Are There Any Fun Tech Jobs Left?

er0ck asks: "My first job out of college was working for an Internet Startup. They gave me some books and told me to learn Perl. Our office was a refurbished factory, with lots of light and open space. Best of all, we could bring our nerf toys in to work (and use them!). Four months later, the company went under. Several dot bomb jobs later, I work for my state government. Is anyone still having fun at their tech job?" I think that with the economic downturn, more companies are concentrating on survival more than being "fun". Are there any "fun" tech jobs left, or have they all suffered from the Economic Darwinism of the early 21st century?

"[Government work is] steady work, but boring at times. (I don't think they'd approve of the Nerf guns). Without the pressure of staying in business, projects sometimes stagnate, leaving us with little to do. During these slow times, I help behind the scenes at NerfCenter.com; It's a fun site, and they are switching to Perl for their admin backend. It keeps my skills sharp, and wards off the boredom.

My questions to the Slashdot community are:

1. Can you have a fun tech job, without the worry of being suddenly unemployed?
2. If you are you forced (as I am) to get your fun on the side what are some good projects to get involved in?
3. What do you to unwind and have a bit of 'fun' in the workplace?"

Comments

Not trying to one-ups

[MeowMeow Jones]

But if you're on a computer, you can actually crawl through a new statement with a debugger. Here on MSVC, new calls malloc, delete calls free. Alot of programmers will write debug versions of all dynamically allocated objects, so that they can do primitive reference counting to see if there are any memory leaks. They're usually implemented with malloc and free as well.

Regarding sbrk() most manpages recommend that you dont use it (From FreeBSD:The brk and sbrk functions are historical curiosities left over from earlier days before the advent of virtual memory management.)

The original question is a bit of a double-edged sword (and might have even been intended as so) Although the only rational way to impement a default new() is {return malloc(sizeof(object));} it's irrelevant, because new is an interface. It's implementation should be considered a black box by the programmer. Even though new and delete almost definately use malloc and free, you can't make that assumption as a programmer.

That's the reason they go out of the way to state that calling free on a new'ed object leads to undefined behaviour. It'll work 99.999% of the time, but when it does bite you in the ass it'll take you weeks to find the bug.