Tricks and Treats

1

How can you change a programming trick into a programming treat? Try this algorithm:

  • If you develop a clever snippet of code, squirrel it away. This snippet is a "trick."
  • If you use the trick a second time, copy and modify the code. The trick has become a "treat."
  • If you use the trick a third time, generalize the idea and write a self-contained function. The trick has become a technique.

Tricks

When I write a program, I decide whether the program is for my own use or is intended for others to use.

If the program is for a one-time analysis, I don't bother to add many comments, I don't check for error conditions, and I hardcode constants such as filenames and dimensions of matrices. Sometimes I don't even save those programs. I create the program, run it, look at the answer, and move on.

But sometimes I come up with a fiendishly clever programming trick, and I decide to save the program in case I ever need to use the trick again.

Treats

When I do have an opportunity to reuse a trick, I find the old program. (This is not always an easy task since I've written and saved more than 8,000 SAS/IML programs! Thank goodness for grep!)

Change a value here and a constant there, modify the IF condition, and...voila! Problem solved. I have given myself a sweet treat: a program that solved the problem but required almost no work. What could be better?

Techniques

Actually, there is something better, and that's to turn the treat into a technique. To do this, I generalize the idea, remove the hardcoded constants, add comments, handle error conditions, and break the code into self-contained functions.

Now I can share my trick-turned-treat with others.

Share

About Author

Rick Wicklin

Distinguished Researcher in Computational Statistics

Rick Wicklin, PhD, is a distinguished researcher in computational statistics at SAS and is a principal developer of PROC IML and SAS/IML Studio. His areas of expertise include computational statistics, simulation, statistical graphics, and modern methods in statistical data analysis. Rick is author of the books Statistical Programming with SAS/IML Software and Simulating Data with SAS.

Back to Top