March 26, 2007

Don't Keep the Bad Stuff

Drew Hamlin: 9:00 PM

This is something I've learned through experience. It might sound obvious but when you're actually faced with the situation, it's not always so easy...

Whenever you work hard on something you really care about, you sew your heart into it. That's a great thing because it means it's honest.

Sometimes though, down the road it turns out you made a mistake. The interface or feature you've spent a week designing or coding is just not feeling right...

The problem is that it can be quite tricky to let that hard work go to waste. When you spend a lot of time on something, you feel like you must somehow justify the work being done in the first place. There is a temptation just to leave it in or tone it down or make it a setting.

Don't do it. Don't keep the bad stuff. Throw it out. You do not need to keep something just because you spent some time on it. Go back to the drawing board. Kill it, start over, and do it right. If it's just not needed, leave it out.

Never let "but I worked so long on it" be an excuse to keep a feature that isn't working or isn't necessary. Mistakes happen. That's part of the process. Move on and feel great knowing you did the right thing.

This process works great as well for something even when you think it is working. Weeks into a project, it can be amazingly helpful to drop everything, get a blank slate, and just start over. You might be amazed at some of the things you learned through your initial attempt that radically change the way you think about it the second time around for the better.

In other words: always throw yourself into your work... just don't become attached to it. Never get so involved that you forget to step back.

5 Comments:

Luke: March 28, 2007 6:26 AM

so true, i've had to do this a few times but worth it in the long run. it's like when your pc crashes (windows user here) and you lose work. more often then not on the 2nd attempt it works out better.

wphj: March 28, 2007 9:33 PM

So true, I know the feeling exactly.

Elliot: April 11, 2007 11:06 PM

This mindset applies to a lot of different areas. I like when thought goes into a product, whether it's software or a tangible item. Too often we see things mismashed together and they don't flow. They lack soul. That's how I feel about Vista...

Michael: April 15, 2007 5:09 PM

Even back when I stared coding in the 1980's (yes, I'm an altakaka), programmers viewed their code as their "children" with a rather propriatary air. One thing I've noticed since then is NOTHING is forever (except maybe entropy and taxes).

As Harvey Fierstein once said in a 1992 college commencement speech:

Change your mind as often as possible. Just because you thought something yesterday doesn't mean you have to think it today. Don't ever become a prisoner of your own opinion.

He also speaks about being weary of anyone who says they "know" and I've been in technical meetings where I've been bitten by that myself ("I could have sworn that was in the code...").

Even though the speech is 15 years old, most of it still holds true today.

Chad: June 13, 2007 7:14 PM

Fred Brooks summed it up nicely like this: "Write one to throw it away."

...or something along those lines.

The first time I wrote a novel, it was an interesting exercise, but I realized it was crap in the end, but it was great experience.