I program stuff.

Saturday, February 10, 2007

A very good recommended read for all coders. This specifically covers web coding but is applicable to any type of code.

Getting Real: The Book by 37signals

A basic takeaway from this book is to keep it simple. Which is not new (and they say so) but pushes the idea a lot further than traditional thinking on development. The keep is simple approach is usually in practice only applied to the actual coding and architecture part, not to the requirements. Usually people throw in as many features as possible to their apps which always makes them bloated in the long run. This is really a marketing problem. They want more customers so they try to appeal to all possible customers so they put in all possible features, which in the end makes a big lumbering app that arguably becomes less useful to all because it has so many random features that they never will use that gets in the way here and there. Most of the most useful apps are the ones that are super specialized like grep, tail, or ls (a.k.a. dir).

Another great takeaway is an important key to business, which I like to characterize by the phrase, "Never underestimate the power of doing nothing." The idea is not to sit around a do nothing all day but not to jump at things like it's a do or die emergency, sit back and allow the alleged emergencies to settle in. Most things are not as important as you think they are when you first encounter them. Sit back a while and let it digest for you and others involved and see how important it looks in a few days or weeks.

The big benefit to this is that doing this errs on the side of doing less and not on the side of doing more so you have much less of a chance getting into the situation where you are spinning your wheels all day fighting fires and not getting much done in the process. The dangerous cycle of doing more to accomplish less. If something is important to work on it will still be important to work on next week. Also this helps the "getting into the zone" process where it give you more freedom to focus on one thing at a time to do good work rather than doing many things at once.

I used to jump at every little thing like it needed to be done right now. When you do that you increase the amount of work you have to do in an exponential way. You begin to meddle in things that are not so important which tends to make things worse. This requires you to keep fixing things over and over again because you have a perception that it's not good enough when it actually was good enough in the first place and you didn't need to do anything about it. It's not that the work you do is bad when you do this it's just that you're adding value to something that is not important as something else more important. There is a huge opportunity cost that you are paying here. Usually no one really cares all that much when they step back from things anyway. Following a ton of little emergencies takes a huge chunk out of your moral in the long run. You'll start saying to yourself, "nothing I do is ever good enough."

Remember that odds are that most things are not that important relative to all the tasks that show up on your radar. Take comfort in the fact that if it is important that it will be pretty obvious that it is important next week or even in a few weeks. Let things compete for your time rather than you competing with yourself in order to do everything.

Start saying to yourself, "this is emergency is not good enough for my time, it's not worth it."

A way that can help to start to think about it is, only do something if it was obvious last week that you needed to do this today. Like, "okay, I knew that I would have to eat lunch today so I'll go ahead and do that rather than fix this bug that I didn't know about yesterday." There are a lot of other little examples that help in this book that I very much recommend reading if only for this single reason alone. They took this idea to heart and it shows on every page. This important concept makes for better work and a better work environment.

No comments: