3 ms·

Wut. Time off does not correlate to dev quality. I, and many others, do their best work outside of the normal parameters. It sounds like you think if Twitter e

by lstyls 9y ago

Wut. Time off does not correlate to dev quality. I, and many others, do their best work outside of the normal parameters. It sounds like you think if Twitter engineers worked harder Twitter would be doing better. Besides your anecdotal observation about laid back engineers, is it not plausable that the company doesn't have a vision that contributes to monster growth?

burnte 9y ago
It certainly does have a level of correlation. The less time you spend "at work" (for whatever definition you care, including work from home/the moon) the less work you get done. Now, no, that doesn't translate to a perfect linear growth rate where working 20 hours a day is twice as good as 10 hours a day, there is a limit you can push people. People need vacation and time away. However, it's a rock star who can be as/more productive in three days a week than a similar guy working 5 days a week most weeks. I'll even grant some people can cram for three days a week and be as productive as a 5d/8h guy, but in the end you still need to put in the hours. Note, this comment doesn't allow for absurd comparisons such as "Forrest Gump coding 8 days a week will never match Donald Knuth coding 1 hour a month" or whatever. Ass comparisons should use reasonable definitions.
- lstyls 9y ago
  Not sure what an "ass comparison" is. I didn't argue that there isn't a correlation (never used that word) but correlation != causation. In case it's not clear, I really doubt that Twitter being an also ran has to do with a culture of laziness. If anything my impression was that they overengineered without making any real product changes.
- lmm 9y ago
  I think the breakeven point is much earlier than you suggest. I'd guess something like 28 hours/week is peak productivity; more than that and you're spending more effort on looking productive and being in the office and getting less done.
guiambros 9y ago
> Besides your anecdotal observation about laid back engineers, ... Who said "engineers"?

by lstyls 9y ago

burnte 9y ago

It certainly does have a level of correlation. The less time you spend "at work" (for whatever definition you care, including work from home/the moon) the less work you get done. Now, no, that doesn't translate to a perfect linear growth rate where working 20 hours a day is twice as good as 10 hours a day, there is a limit you can push people. People need vacation and time away. However, it's a rock star who can be as/more productive in three days a week than a similar guy working 5 days a week most weeks. I'll even grant some people can cram for three days a week and be as productive as a 5d/8h guy, but in the end you still need to put in the hours. Note, this comment doesn't allow for absurd comparisons such as "Forrest Gump coding 8 days a week will never match Donald Knuth coding 1 hour a month" or whatever. Ass comparisons should use reasonable definitions.

lstyls 9y ago
Not sure what an "ass comparison" is. I didn't argue that there isn't a correlation (never used that word) but correlation != causation. In case it's not clear, I really doubt that Twitter being an also ran has to do with a culture of laziness. If anything my impression was that they overengineered without making any real product changes.
lmm 9y ago
I think the breakeven point is much earlier than you suggest. I'd guess something like 28 hours/week is peak productivity; more than that and you're spending more effort on looking productive and being in the office and getting less done.

guiambros 9y ago

> Besides your anecdotal observation about laid back engineers, ... Who said "engineers"?