Ask HN: What do you do when you don't have an itch to scratch?

https://news.ycombinator.com/item?id=2407697

by angrycoder • 15 years ago

9 1 15 years ago

news.ycombinator.com

So the saying goes that if are hunting for ideas about what to build, that you should scratch your own itch. My question is, what do you do if you are unsatisfied with the solving other people's problems, but don't have an itch of your own to scratch?

I have build a career out of scratching other people's itches. I've gotten away from the daily grind to a good extent by going independant a few years ago. In the end, it is still largely unsatisfying. It might be that I have been doing development for so long that I lack the foolishness or innocence necessary to accomplish anything on my own. When I have an idea, my brain is to wired to think of all the reasons why it wouldn't work, or that it would take to long, rather than possessing that go-go-go attitude.

I think what it boils down too is that I feel I have the skills and experience to accomplish just about anything, but lack the passion for any particular problem or idea.

bartonfink• 15 years ago

Hey, angry -

I'm not quite where you are (I haven't broken free and gone completely independent yet), but I sympathize with the feeling. I also tend to be very critical and don't very much like to solve problems I don't perceive as problems (e.g. writing yet another CRUD app).

My best guess, and this is honestly just a guess, is that this is a similar situation to a rut. The best way I've found to get out of a rut is to start saying "yes" to things I normally wouldn't do. That introduces a level of "randomness" to an otherwise stable system and can break me out of what I perceive as a local maximum.

Try coding something that you believe you understand but whose implementation will take some figuring out. For example, a couple of years I wrote a genetic algorithm that evolved an arithmetic expression to approximate a square root in C++. After a hell of a lot of wrangling, I got it to work but thought to myself that I spent more time fighting C++ than I did fighting the problem itself. That prompted me to start working with Lisp because Lisp is better suited for that problem than C++ was. I never had an itch to work in Lisp before, and that certainly wasn't my goal, but I think that the experiment was fruitful because it led to a challenge I genuinely want to meet.

You never know when you'll run into an interesting implementation wrinkle in an otherwise "solved" problem, and that can provide plenty of itches you might never have seen at a first glance.

tl; dr -

If you don't have an itch, and haven't found someone else with an itch you'd like, then you need exposure to more itches.