3 ms·

Here's my problem with people who put in 80 hours a week: they don't write good code. At my first job - way back in the last millenium - we were creating a carr

Here's my problem with people who put in 80 hours a week: they don't write good code. At my first job - way back in the last millenium - we were creating a carrier class network device. I was hired to run the mail servers but ended up building an embedded linux that ran the PowerPC on our hardware - we also had a Strong ARM that ran a gig ethernet but that was under another guy. So, how did I go from mail server to an integral piece of the product? I took over for the guy who put a lot of time into not solving a problem. Before you can even get linux running on your hardware, you need a boot loader to initialize your hardware and get it into a state where linux will run. Our "hard worker" volunteered to write a boot loader and spent 60+ hours a week for the next 6 months writing it. But it never worked. Management decided to let me take a crack at the boot loader. Now, I was in my early 20's. I had bars to drink at, a girlfriend to get some lovin' from, and parties to go to. I didn't want to spend 60 hours a week at my desk. So, I did what any lazy hacker would do: I found an open source project that was close to what we needed. The project I started from was PPCBoot and it was started by Wolfgang Denk for the purpose of booting hardware running a Power PC processor. I spent 2 weeks telling everyone else in the company that I needed their help on X, Y or Z; getting them to write some code; and burning the new version of PPCBoot onto the flash chip. After 2 40 hour weeks, something amazing happened: it worked. It configured the hardware and handed control to linux which booted up. Anyway, that's my reason not to spend more than 40 hours at work in a week. All of the people who I've seen put in all those hours aren't really working. They're just playing with a neat project because they've always wanted to write a boot loader. The purpose of you job isn't to write code, it's to ship a product. If you keep that in the front of your mind at all times and focus your efforts on shipping product, you can "work" 40 hours a week and code a pet project on your own time.

ajross 15y ago
Honestly that anecdote just proves that you were a better engineer than the 60 hour boot loader guy. It doesn't say anything about hours other than that some people will work hard even if they aren't making progress. You aren't claiming that he'd have done better by spending only 2/3 the time, just that a solution (that he couldn't find) existed that fit. Honestly I find myself amused at the extent of the argument here. Most people I've known in the tech world who work long hours generally do so because ... they like it. It's a way to define your life. It's a way to feel good about yourself. I'm sure there are programmers trapped in death march dungeons somewhere for exploitative wages, but they are the exception, not the rule.
- jiggy2011 15y ago
  I think the point is more that having some sort of time constraint perhaps helps force you to come up with more realistic solutions. For example the number of startups with VC investment who seem to have spent 50%+ of their time creating a new NoSQL DB or something.
j_baker 15y ago
I agree with you on this one. Working long overtime doesn't just make you unhappy, it isn't productive either (although I certainly have to grant that it's necessary sometimes). The biggest problem is that it tends to become self-sustaining. You're overworked, so you code something poorly. Then, next time you work with that code, you have to work harder because it's so poorly written.
- lutorm 15y ago
  I wondered about this at a place I recently interviewed where the team said that I'd better be ready to put in 60h/week. I said that I don't think I get more done working 60 instead of 40h, and certainly not better work. They disagreed that it was a problem.

ajross 15y ago

Honestly that anecdote just proves that you were a better engineer than the 60 hour boot loader guy. It doesn't say anything about hours other than that some people will work hard even if they aren't making progress. You aren't claiming that he'd have done better by spending only 2/3 the time, just that a solution (that he couldn't find) existed that fit. Honestly I find myself amused at the extent of the argument here. Most people I've known in the tech world who work long hours generally do so because ... they like it. It's a way to define your life. It's a way to feel good about yourself. I'm sure there are programmers trapped in death march dungeons somewhere for exploitative wages, but they are the exception, not the rule.

jiggy2011 15y ago
I think the point is more that having some sort of time constraint perhaps helps force you to come up with more realistic solutions. For example the number of startups with VC investment who seem to have spent 50%+ of their time creating a new NoSQL DB or something.

j_baker 15y ago

I agree with you on this one. Working long overtime doesn't just make you unhappy, it isn't productive either (although I certainly have to grant that it's necessary sometimes). The biggest problem is that it tends to become self-sustaining. You're overworked, so you code something poorly. Then, next time you work with that code, you have to work harder because it's so poorly written.

lutorm 15y ago
I wondered about this at a place I recently interviewed where the team said that I'd better be ready to put in 60h/week. I said that I don't think I get more done working 60 instead of 40h, and certainly not better work. They disagreed that it was a problem.