Developer == Long Hours

[Elle Atechsy]

Hi All,

What is the norm when it comes to developers having to work long hours a lot of the time?

I've only worked in IT at one company, in one group. Most others on my team who have worked in other companies claim that most developers they know don't work overtime so often, as we do. So, they blame the need for long hours here on bad management.

I work for a financial firm, which is different from something like product or .com companies. But, for the most part, what is a developer's work load really like in other environments out there?

[Deepak Bala]

I had wanted to post this question for a very long time. In general I was looking for the kind of working hours put in by employees around the world and how it varies in the US, India, Australia etc etc.

I work for about 10 hours a day. When its crunch time I have to sacrifice Saturdays as well (May be 6 hours). When the work load is heavy or there is an emergency, work can extend to 12 or 14 hour days + the Saturdays that i mentioned. Work timings are flexible most times. You can come in to work at 11 or 12 in the morning if you choose. But leaving early is the problem, since almost always you have to speak to some one over the phone in another country before you can leave. Add day light saving changes to that and it can leave you frustrated (I cant wait for march ). I dont enjoy days when I have finished my work but I have to stay back for a phone call. Damn !

How many hours do you put into your work Lulu ?

[Srikanth Raghavan]

Interesting...

I don't like to work for more than 9 hours simply because I will exhausted by that time. I am so curious to know how people work for 12 and 14 hours at a stretch.

And I hardly worked on Saturdays, may be 1 or 2. And I always finished my work on time, mostly before time. So, I am worried as to whether I am doing very less work and losing the oppurtunity to learn more or I am just a genious who does everything on time. (which I think is not the case).

Any comments?

-- Srikanth

[Elle Atechsy]

How many hours do you put into your work Lulu ?

John, I tend to work anywhere between, 9 to 11 hours a few times a week. Weekend day(s) only when required or during critical crunch times.

[Mark Herschberg]

I'ved worked in startup companies. I've done regularly 60 hours weeks and have had them run into the 80s when I was younger. When I run teams these days I try not ot let them get above 50. Working in a major corporation, I saw people putting in 8 hours days. Government work is notorious for strict 40 hours work weeks. My friends in investment backing, including those in IT, can regularly work 60-80 hours weeks, longer during big crunch times. In short, there is no norm in the US, but there are trends by industry here.

--Mark

[Tim Holloway]

Hours are one of the metrics by which I rate a (potential) employer. As a salaried non-union employee - and I've never worked in any other kind of shop, I don't expect every day to be uniformly 8 hours. Sometimes a project deadline or an emergency requires me to work late and/or on weekends. I understand. Some of those emergencies have been of my own making. I grit my teeth and soldier through. Any decent employer has a comp-time policy where I can rest and recover my life when things are less hectic.

However, I have occasionally been pressured to work extended schedules as a matter of course. My response has been and continues to be: I'm being paid salary "X", which is presumably based on a 40-hour work week. If I routinely am expected to work extended work weeks at the same compensation level, that's cutting my effective hourly compensation. If I really felt obligated to work 12-16 hours a day, I'd much rather just get a part-time job at the local convenience store and a full-time job at a lower-paying IT shop. That way, I could get the same net pay but at least I'd get a change of scenery every few hours.

I work to live, not live to work. If I ever get that hard-up for an IT job, I'm quitting and going to work at Radio Shack. Aside from everything else, after 25 years, I've learned what times of day my different types of productivity ebb and flow, and that simply being physically present isn't going to make me any more creative or productive. In fact, by eating into my refresh cycles, it may do the opposite.

Of course, there are the "hamburger grinder" shops, who don't care about my personal productivity, since it's only gross-hours-on-job that they consider important, not the product itself. But I try to avoid those places for another reason as well - the fun projects aren't located in such places.

[Mike Isano]

I average 60 hours a week and often work on the weekend. Once and a while I do an 80 hour week.


I get paid based on a 40 hour week. There are rediculous deadlines from our client. Sometimes I feel overworked and underpaid.

[Tim LeMaster]

Don't blame your clients, blame your management for accepting unreasonable deadlines.

I blame the need and/or pressure for extremely long hours on management nearly all of the time. As someone else stated my salary is based on normal 40 hour work weeks - with some crunch times. I've noticed a trend in corporations to make all projects all crunch all the time. This is unacceptable, you can't work at 150% all the time - thats what leads to burn out, mistakes, and apathy.

