[UK-Al05]

Well it depends. You could probably never stop learning the tools or the new languages; they're continually re-invented. However you can certainly become bored with it; learning new tools and languages becomes a skill in its self. At which you think i've learned a new tool, so what?

[jgw]

I have no beef with someone getting bored of what they're doing, but then they should say just that. To say "I know everything there is to know about programming" after programming for a year is laughable.

You're right that if you know C++ and you learn Java, you're not really learning all that much. But if you've spent your entire life programming in those types of language, and then learn Lisp, or Forth, or even just assembly language, your entire mental model of computation is turned on its head. Heck, learning C would be an eye-opener for someone who has known only Java.

You can go even further. My own trade is ASIC verification - writing testbenches to test functional correctness of chip designs. I've done a bit of FPGA design, too. I've chatted to software guys far above my humble skill level who don't grok either of those two domains because they're completely foreign to their way of thinking. But I'd file both under the broad umbrella of "programming".

I think you can learn constantly for a lot longer than a year or two without learning any new languages or tools, just working on different project, in different domains, using different paradigms. [EDIT:clarification]

[jiggy2011]

They probably didn't mean that they "know everything there is to know" unless they are very arrogant.

Perhaps they meant that they knew the syntax to at least one programming language and where at a point where they felt it would be easy to learn more if needed but they no longer felt the need to study it as an end in itself for their purposes.

having said that it's easy to overestimate your level of knowledge when you haven't been introduced to anything more difficult.

You may believe for example that you can easily solve say traveling salesman because you figured out how to write the naive n! solution and tested it with 5 nodes on a fast computer and it worked fine. If you never have to test it with 1000+ nodes then you may never find the performance inadequate so will never have to open the dynamic programming rabbit hole.