5 ms·

This has happened way more times than I can count. Spending hours staring at a line of code, up all night. Wake up next morning, and it's a simple spelling mist

This has happened way more times than I can count. Spending hours staring at a line of code, up all night. Wake up next morning, and it's a simple spelling mistake I've missed. Grinding out work has never been productive. Mistakes happen when you're tired. This applies to study cramming as well, and it's completely pointless. Study as much as you can, until your NORMAL bed time, then stop and get a full nights rest before the test. Side note, it's no wonder these big games come riddled with bugs these days. Pushing developers to the extreme during crunch time, just to meet a deadline. There is NO WAY they are doing everything correctly.

brianwawok 10y ago
I think it does sometimes work, which is why at last I fall back to it. Grind out fix.. 90% success rate, average time to solve - 1 hour. Sleep and fix in the morning.. 100% success rate, average time to solve - 15 minutes. Clearly better to sleep but if something needs fixed NOW... then it's time to grind I guess...
- overcast 10y ago
  Well to be fair, we're not talking about emergencies, obviously those get priority. And of course, as much as people would like you to believe, most things are not an emergency!
- inglondon 10y ago
  What about the 3rd option, getting two tasks done? - Grind for 1 hour and assuming 90% success rate, finish the task 9 out of 10 times - Sleep - Finish another task in 15 min with, again based on your assumption, a success rate of 100% In start-ups there is usually a near endless list of tasks to solve so this would be the most productive approach. Again, based on a lot of assumptions. [edit: formatting]
  - st3v3r 10y ago
    Except that success rate was assuming that you did get the rest you needed, mainly from leaving at a reasonable time. Staying and grinding directly contradicts that, so you're not going to be rested and well in the morning, and you're not going to have the success rate of 100%. Basically, you just advocated for crunch time and death marches. Things that are known to not work.
  - jimcsharp 10y ago
    You haven't prepped and then slept on task 2 so I don't think you're finishing it in 15 minutes. And now you're sleep deprived.
- st3v3r 10y ago
  In reality, though, how often does it need to be fixed NOW? Like, actually needs to be fixed now, or there will be serious consequences, and not just some manager thinks they want it now? I'd be willing to bet it's less than 10%.
  - overcast 10y ago
    Reality is basically never, unless it's a major system outage, or security issue. We don't even bother looking at issue priorities, because everyone raises them as critical.
    - PaulRobinson 10y ago
      Even then, is it really an emergency? What's the consequence of it being down? Death? People being fired? IME, 90% of emergencies are "such-and-such a manager is going to look a little bad in a spreadsheet next Monday". That doesn't mean it doesn't need fixing, but we should start being honest with ourselves. Defining P1, P2, etc. is useful.
      - overcast 10y ago
        Well in the case of Amazon, $66,000 per MINUTE. If you want to make a case for money only being money sure. But as far as business is concerned, that is an emergency.
softawre 10y ago
Spelling mistakes are the worst! You need less stringly-typed things in your life.
awinder 10y ago
There's a lot of science behind this phenomenon, this article does a good job of discussing the process https://www.psychologytoday.com/blog/your-brain-work/201209/stop-trying-solve-problems https://www.psychologytoday.com/blog/your-brain-work/201209/... But tl;dr your brain is adept at working on problems in the background after you've defocused on them. Your work is still getting done, even if you cut out of work and unwind, no guilt necessary ;-)

brianwawok 10y ago

I think it does sometimes work, which is why at last I fall back to it. Grind out fix.. 90% success rate, average time to solve - 1 hour. Sleep and fix in the morning.. 100% success rate, average time to solve - 15 minutes. Clearly better to sleep but if something needs fixed NOW... then it's time to grind I guess...

overcast 10y ago
Well to be fair, we're not talking about emergencies, obviously those get priority. And of course, as much as people would like you to believe, most things are not an emergency!
inglondon 10y ago
What about the 3rd option, getting two tasks done? - Grind for 1 hour and assuming 90% success rate, finish the task 9 out of 10 times - Sleep - Finish another task in 15 min with, again based on your assumption, a success rate of 100% In start-ups there is usually a near endless list of tasks to solve so this would be the most productive approach. Again, based on a lot of assumptions. [edit: formatting]
- st3v3r 10y ago
  Except that success rate was assuming that you did get the rest you needed, mainly from leaving at a reasonable time. Staying and grinding directly contradicts that, so you're not going to be rested and well in the morning, and you're not going to have the success rate of 100%. Basically, you just advocated for crunch time and death marches. Things that are known to not work.
- jimcsharp 10y ago
  You haven't prepped and then slept on task 2 so I don't think you're finishing it in 15 minutes. And now you're sleep deprived.
st3v3r 10y ago
In reality, though, how often does it need to be fixed NOW? Like, actually needs to be fixed now, or there will be serious consequences, and not just some manager thinks they want it now? I'd be willing to bet it's less than 10%.
- overcast 10y ago
  Reality is basically never, unless it's a major system outage, or security issue. We don't even bother looking at issue priorities, because everyone raises them as critical.
  - PaulRobinson 10y ago
    Even then, is it really an emergency? What's the consequence of it being down? Death? People being fired? IME, 90% of emergencies are "such-and-such a manager is going to look a little bad in a spreadsheet next Monday". That doesn't mean it doesn't need fixing, but we should start being honest with ourselves. Defining P1, P2, etc. is useful.
    - overcast 10y ago
      Well in the case of Amazon, $66,000 per MINUTE. If you want to make a case for money only being money sure. But as far as business is concerned, that is an emergency.

softawre 10y ago

Spelling mistakes are the worst! You need less stringly-typed things in your life.

awinder 10y ago

There's a lot of science behind this phenomenon, this article does a good job of discussing the process https://www.psychologytoday.com/blog/your-brain-work/201209/stop-trying-solve-problems https://www.psychologytoday.com/blog/your-brain-work/201209/... But tl;dr your brain is adept at working on problems in the background after you've defocused on them. Your work is still getting done, even if you cut out of work and unwind, no guilt necessary ;-)