[Deepak Bala]

I'm quitting and going to work at Radio Shack

haha !

all projects all crunch all the time

Thats true. Some organizations do that. Why cant they understand that when people work for more than 8 hours they are no longer productive ? When it is an emergency, once in a while, I can deliver by working for 14 hours a day for may be 2 weeks.

Another thing that promotes long work timings is when people stare at you for leaving early, even though your work is complete. Some people can take it while others cant, which forces them to stay late just so it looks like they are working.

[Mark Herschberg]

Originally posted by Tim LeMaster:
Don't blame your clients, blame your management for accepting unreasonable deadlines.

Don't simply blame your management, blame yourself, too. Yes, management should know better. (Note: I'm not talking about the occasional push for a long work week to hit a deadline, I'm talking about extended long hours.) But if you are complicit and simply go along and passive-agressively complain, you're not helping the situation any.

If management picks an inferior database for your project, you have an obligation as an engineer to speak up and let them know why it's inferior. Likewise, if management is mismanaging your project by making the team work long hours, you have an obligation to speak up. In both cases *maybe* there is justification, but you should make sure that the justification to management is also sufficent justification to yourself.

More to the point, if management is so bad (as I often hear people complain), they should be "punished" with a high turnover rate, i.e. you should leave and work elsewhere. (For those interested in learning more about the culture of long hours and how to handle it, I strongly recommend Ed Yourdon's Book, Death March.)

--Mark
[ February 14, 2007: Message edited by: Mark Herschberg ]

[Tim Holloway]

You laugh when I say I'll go to work at Radio Shack. I'm serious. I effectively lived in the college DP labs, spent innumerable hours reading technological publications when others had social lives, have made a lifelong effort to be at the leading edge in learning and applyin new technologies. If that isn't worth something, I sure as heck ain't giving it away. Better to program for fun and earn my living doing something mindless than see my sweat treated like it's just salty water.

A number of years ago, I had the Job From Hell. Not as hellish as many I know have had, because I won't stand for it, and I have enough creds to be able to draw the line. However, this was one of those places where they like them young, cheap, and fearful. They were constantly pushing for "extra effort". After a while everyone got so blas�, they started pushing for extra extra effort. I kid you not.

A less hellish job than that had the boss explaining to our IBM rep that they needed to "educate" our customers to keep their systems running at 100% CPU utilitization. The IBM rep gently pointed out that in hardware, 100% means 100%, and if you get a flood of unexpected interrupts or the like, the lack of reserve capacity could trigger a system "train wreck". OK, so it was OS/2 we were talking about, but still...

Good generals don't commit 100% of their forces. They keep reserves. Good managers don't overcommit or overschedule either. Murphy doesn't sleep. I've made no few superiors unhappy over the years by giving accurate time estimates in place of the ones they really wanted to hear, but someone has to. Enough IT projects fail as it is.

I guess what really hurts is that our employers often demand so much in the way of work, yet demand so little in the way of quality. If you're going to kill yourself on a project, the delivered product should be higher calibre than the grossly sloppy stuff I see everyday at major corporate websites.

[Deepak Bala]

You laugh when I say I'll go to work at Radio Shack

Wow. I thought you were kidding.

delivered product should be higher calibre than the grossly sloppy stuff I see everyday at major corporate websites.

++

Enough IT projects fail as it is.

++

Getting work done faster is what I like to do. First you plan, then you execute and then you test. Once you plan well for something the rest is good to go (unless you have some kind of last minute requirement). With a reduced time frame its hard to do any planning at all. Sigh !

Don't simply blame your management, blame yourself, too.

People who stand up to it get blamed with having an "attitude problem" I did this once and got stamped with the "attitude problem" brand. I will continue to do it if that is what is slapped down on me. I left the project anyway. You are right about the "stand up for your rights" part, it can get people in trouble but working long hours because of crazy deadlines is far worse. What makes it even worse is if you dont learn much out of the experience. Reminds me of an article about toxic workplaces. If i can only remember the link !

[Ramna Reddy]

Another thing that promotes long work timings is when people stare at you for leaving early, even though your work is complete. Some people can take it while others cant, which forces them to stay late just so it looks like they are working.

John,You are absolutely correct.

[Rohit Nath]

I work to live, not live to work.
Period.

[Jan Cumps]

I'm working 4 days per week, 8 hours per day.
I have a fixed non-working day.

