The Developer's Dystopian Future

I started programming in 5th grade, and still love it after 40.

But the world has so many large problems these days. I think many people decide at some point, there is more reward and more meaning in working on those things, than following the latest Android API diffs.

Technology is still interesting to me, but only as a tool to accomplish more important things.

Unfortunately, as a software developer - I view the most important technology not to be software, web frameworks, or mobile, but developments in energy, material science, etc. Things that are out of reach for me as a developer.

How can i solve any meaningful problems, when the main tool at my disposal is mostly just copying bits from one memory location to another?

Thats what makes me depressed and disenfranchised with software development at 40. I'd like to work on more meaningful things, but I find my skills rather impotent given the enormity and complexity of the world's problems.

I have felt the same from time to time (45 here). However, as software technologists we have one advantage: our work can be completely independent.

In this case I mean if you can conceive a project to better a part of the world, it may be possible to implement with resources you already have at your disposal.

The materials energy researchers need resources beyond the reach of most individuals (and their personal budgets), and in some cases corporate resistance to embedded profit centers.

I'm probably considered an "old guy" in the software world. But I realized a few years ago that there is little new under the sun. We get new names on stuff all the time; we get old concepts repackaged as "new." Data still must be organized and searched, transmitted and received. Instructions must still be transcribed and turned into something the silicon can use. The only differences are in the syntax. Sometimes, those differences are what we needed to make a breakthrough in comprehension of complex systems. Oftentimes, there's little difference.

I was once worried I'd be left behind. I don't worry about that now. Mainly because I've got the skills to grok the "new" and catch up. But also because "new" is just repackaging that research paper from 1964 into another package for $YOUR_FAVORITE_LANGUAGE.

Woah woah woah, relax. OK, C++ and client-side tech (HTML/CSS/JS) is enough to go for some time. Common Lisp is a good way to go next, but you'll see most of that return in how it affects your use of other languages. For the rest in your list: they're solving overlapping sets of problems with overlapping sets of techniques.

You have to know very few of them to be quite productive. You also have lots and lots of time. I'd honestly focus on getting better at those that you already know than worry about adding in more. And just start building shit. Nobody cares what technology you use, as long as it's not PHP[2].

The more technologies that you know, the easier the new ones will be. I don't know why, but people keep adding them on their resume like merit badges. They're tools, and you aren't responsible for learning tools, you're responsible for making solutions.

But, if you want to learn a bunch of them, here's how you get started. When you start a new technology, ask yourself a few questions about it:

- What's the problem it's trying to solve? (hint: this applies to programming languages as well)

- What're the constraints imposed by the environment?

- What fundamental techniques is it using?

- What are its predecessors?

Those four questions will explain most of the technology[1], before you even look at the details. Do this a few times, and you can predict much of a new technology just from the answers to these questions.

Example: C++:

A systems language for large programs. Must have as close to minimal runtime overhead and optimal code as possible. It's native-code compiled with the platform ABI, usually with the same compiler backend as the C compiler, and without any runtime interpreter or compiler. Predecessors include C and Simula. Footnote 1 applies here, as it is designed by committee, with a "shove a lot of stuff in the language, and let each user figure out which parts are useful" attitude.

[1] Modulo random experiments, social system issues, laziness, and bad ideas.

[2] The only reason to use PHP is "someone made me."

I find myself moving farther down the stack rather then picking up the "new hotness" framework etc. Right now I'm very interested in embedded / low level systems. Someone still needs to maintain and build the actual technology that these young'uns use. :P