Blog Stats
  • Posts - 102
  • Articles - 32
  • Comments - 223
  • Trackbacks - 16

 

Treating the symptoms, but not the problem

A few days ago, I was riding up the elevator to my office and as usual, was annoyed when it stopped at the second floor. My office building has a parking garage right behind it and there's a connection to the second level of the building. I have a parking pass in another garage because there's a two year waiting list for this one, but I digress.

What has often bothered me is the fact that I need to get to the eight floor and more often than not, on my way up in the morning I'm stopped by people who take the elevator from the second floor to the third floor. To get to the elevator on the second floor, you have to walk past the stairs to the third floor. But these people will wait for an elevator so they don't have to climb 13 steps. It's a little ridiculous.

Today, was even more so. The man who got on the elevator had two bags over his shoulders. One was a gym bag that had an old pair of sneakers sticking out of it and some gym clothes next to it. The other bag had some office documents I presume and a Lean Cuisine meal sticking out the top.

This is a classic case of treating the symptoms and not the problem. "I'm overweight so I'm going to go to the gym and get on a diet to lose weight." Nevermind changing the basic habits that made me that way to begin with to solve the problem long term.

I've seen this is software too and it's just as annoying. The following is a recreation of some code I worked on a long time ago.

// Fixed: autocorrect if this variable wasn't set to keep the progg from crashing
if ( selectedOption == null )
{
    selectedOption = 1; // default to the first option
}

Hmmmm. Here's a thought. Instead of just avoiding the problem, why don't you find out what's causing it and fix it!

If the user didn't select something, chances are they didn't mean to select option 1. You can't make the assumption that the user even knew what Option 1 was, let alone that they wanted to select it. Very poor bug fix if you ask me.

When you're debugging a problem, make sure you understand the root cause of it and treat that. Don't assume that by fixing the symptoms, you've fixed the problems. That's like getting treated for lung cancer while you continue to smoke like a chimney.

Feedback

# Be it life or software « Entertaining Research

Be it life or software « Entertaining Research 7/11/2007 1:51 PM |
Gravatar

# re: Treating the symptoms, but not the problem

I've seen this way of "fixing" bugs more than once.
Usually by people who are real bad at debugging e.g.:when faced with a problem in the application can't or take very long to find the source (hint : the debugger can help you if you learn how to use it).

The reasoning seems to go this way:
"Hey I found the source of the bug" (they actually onle found some code that is somehow related to the buggy code). "Mhh the system doesn't work when used this way so I'll just use it another way".

Nevermind the fact that
-the "fix" only heal the symptom instead of curing the disease.
-the same problem can occur elsewhere
-it's ugly
-it may break when the real problem is fixed 7/15/2007 4:16 PM | Monkeyget

# Entertaining Research

Entertaining Research 7/23/2007 5:44 PM |

Post a comment





 

Please add 3 and 5 and type the answer here:

 

 

Copyright © Mike Taber