.org
