I'm so jealous of middle-aged developers. Are we unlucky? [closed]

Question

First of all, I couldn't decide the right words for title after reading How old is "too old"? question. I still can't decide with old, middle-aged, expert or professional.

As you know, there are a lot of developers over 35-40 years old in StackExchange. They are helping to beginners and having less experience people. They are so valuable users. And mostly, their some answers are awesome becauase of good analyzing and adding their experience to answers.

When I see this kind of users, they all started programming with kind of Lisp, Pascal, Cobol or Basic type programming language. These are the first modern programming languages. And I think these people are so lucky learning programming with them first because of they are close to machine language. Maybe they learn the basics of programming better. Now it looks they are working easly with object orianted programming.

I'm graduated from computer engineering a year ago . First year, we superficial learned C programming language. Second year, there are C# and Java course. But when I try to learn this languages in colluage, I always feel like I missed something. Learning languages, structures, libraries.. if fine but teachers never teach us "How the codes working in machine?". They teach us the programming language which is popular in software market. So kind of they prepare us for companies. They think If we learn assembly or fortran, we can't easly find a job according to Java or C#. Unfortunately, at these days, this is kind of true.

Could be better learning first Assembly, Fortran, Cobol instead of Java and C#? Are we unlucky?

Answer 1

Consider yourself lucky. We have an abundance of mature languages today, blazing-fast hardware is available at commodity prices, and virtually any programming environment you want, including compilers, editors, documentation, and a world-wide community to support you, are available at your fingertips, for the price of a consumer-grade internet connection.

You can have everything; make use of it instead of whining about your college teaching you the 'wrong' things.

Answer 2

COBOL is a zombie lurking in backrooms of Enterprise whilst Fortran is almost a domain specific language these days anyway. Assembly is barely used outside of embedded and other performance conscious circles. Also only Fortran, C or Assembly are "metal" languages as such. COBOL, Lisp and the rest were and are high level programming languages.

It's also the law of the jungle at work. You've got really good veterans answering questions because they are good enough to have survived that long and not burn out. They may have been better educated in the sense that they had to contend with more direct problems of engineering than what shows up today but I can guarantee you that the courses being taught in my program are quite rigorous and the ones down the way at Georgia Tech are even more so in many cases. Tech for instance has you doing bit slicing in Java, teaching you the same thing you'd have learned with C or whatever. And they make you Unit Test it. Agile shows up earlier these days too. Academia is beginning to catch up the social coding methods of industry.

If there is any validity to "older is better" it would be one factor beyond experiential. It's mathematical. My CS department split from the Math department just over 10 years ago and until very recently had far more explicit math in the program. So you've got 40 year olds running around with physics and math degrees who happen to be programmers. They are problem solvers by training and nature, real engineers. I'm not knocking CS programs by any means but I'd wager the general level of training in mathematics and physics that permeated older programs rather than "pure" CS helps a lot in making "older is better". There's a certain pragmatism that shows up with engineering and science programs that doesn't show up with newer "CS" programs due to CS being a more theoretical discipline. I'm not saying it's "worse" per say just different priorities.

Lastly, why are you waiting around for teachers to teach you everything? If you do in fact love this discipline, then love. Go really learn how computers work on your own, no one is stopping you. If you want to learn Lisp or Fortran then do so. You have an internet connection. You are limited only by your motivation to be as good as you can be. I for one, plan to be as good as the veterans when I get around to that stage.

Answer 3

Grab a copy of The Art of Computer Programming. It's a favorite collection of books to sit on the bookshelf so people can feel like they're smart. If you actually open and read them in sections, however, it's the Rabbit Hole. First off, this work was started over 40 years ago, so you'll see some old stuff that has passed in time (MIX processor language has 6-bit bytes).

Aside from that, the nature of the book is that it's comprehensive in language, math, understanding, and approach. If you want to look at things at a lower level and understand how your modern languages relate to the processor, I feel like this is one of the best ways beyond what you get out of your degree program.

Now, for the order of learning things, I think that any functional modern language is a great start (functional as in useful, so Java & OO languages are in). When you really look at data structures, learning C is the next thing I think matters. Understanding the power of simple arrays, pointers, and the stack are there. Writing assembly is something you do because you must, because you think it's fun to be MEL, or if you're doing reverse-engineering / hacking.

Answer 4

Those other languages have been superseded for a reason. They're not better just because old people, or even good people or smart people, started with them. DMR didn't sit down and go "Gosh, you know what's a great language? Assembly. And I'm going to call on all my experience and knowledge to make that call." He used it because there was nothing else. And then it sucked so badly, he invented something else. OO exists and is the current dominant paradigm because non-OO languages suck. Take a hint from evolution.

Moreover, people don't choose their starter languages. A language isn't better just because some famous guy or some experienced guy started out with it. The overwhelming probability is that they had absolutely no choice in the matter, and even if they did, they probably wouldn't have known goto from polymorphism when starting out, making it a choice totally disconnected from their current status anyway.

And even if they did, you'd still have to ask them and get a solid justification before drawing any conclusion for yourself anyway.

As such, the fact that a language was used to start someone else is utterly irrelevant in the highest possible degree. You'd get more significant data deciding that your faeces look like a language's name, or the face of it's creator, or trying to divine meaning from the stars.