I do make exceptions when traveling or during go-lives,
but don't make exceptions to meet other peoples deadlines.
I decline team meetings scheduled on my non-working day.

[Richard Parker]

Yikes! - You guys are scaring me.

I currently work at a "cushy state job". I come in when I like, leave when I like - just so long as the job gets done on time. Heck - I even go work out for an hour in the middle of the day. If a deadline seems unrealistic, I speak up and things get adjusted. Sometimes I work late - but its not a big deal.

However - I'm getting bored and would like to look for a more challenging job in another part of the country.

I'm worried about getting a "hellish" job where they make you work 14 hour days. There's absolutely no way my psyche can handle that. I'd have a breakdown for sure.

[tambi durai]

Originally posted by John Meyers:

Another thing that promotes long work timings is when people stare at you for leaving early, even though your work is complete. Some people can take it while others cant, which forces them to stay late just so it looks like they are working.

Try leaving quietly, without saying goodbye to anyone; just leave as if you are going for a coffee. Don't carry any bag, coat; leave that in your car. Most people won't notice that you have left early...at least not immediately...

[Rohit Nath]

8 hours per day, 5 days a week should be more than enough for an IT job.
You got to have a life of your own as well right! But once we are into it then it seems quite unreasonable anyways... but I think total of around 40 hours per week should be good. I really dont appreciate when someone requests me to comedown on weekends as well! I hardly remember working on a weekend for last 2.5 yrs except if there is some event in the organization that I am organizing or part of. Workin on weekends is my last priority! Dont know if its good or bad but I like it that way and it remains my way!
I think 5 days a week are enough to get the job done! Efficiency? Well that my opinion and who cares anyways....!
[ February 19, 2007: Message edited by: Rohit Nath ]

[Richard Parker]

Rohit, thanks for your empowering post!
That makes me feel better!

[Deepak Bala]

Try leaving quietly

Unfortunately some one invented the cell phone. You manager will call you back in to office. If your cell phone is switched off you will be asked about it next morning and it kind of becomes like a nag thing. However I have the liberty of leaving early if I so wish. My work timings are not crazy. But I know of some developers whose work timings are crazy AND they dont get to learn as much as some of the others do. That sucks.

[Mark Herschberg]

Originally posted by tambi durai:

Try leaving quietly, without saying goodbye to anyone; just leave as if you are going for a coffee. ;) Don't carry any bag, coat; leave that in your car. Most people won't notice that you have left early...at least not immediately...

More to the point, this addresses a symptom, not the root cause and in the end is unhelpful.

Originally posted by Rohit Nath:
8 hours per day, 5 days a week should be more than enough for an IT job.

That seems pretty aribitrary. I would argue that a 44 hours work week would gives you roughly a 10% improve because i think for 99% of the population those extra four hours don't really come at a cost of effiency. I' not arguing that we should work 44 hours weeks, just that "40 hours a week" as you claim is an aribtrarily number dating from historical reasons.

The extension of the 40 hour work week is not unique to IT. I've seen people in all walks of life put up with it, including marketing, sales, accounting, etc.

As for other industries... Consulting firms often have people work 60-some hours work weeks. Ask a doctor at a hospital about their work week (a few years back residents threatened to unionize because of the issue of long hours). Lawyers? Forget it, if you're at a major form give up a life for the next 20 years. Know an academic trying to get tenure at a top 50 school? See how much they work. I-banking? It's notorious for long hours and burn outs. Chat with someone in fashion or entertainment, especially someone with less than 10-15 years of experience--it's like being a pledge at a fraternity complete with very long hours. Heck, ask a high school teach how many hours she works!

In the US, the 40 hour week, outside of government work and government related industries is pretty much a thing of the past for salaries employees. I'm not saying it doesn't exist at all, but it's rare, and has nothing to do with IT.

--Mark

[Elle Atechsy]

Thanks, Mark. That has been my thought for a long time. I have a co-worker that teases me about working overtime because he refuses to do so, outside of prod releases & emergency prod support. So, certain things just don't get done or get done late.

The funny thing is he's married to a doctor. She works very many hours. I asked him one day, 'What is the difference between your wife working 80 hours a week as a doctor for humans and me working 50 hours a week as a doctor for computers?' His answer was that his other developer friends in other groups/companies don't work so many hours, so he doesn't understand why any of us should have to. Grant it, human life & well being is 100% more important than a computer program's, but society cannot survive without properly run technology, either.

