Who is the first programmer in the world?

To make a strong general programmer career, what kind of programming knowledge should we pursue?

• Disclaimer: I'm asking this question because I just got declined a job in a big gaming company, after a short interview process. It's a job I was positive I could do, being a professional programmer for over 15 years. Maybe I simply don't have enough knowledge indeed, like they implied. So I want to improve. But how? Have you watched http://www.imdb.com/title/tt0119794? It's a movie about a programmer in an age in which programmers don't even code any more. They just build the virtual world. Almost like architects, but with knowledge of bytes instead of physics. At least, that's how I recall it. I think the most important skill a programmer must have is self-learning, and the most important trait is patience. With that given, what should we pursue to know? It's certainly not learning a specific language that will make you a great programmer. We need to be able to code in any. Nor will you be able to build the first AI program just because you made the biggest iPhone hit app. Practical accomplishments such as this may mean nothing about how good you did it. It can be just luck, even many times. I just told to a friend who wants to begin to code (he's over 30 years old and never did anything) what I'd guess he should focus on (for our age at least): basically mathematics and programming logic. Matrices, vectors, functions, algorithms and performance optimization. I now guess it's also important to know eletronic basics, how a circuit works, etc, but that would be a second step. But is it? Or am I missing something? Not every school will teach you those. Certainly mine didn't.

• Answer:

  Well I can see the irony behind your question as you mentioned 15 years of experience and still cannot figure out what should be the quick learning path. You must know a 4GL, get the syntax of it, do play with it to solve multiple problems - simple to complex and always need to feel the passion to thrive for the best possible code you can produce. Once the logic building becomes like you can resolve anything on the fly not even needing to look at 90% of the syntax and semantics hitting the HELP button I guess you are ready for patterns and various design paradigms such as Gang of Four. Once you can apply your code with any design or pattern then you can look into the big picture where interfacing with databases, networks, different OS etc. comes into play. By this time, you should be able to look at codes written in any other language and still be able to find flaws, debug efficiently, develop with ease by the prolonged practice that you earned through the time you spent for a strong foundation. It is not easy, time consuming but it should always be fun, if the fun is gone then you may not be cut for it or that would be time out. About the interview dilemma I would consider that few things most companies do not understand is memorization of common interview questions does not prove anyone a good programmer. A rather better approach for interview could be given a problem, ask the candidate to resolve it, this can be different parts of SDLC related with the company culture of development process. The resolved answers then can be analysed to find out who could be appropriate for the position - along with soft skills and other factors. The reason I say that cause in my field most job offers are recruiter controlled and contractual. That way companies end up paying huge hourly rates for consultants who most cases do not even go close to 30% of doing whatever the responsibilities, by themselves and still end up earning 70% of the rate paid the least. Due to contract issues there is always a huge glitch of losing it or lingering it further - which gets higher priority than getting the job done accurately. Ratio of implementation success of my specialized area is bleak due to this surprising factor making it volatile - not only in the US but in all corners of the world.