October 4, 2012 at 13:40Do you really want to continue doing this when you turn 50?
- Transfer
When I was still a professional programmer, my colleague once asked me: “Do you really want to continue doing this when you are 50 years old?”
I have to say, it made me stop and think.
I am a little disappointed in programming. Not because you have to work on solutions to complex problems. This requires careful thought, but this is the same way a writer structures a story or writes a dialogue that sounds plausible. Solving problems of this kind is fun and even fun.
But, unfortunately, programming is different. This is an attempt to create a working solution for problems that you do not fully understand and you do not have time to understand.
This is an attempt to cover the vast ocean of APIs, which can take years to study, but the market is moving forward, urging you on, so you copy and paste the code from the examples and try to make it work without a complete picture of the architecture of the application that you support.
This is reading documentation between the lines and trying to predict how extreme cases will be handled, and whether your assumptions will be true in two months or two years.
These are constant evolutionary changes in the language, compiler, libraries, framework and underlying OS. This whole snowball keeps you in maintenance mode, preventing you from making real improvements.
This is when all the work can stop due to a minor bug in seemingly reliable tools. And you turn out to be the first person to encounter the fact that a PNG image with four bits per pixel and alpha channel crashes a decoder and you need to get around this somehow.
One approach is to delve deeper into the problem and force your way through obstacles. If you are young, the office has an unlimited supply of coffee and all your friends are also in the office at 2 a.m., then ... of course, this is an option. But then, you need to do it again. And again. It is always like a skid at a speed of 200 km / h, in which brakes smoke and tires break, determining success or failure. But you miraculously stay alive to do it again.
I still like to do different things, and if there is nobody else to do it, I will do it myself. I continue to improve my little Perl script, which compiles this site, because this little script is unobtrusive and reliable, and allows me to focus on writing articles. I have a handy little image processing tool that is written in C and Erlang, and its source code takes less than 28 kilobytes. I know how it works inside and out, and for me it's faster to make changes to it than to get ImageMagick to do what I want.
Is there more stress on a large scale development? I have to admit that yes. Still, this is the lot of young people.
I have to say, it made me stop and think.
I am a little disappointed in programming. Not because you have to work on solutions to complex problems. This requires careful thought, but this is the same way a writer structures a story or writes a dialogue that sounds plausible. Solving problems of this kind is fun and even fun.
But, unfortunately, programming is different. This is an attempt to create a working solution for problems that you do not fully understand and you do not have time to understand.
This is an attempt to cover the vast ocean of APIs, which can take years to study, but the market is moving forward, urging you on, so you copy and paste the code from the examples and try to make it work without a complete picture of the architecture of the application that you support.
This is reading documentation between the lines and trying to predict how extreme cases will be handled, and whether your assumptions will be true in two months or two years.
These are constant evolutionary changes in the language, compiler, libraries, framework and underlying OS. This whole snowball keeps you in maintenance mode, preventing you from making real improvements.
This is when all the work can stop due to a minor bug in seemingly reliable tools. And you turn out to be the first person to encounter the fact that a PNG image with four bits per pixel and alpha channel crashes a decoder and you need to get around this somehow.
One approach is to delve deeper into the problem and force your way through obstacles. If you are young, the office has an unlimited supply of coffee and all your friends are also in the office at 2 a.m., then ... of course, this is an option. But then, you need to do it again. And again. It is always like a skid at a speed of 200 km / h, in which brakes smoke and tires break, determining success or failure. But you miraculously stay alive to do it again.
I still like to do different things, and if there is nobody else to do it, I will do it myself. I continue to improve my little Perl script, which compiles this site, because this little script is unobtrusive and reliable, and allows me to focus on writing articles. I have a handy little image processing tool that is written in C and Erlang, and its source code takes less than 28 kilobytes. I know how it works inside and out, and for me it's faster to make changes to it than to get ImageMagick to do what I want.
Is there more stress on a large scale development? I have to admit that yes. Still, this is the lot of young people.