"Learning", I have been thinking about this topic lately. There are several things I felt about learning, from my point of view as a Computer Science student as well as an independent person. Lately, I am feeling like, I am not learning fast enough. There seems to be so many topics to be learned (mostly are not about the course lessons though, they are about the Algorithms). Sometimes, while learning, I realized that out of these many topics (most are some subtle algorithms for solving some problems), there seem that many person out there (especially competition enthusiasts) have mastered much more than I do. Thanks to the "flat world", this is becoming a problem, because most competitions I join are international, and mostly online as well. Somehow, I feel like I have to accelerate my learning, otherwise I will never catch up with the champions.

However, learning requires two things : materials, time, and motivation. Acquiring the text materials is no more difficult thanks to the internet, and motivation is certainly not a problem for me. However, time is a problem. I realized that I can’t learn much while I’m multitasking (which, means, having the long term projects, especially big software projects while learning). Having these extra burden while learning significantly distracted my concentration on a single topic, which means, wasting my time when learning things.

These works (project - especially e.COM System) however, can not be abandoned anyway. Although I don’t learn much doing projects, I have committed to do it. Therefore, I think the best solution is to have the projects done as soon as possible, and for the future, I should limit the number of tasks to do at one time. I think it is about at most one project at a time in the future, which should not be a big project, one week of work at most. For this time, I aim to finish the whole e.COM System by the end of July. After that, I will focus on learning algorithm, in preparation for the ACM ICPC Regionals 2007. Last time, I did not prepare enough to achieve the top positions. This time, there should be no more compromise, as this will be my last chance of joining the ACM ICPC. The world is changing, the competitors are growing, and I’m afraid that if I don’t learn fast enough, I will never reach the top.

In the past, I focused more on software development. Later on, I realized that the world of software development does not mean much : it is something that an average programmer can learn easily. Software development is more about work : if you work, you’ll be able to do it. However, algorithms and competitions are different. Here, learning means something, and time is important as well. In a simpler way : if I practice too much software development now (J2EE or web development, for example), the skills I get will be useless in a few years to come. It is different with algorithms, which is growing, rather than changing. Once I master all of the common algorithms, improving should require less work because new algorithms don’t come up that often. Most algorithms in use today basically comes from 60s, 70s, 80s, but so few comes from 90s or the last few years. Therefore, it’s less likely to become obsolete anytime soon. Compare with web development, where a handful new technologies comes out every month. Without understanding these changes, any programmer will most likely either end up jobless in ten or twenty years, or otherwise work as a low level programmer for the entire career.

The world of computer science is growing, while software development is changing. Most programmers are software developers instead of scientists. Learning software development is much easier than learning the science, but software development require a more of constant learning due to its changing nature. Learning is the only way for a programmer to survive the future, and that’s why I don’t want to be a programmer for my entire career, it’s not a stable job. What do you think?

-Andrian Kurniady