4 ms·

I don't think that many people working 80 hours a week for months can be just as good as people of the same skill level working only 40.

matwood 10y ago
There are times in a startup when it is a race. Those working 80 will beat those working 40. Can those working 80 do that forever? No, but if it helps them put the ones working 40 out of business they do not need to work 80 forever. I think the real key is recognizing when working more will help.
- samastur 10y ago
  An example would be nice as I cannot think of any.
  - jasode 10y ago
    Basically most of the companies mentioned in Founders at Work.[1] There are several common themes to why all those companies had to put in crazy hours but one theme that sticks out is the "oh shit, the user count is growing like crazy and we've got to scale out our servers to handle the exponential workload like YESTERDAY or we're going to DIE!" It happened to the Youtube engineers, Google, Napster, Etsy, etc. It's a stressful combination of fighting fires and adding capacity at the same time. For horizontal scaling re-architecting & re-engineering, there is no commercial off-the-shelf or open-source software that let's you install it in one afternoon such that all programmers can just work 40 hours a week without a hiccup. Today, you can't just install a distributed database like Cassandra and fire up more AWS instances and assume those newer technologies that the previous Founders at Work didn't have will prevent your employees from putting in 80 hour weeks for months. Going from a single instance database to handle 10k to 100k users to a sharded distributed db to handle 100 million is not going to be a 40-hour a week job for startup employees. If any companies have done it with zero overtime, please write a blog highlighting your user traffic growth, database sizes, and migration strategy. That would be a very impressive achievement. [1] http://www.foundersatwork.com/interviews.html http://www.foundersatwork.com/interviews.html
    - samastur 10y ago
      You are right, thanks. I have misunderstood it as being a race against some other company/team.
wmeredith 10y ago
I agree completely. I've been is software for over 10 years now and barring a few extreme outliers most people do not have 80 productive hours in them every week unless they literally do nothing else. And even then, it cannot last. There is a place for hard work. And it is certainly a factor in success, but working nights and weekends only gets you so far and there is a significant diminishing return. Go through a code base and look at commits made past midnight, or 8+ hours into a shift. They are going to lead to more regression issues and technical debt nine times out of ten. Considering all the tone-deaf ham-fisted nonsense that's come out about Mayer's management style while she was at yahoo, I'm surprised people will still listen to this nonsense. It's just a bunch of people following their selection bias into a cult of personality.
latch 10y ago
I hear this argument/fact often, and while I'm certain that it's as true in the aggregate, I'm also certain that it ignores outliers. There are people who can be productive (and happy) working more than 40 hours a week.
- samastur 10y ago
  Sure, but doing what? And can you get a whole team of them? I very very much doubt you can when it comes to programming. I've been doing this for almost two decades and in this time have only known ONE person who might fit the bill.
  - latch 10y ago
    I dunno, early Id Software seems like a good example. Personally, I think it directly relates to having people do something their passionate about. 20, 40, 80 hours...doesn't matter, getting a group of programmers excited and passionate about getting a taxi from point A to point B is probably unlikely. But get a group of people who grew up playing D&D and tell them to make it real...
bluedino 10y ago
Sometimes you don't have a choice. Pierre Omidyar worked on eBay on weekends and in the middle of the night to keep the site running because of growth and troubleshooting problems.

matwood 10y ago

There are times in a startup when it is a race. Those working 80 will beat those working 40. Can those working 80 do that forever? No, but if it helps them put the ones working 40 out of business they do not need to work 80 forever. I think the real key is recognizing when working more will help.

samastur 10y ago
An example would be nice as I cannot think of any.
- jasode 10y ago
  Basically most of the companies mentioned in Founders at Work.[1] There are several common themes to why all those companies had to put in crazy hours but one theme that sticks out is the "oh shit, the user count is growing like crazy and we've got to scale out our servers to handle the exponential workload like YESTERDAY or we're going to DIE!" It happened to the Youtube engineers, Google, Napster, Etsy, etc. It's a stressful combination of fighting fires and adding capacity at the same time. For horizontal scaling re-architecting & re-engineering, there is no commercial off-the-shelf or open-source software that let's you install it in one afternoon such that all programmers can just work 40 hours a week without a hiccup. Today, you can't just install a distributed database like Cassandra and fire up more AWS instances and assume those newer technologies that the previous Founders at Work didn't have will prevent your employees from putting in 80 hour weeks for months. Going from a single instance database to handle 10k to 100k users to a sharded distributed db to handle 100 million is not going to be a 40-hour a week job for startup employees. If any companies have done it with zero overtime, please write a blog highlighting your user traffic growth, database sizes, and migration strategy. That would be a very impressive achievement. [1] http://www.foundersatwork.com/interviews.html http://www.foundersatwork.com/interviews.html
  - samastur 10y ago
    You are right, thanks. I have misunderstood it as being a race against some other company/team.

wmeredith 10y ago

I agree completely. I've been is software for over 10 years now and barring a few extreme outliers most people do not have 80 productive hours in them every week unless they literally do nothing else. And even then, it cannot last. There is a place for hard work. And it is certainly a factor in success, but working nights and weekends only gets you so far and there is a significant diminishing return. Go through a code base and look at commits made past midnight, or 8+ hours into a shift. They are going to lead to more regression issues and technical debt nine times out of ten. Considering all the tone-deaf ham-fisted nonsense that's come out about Mayer's management style while she was at yahoo, I'm surprised people will still listen to this nonsense. It's just a bunch of people following their selection bias into a cult of personality.

latch 10y ago

I hear this argument/fact often, and while I'm certain that it's as true in the aggregate, I'm also certain that it ignores outliers. There are people who can be productive (and happy) working more than 40 hours a week.

samastur 10y ago
Sure, but doing what? And can you get a whole team of them? I very very much doubt you can when it comes to programming. I've been doing this for almost two decades and in this time have only known ONE person who might fit the bill.
- latch 10y ago
  I dunno, early Id Software seems like a good example. Personally, I think it directly relates to having people do something their passionate about. 20, 40, 80 hours...doesn't matter, getting a group of programmers excited and passionate about getting a taxi from point A to point B is probably unlikely. But get a group of people who grew up playing D&D and tell them to make it real...

bluedino 10y ago

Sometimes you don't have a choice. Pierre Omidyar worked on eBay on weekends and in the middle of the night to keep the site running because of growth and troubleshooting problems.