[Jesus Angeles]

I think it is ok if we work late,, as long as it is our 'plan'.

If it is spontaneous (that we cant afford) or kicks us off our planned path, then we should control ourselves from doing it.

Since we are talking about work, then I mean 'career path'.
[ February 19, 2007: Message edited by: Jesus Angeles ]

[Elle Atechsy]

Since we are talking about work, then I mean 'career path'.

How many of us believes that willingness to work late & hard is based on if we consider what we do "just a job" vs. "a career"?

For myself, I consider what I do a career, and feel, if I want to move ahead, I need to convince my managers that I'm willing to do what it takes to get he job done, as long as it's not ridiculous. What I've noticed in my company is that people don't get noticed for just working 40 hrs a week, because that means that they are only doing what is requested, or not getting what is requested completely done. I believe that taking the initiative is taking on tasks that one is not necessarily assigned to do, so then it would require one to work overtime, at times.

[Jeanne Boyarsky]

Originally posted by Mark Herschberg:
In the US, the 40 hour week, outside of government work and government related industries is pretty much a thing of the past for salaries employees.

There's quite a range in the examples here. Doctors and people working long hours at consulting firms know this part of the job. Investment banks are pretty honest about the hours too. However in many IT jobs, you are told it is a 40 hour work week when you interview. If it is a 60 hour work week, call it that up front.

Originally posted by Lulu Carr:
How many of us believes that willingness to work late & hard is based on if we consider what we do "just a job" vs. "a career"?

Which means I have the pleasure of being promoted to a job that requires even more working late. Hmm. Is this a good thing? Note that working late and working hard aren't the same thing. I think a "career" is based on skills, not just the length of time working.

Originally posted by Lulu Carr:
What I've noticed in my company is that people don't get noticed for just working 40 hrs a week, because that means that they are only doing what is requested, or not getting what is requested completely done.

This always bugs me - the expectation that you should "want" to work extra. As to not getting what is requested done - maybe that means too much is being requested. It quickly becomes a self-fulling prophecy. You show you are ok with working X hours and then become expected to do so for the rest of your time at that company. Of course for unexpected things, overtime is sometimes needed. But to actively plan for it over a long period of time and then imply that you are doing a poor job seems unfair.

I am one of the lucky ones who has a 40 hour work week. I will work overtime at times. However if it is more than 30 minutes it is for production support, production releases or a very specific reason. That said, I work quickly and focused for those 40 hours. I don't check JavaRanch, the paper, my e-mail, etc. I also do personal career development (learning Java 5 etc) on my own time. Some of the places described earlier in this thread (where face time matters), people may be there longer but they aren't always being more productive.

I agree with Mark that 44 hours is certainly sustainable. However if we had a 44 hour work week, they would be asking for 50. (or 60+) You do hit a point where the overtime is not sustainable though. We had a subway strike in New York City about 14 months ago. I worked 7 hours each day of the strike and spent about 3 hours each way commuting. So this accounts for 13 hours of the day. Adding sleep and human needs (eating, showering, etc), I had just enough to check my e-mail and that was it. This wasn't sustainable for even a few days. By the second day, I was present, but felt like a zombie. Companies often forget that tired people make more mistakes that require rework and then the overtime wasn't so useful.

My point being that I think we put too much value in # hours over results as an industry.

[Jeanne Boyarsky]

What I've noticed in my company is that people don't get noticed for just working 40 hrs a week,

This brings up another interesting point. Did we as an industry get into a situation where we are competing with each other to see who works the most overtime?

The reason the perception that people don't get noticed for "just" working the amount of time they are supposed to is that other people are voluntarily working more hours as their base.

[Mark Herschberg]

There are a number of good books on US work culture. Of course, what we have here in *nothing* compared to the work culture in Japan!

Again, 40 is an aribtrary number stemming from historical influences. There is nothing inherently wrong iwth a 60 hour work week. Personally I think people burn out doing that for weeks on end, but if you want to run a company that has a 60 hours work week, that's your choice. If people don't like it, they won't work there. if you pay them enough or reward them enough in other ways, they'll put up with it. Some people actually like it.

I *always* ask a compnay what the typical work week i slike and I *always* tell candidates what I expect. As I work for startup companies, I expect about 50 some hours a week because we need to work long hours to win, but I've found going much above 50 the marginal returns aren't worth it.

I've yet to find an employer who wasn't honest with me about the hours. (Or maybe I simply only took jobs from employers who were honest about it.)

