[reuven]

It's true that certain languages are particularly hot, and thus particularly in demand. The three that come to mind right away are Ruby, Python, and JavaScript, for which there seems to be unquenchable demand, even if the absolute number of such jobs is dwarfed by Java and C#. I'm sure that there is a shortage of good Erlang, Scala, and Clojure developers, as well.

But no matter what language you're looking to hire for, you're likely to come up short. That's because you don't just want someone who can write code in XYZ, but someone who can write amazing code, and interact well with the others in the company, and be sensitive to business needs, and communicate well. That combination is rare, no matter how popular and new (or unpopular and old) the language is.

Heck, everyone's favorite punching bag, Cobol, is still used in a huge number of businesses. I've seen in a number of places that if you want to make lots of money doing un-sexy work, you should become a Cobol specialist. Believe me, the businesses who need Cobol developers don't really care that Ruby is the new hotness; they've got billions of dollars invested in existing code that needs to be maintained and improved.

[humanrebar]

> That's because you don't just want someone who can write code in XYZ, but someone who can write amazing code, and interact well with the others in the company, and be sensitive to business needs, and communicate well.

This is really an issue of price points. If companies went around offering contracts for double-current-salary, full benefits, and guaranteed employment for 5 years, positions would be filled fairly quickly.

If the required price point is higher than a company can afford, maybe key goals, plans, and roles need to be reevaluated. As I say to my coworkers all the time, software project estimates are not very accurate, so if business will die without 10% returns on their software development investments (or being within 10% of schedule estimates), everyone's already in trouble.

> The three that come to mind right away are Ruby, Python, and JavaScript

That's highly dependent on the problem domain. Again, perhaps those are in demand because companies' expectations aren't keeping up with market prices.

[lectrick]

There's also lots of money to be made in pivoting/combining your career learned lessons to date and applying them to a different area.

For example, you could take all that great knowledge about unit testing you learned in the Ruby space, and jump into Cobol and in all likelihood you will not only be able to share a new thing or two to some Cobol folks, you might contribute to far better maintained Cobol code.

The new ideas tend to congregate to the language-du-jour.

Of course, most of the "new ideas" were done on Smalltalk and Lisp years ago, but I digress.

[humanrebar]

For the most part that's true, but (for example) unit testing isn't as important in Haskell as it is in Ruby (due to stricter typing rules). Even something simple like mocking an interface requires a lot of boilerplate in C.

That being said, there are probably some things that can be pivoted. The value is in having attempted them in pilot projects so that you already know what works.

[kozhevnikov]

Sold. Where do I learn Cobol? Is it web-scale?

[miguelrochefort]

Funny guy.