you know what none of it matters
10k is alot of money regaurdless, if i was offered 2 jobs with the same pay but one of them told me they would buy me a brand new honda cb600rr every year, i take the one thats giving away sweet ass motorcycles

honestly i just got in an argument with a guy today who was talking shit on cobol saying nonsense about "they just support legacy code for companies that don't wanna pay to move over to a real programming language" and then saw a couple people in this thread saying the same nonsense and saw red

This post was edited by Ideophobe on Jul 9 2016 10:39pm

i never disagreed that your college's self reported survey showed cobol developers making more money than java developers on average. i get the impression you're arguing something that nobody is disagreeing with.

or are you arguing about the usefulness of averages? here's a simple example of why i dont find it super useful to compare salary averages:

http://www.indeed.com/salary/Cobol-Developer.html
http://www.indeed.com/salary/Java-Developer.html

that site shows cobol developers make 14k less than java developers on average. but i'm earning more than that, and so do the people i know. so these averages aren't very useful for me. i would expect cobol devs to make more than just 88k.

This post was edited by carteblanche on Jul 9 2016 10:37pm

ya, i shoulda just stuck to facts that illustrated my point that cobol is more valuable than "just supporting legacy systems"
i edited that post out to say it doesnt matter but unfortunately it was seen before i pressed the button

i still think it's an interesting fact, and more useful than averages from things like a job directory because it's the pay that people actually got for the job that is expected to be preformed with a CS degree at the same point in their career.
i would never bother thinking about averages of salaries posted based on keywords because you're not comparing the same positions

I'd say

Java/C++ - General purpose applications
Python/shell scripts - useful for automating tasks
sql - standard for anything using a database
php - for the web, not that important tbh

Java code looks so much like C++ code that if you understand one you can read the other and know what's going on, so it really doesn't matter which you learn, so it just kind of depends on what your team uses.
Python serves the same function as Perl, which is an older and kind of dying language, but if it's a large tech company that's been around a while they might use Perl, so again it just depends on which your team uses.
The basics of SQL are really easy to learn, just focus on SELECT, INSERT, DELETE, and JOIN statements.
PHP is shitty.

/e If you're looking to get started I'd start with Python. It's very beginner friendly, powerful, and lots of resources to learn it. You can even start here:




This post was edited by Mastersam93 on Jul 10 2016 09:03am

Learn two things:
- SQL. It will give you great insight in how data works, and what the possible difficulties can be. As an IT consultant, I can tell you that many business analysts would be much better at their jobs if they had a basic understanding of relational databases.
- Anything that's object oriented. Your intention is not to become an expert who actually solves issues, so learning something specific like JavaScript is probably not useful. If I were you, I'd go with Python, C# or Java, simply because those give you a good understanding of how people write 95% of the software that your business uses. JavaScript is useless for this purpose, and C / C++ have a much too steep learning curve.

javascript is the future

get outa here

Straight wrong.

This post was edited by umeshieee on Jul 11 2016 03:28pm

Well yeah, nobody is actually pumping out any kind of apps in cobol anymore, let alone crappy ones. Cobol developers are needed to support huge legacy apps that are so ingrained that they essentially can't be replaced. But you are gonna be stuck working with decades-old software and technologies, always doing support instead of greenfield development, never learning anything new on the job, never trying out any cool new tech. You'll go to work and every day you'll do the same thing, day-in and day-out, a digital equivalent of flipping burgers. It pays very well, but who needs money when you blow your brains out from boredom in a couple years?

I'll take my modest $150k a year for doing Java/C# but actually enjoy going to work every day.



It's not complete nonsense. Cobol is certainly a real programming language, but it's true that companies pay through the nose for developers because they won't (or can't) replace the software with something newer. Cobol isn't special in any way, it's just not feasible to replace these millions of man-hours any time soon. The cost in both time and money is too high.
And I'm not ragging on Cobol, it's a programming language and it works. But there's a reason why languages evolve and new ones are developed. They support new features, they make developers more productive, they improve code quality though new tools, they improve performance, they improve re-use, maintainability, etc. Cobol works, but you wouldn't want to use it for new software when there are far better options available.

This post was edited by russian on Jul 11 2016 03:48pm

People hating on Cobol are idiots.

Yes indeed, you won't be doing green-field implementations. But that doesn't mean that you're limited to doing support.

In environments like these, implementing a new feature is challenging because of the enormous importance of keeping the code (and the software as a whole) maintainable. Think of software that's 30 years old, and that has had many features added to it through time. If developers are even a tiny bit lazy regarding code structure & guidelines, the whole thing turns into a complete mess and will probably implode sooner or later.

That's a challenge that's fun to work on for some people. In fact, I'd almost say that this type of programming is more "real" than modern Java / JavaScript / C# programming, because Cobol still directly requires you to go for the optimal solution at all times. In modern languages, there's many idiots who write spaghetti code that happens to get the job done.

Anyway, it's obvious that AspenSniper isn't looking for a language like Cobol. That language holds barely any relevance to 95% of the software industry.