--Mark

[Jeanne Boyarsky]

Mark,

There is nothing inherently wrong iwth a 60 hour work week.

I have no problem with a place having a 60 hour work week if they tell people about it up front. I wouldn't work there, but it's their choice to function that way.

I've yet to find an employer who wasn't honest with me about the hours. (Or maybe I simply only took jobs from employers who were honest about it.)

That's good! When I interviewed, all the employers except one stated it was a 40 hour week. Now I can't prove they weren't telling the truth. However, one of them (and not the one that admitted to a longer week) was an investment back that is known for long hours. And the one that admitted to more hours stated that while it is a 40 hour week, you are expected to work more. I think that some of this difference in expectation might be coming from the "you should want to work more hours" mindset. They can honestly say it is a 40 hour week even if nobody actually works that number of hours because they can say overtime isn't required.

Now I was interviewing while in college, so maybe they felt no reason to be honest as I was in college. One of my friends worked for one of the big X accounting firms out of college (not as a techie.) It is common knowledge that the hours are long. However, they wouldn't admit to what they were at the interview. I guess I'm just a cynic when it comes to this stuff...

[Tim Holloway]

The problem is, it's been pretty well established that in the IT world, hours worked have virtually no relationship to productivity whatsoever. It's what I like to call the "hamburger grinder" fallacy. Back when I was starting out, it was called trying to speed up the process of a woman having a baby in 9 months by employing 9 women, but the reality is worse than that, as not only does productivity vary by orders of magnitude between people, but the people themselves vary wildly. In my own case, the very way I think about things is seasonally cyclical.

I've worked in offices where "the last one who leaves, wins!", but I never noticed that that last person was, in fact, getting any more done. One of the Murphy laws concerns how projects fill to fit available time, and I've seen people doing nothing more productive than rearranging icons on the desktop in order to be "busy". Or, worse yet, mucking up other people's machines. Guess who had to divert time to repairing them?

Software production is inherently paradoxical - part deterministic and mechanical and part creative. The mechanical parts are identified and automated, so they generally take relatively little time. The creative parts, however cannot be forced. Even hamburger grinders can only be sped up so much. After that, what comes out isn't hamburger, any more, it's cooked meat.

[Mark Herschberg]

Originally posted by Tim Holloway:
The problem is, it's been pretty well established that in the IT world, hours worked have virtually no relationship to productivity whatsoever.

Bullshit! (Nothing personal Tim, and I only use language this strong because you're a fellow moderator.)

Well established by whom? Unless you've got some authorities proving this, I'm call such a claim anecdotal complaining.

You give me 2 software developers who are about the same, and if mines works 40 hours a week and yours works 20 hours, mine will be almost twice as fast. Sure it's not always linear (Mythical-Man Month), but that sure sounds like a causal relationship to me. Now at what point an individual gets to the point of no return may vary, but there is a time/productivity relationship.

Yes, I would grant you that there are a concerning number of places that think asking people to work long hours means higher productivity, but you're claim seems excessive to the point of being flammatory. So again, what authority who has surveyed the industry makes this claim?

--Mark

[Deepak Bala]

The problem is, it's been pretty well established that in the IT world, hours worked have virtually no relationship to productivity whatsoever.

In the general sense that is true. If I work 45 hours and put some real effort into it and the other person works 55 hours but rearranges icons, browses magazines, takes tea/coffee breaks for 10 hours, it would make him/her less productive.

I also agree that if your developer works for 20 hours and mine works for 40 hours, my developer is going to be better, as long as he/she does not waste half the time by doing things other than work.

A 44 hour week is quite sustainable. At crunch time it can be raised to 60 hours for a period of 3 weeks max. Anything beyound that is probably not going to keep the engine running because people can become very unhappy because of long working hours. When you cant control the amount of time needed to be spent on work most of the time, it can suck.

[Rohit Nath]

I wont claim any conclusions but here are my observations..

I started my career with a startup company.
We were a team of 7 people and typically I would put in around 9-10 hours per day regularly as it was the need of the company. Over a period of time I started analyzing my time and task management.

I tweaked around with my work style:-
1. Spend time on clear understanding of task.
2. Cut down on "time wasters" like email/phone/surfing/friends during work hours!
3. Plan the day for the task. Set my own deadlines.
4. Try to complete task before each deadline.
5. In the saved time do extra things that help me provide that "extra".
6. Make the *Task my first priority* and everything else after work time.
All this added up to efficient usage of my time.

