4 ms·

Sometimes I just wonder, is the frequent burnt-out we see for programmers caused by not having a fast feedback loop. For me, I feel excited when doing performa

Sometimes I just wonder, is the frequent burnt-out we see for programmers caused by not having a fast feedback loop. For me, I feel excited when doing performance tuning work, front-end stuff etc. However, I would quickly become bored after working on work that take more than 5 days or so, even if I love the subject. Although I'm not really into TDD, but I think sometimes using TDD could help you get some more feedback when you work on some abstract stuff. (Although writing the tests could be boring for the beginning...) I would argue that having no expectation on the first-draft may not be the best way to be more productive. Having some concrete goals or metrics could be beneficial if you can get the feedback-loop working. Under-defined, subjective goals could be bad as achieving the goal may seem never-ending if you want perfection. (at least for me)

Ottolay 6y ago
Interesting. I love hunting and fixing significant bugs or issues. It could be because the immediate dopamine hit if successful. Harder to get motivated for things with less certain impact, even if successful.
bonniemuffin 6y ago
Some other fields (e.g. long-running academic research experiments) involve a feedback loop of months or years. I've done experiments where the experiment subjects (mice, human cells, etc) look exactly the same for weeks and weeks, and you have to work on them every day not knowing whether they'll turn into something useful in a month or two. Writing grants and papers has a turnaround time of months--you work so hard on it, lob it into the ether, and just wait 6 months to see if anything will come of it. Coming from a background like that, I find that the days/weeks feedback loop time in tech makes me positively giddy with joy. It's hard for me to imagine thinking of 5 days as a long time to work on a problem.
LZ_Khan 6y ago
Totally agree, at least anecdotally. Working in ML for a large company has really tested my liking for programming. The amount of plumbing/testing/code reviewing needed just to get a tiny feature usable for a ML model is absurd. Taking 6 months to push one model improvement is really draining. The pay's nice though.

Ottolay 6y ago

Interesting. I love hunting and fixing significant bugs or issues. It could be because the immediate dopamine hit if successful. Harder to get motivated for things with less certain impact, even if successful.

bonniemuffin 6y ago

Some other fields (e.g. long-running academic research experiments) involve a feedback loop of months or years. I've done experiments where the experiment subjects (mice, human cells, etc) look exactly the same for weeks and weeks, and you have to work on them every day not knowing whether they'll turn into something useful in a month or two. Writing grants and papers has a turnaround time of months--you work so hard on it, lob it into the ether, and just wait 6 months to see if anything will come of it. Coming from a background like that, I find that the days/weeks feedback loop time in tech makes me positively giddy with joy. It's hard for me to imagine thinking of 5 days as a long time to work on a problem.

LZ_Khan 6y ago

Totally agree, at least anecdotally. Working in ML for a large company has really tested my liking for programming. The amount of plumbing/testing/code reviewing needed just to get a tiny feature usable for a ML model is absurd. Taking 6 months to push one model improvement is really draining. The pay's nice though.