With above changes I was able to do the *same* work in 8 hours instead of 9-10 hours.

So by focussing only on task and not going to the pantry and having a coffee and a chat with my friends or having a walk outside... I was able to save a couple of hours and use them to invest time in myself. Investing time in myself daily after the work helped me beat the work pressure and also improve my own health and mind and keep fresh throughout the week and walk everyday to the office with new enthusiasm.So I never had to wait for the Friday to come! Also it boosted my overall productivity and quality of work.

Does that mean the previous me (working 10 hours per day) is doing a better job than the me (working 8 hours per day) ?

Also one more incedent happened in our company that I really appreciate..
Once our manager asked us how man hours of work we put in per week?
Most of them proudly said 50 hrs, 60 hours per week..and some even more while some only softly said 42 hours per week..
After all this one of the employees asked why did he ask all this?
He replied that all of you working more that 45 hours per week are not being efficient! (*******Given the fact that all in that group were doing similar work.)
There IS a reason for this. While some would go out for a breakfast in morning during office hours others would be busy focussing on their work and task on hand. While others would go out daily for evening tea outside the others would be still busy doing their task.
So in effect you can see that having a shorter work day means sacrificing on other things! (This opinion should not be generalized but is a good case study in itself.)

These are some tried and tested techniques which worked for me! Eveyone is free to have their own beliefs and independant thought process.
I get to the work to do justice to the work and not to impress someone. Afterall we are not running a popularity contest!

There are genuine cases when long hours are *Necessary* but so are cases when they are *made* long.
[ February 20, 2007: Message edited by: Rohit Nath ]

[Srikanth Raghavan]

I tweaked around with my work style:-
1. Spend time on clear understanding of task.
2. Cut down on "time wasters" like email/phone/surfing/friends during work hours!
3. Plan the day for the task. Set my own deadlines.
4. Try to complete task before each deadline.
5. In the saved time do extra things that help me provide that "extra".
6. Make the *Task my first priority* and everything else after work time.
All this added up to efficient usage of my time.

I do the exact same thing and now I understood why I don't need more than 8 hours. Today I am just sitting without any work. It sucks. And I don't read or do any useful thing when I have some free time. I always need work and only if there's a need for me to learn XML, I will learn it (just an example). That's the way my brain works.

I like finishing tasks that gets me some reward, some driving force --a goal. But working 12 - 14 hours really sucks if we have to do it daily.

-- Srikanth

[Jason Cox]

I'm of the somewhat odd mind that if I can do my job in 40 hours then I am going home at the end of the day. If someone else needs 50 hours to do their job, I don't feel obligated to hang around the extra 10 hours a week just to make them feel better. If there is something I can do to help keep the project on track, even if it means I go over 40 hours, then I generally will do it.

In terms of development I am one of those people that tends to be at least twice as productive as anyone else on the team. I have completed numerous tasks, helped others with their issues, and mentored junior developers all in a 40 hour week. I generally don't have to go over and if I knuckle down and get the work done I can do a lot in a surprisingly short amount of time. However, my productivity drops dramatically as I get tired. At 14 hours straight I am pretty much writing 80% bugs. That's with a good amount of rest the night before. I then have to spend the next two or three hours of the following work day undoing the bugs I wrote the night before. I simply do not believe more hours is more productive.

While the 40 hour work week may seem arbitrary, there seems to be something to it. I've seen shops that routinely work their people slavishly and I see very little in terms of creativity or innovation. Code quality tends to be bad and bug fixes are endless. There are days I'm surprised those places actually write any functional code at all. On the flip-side, I have worked at a client site where the regular employees routinely worked between 30 to 35 hours per week and it was also frustrating. There was no sense of urgency and having employees effectively miss a full work day worth of hours had about the same impact as having employees work too many hours.

If someone likes working long hours I think that is fine. I have met very few people who are actually effective when they are working every waking hour.

From a career standpoint, I don't think long hours means better opportunities. I know senior people who routinely work 60+ hours that have peers that work a standard work week. To be honest, I kind of admire the guy who can do the same job in less time. I will say that it is damn hard to climb the corporate ladder if you're never willing to put in the time. My general philosophy is that most things can be done in a 40 hour work week and I'll be right there with you walking out the door at quitting time. However, if there is work to be done or a problem crops up, I will work to get the problem solved. Some things, no matter how tired I am or how long the day has gone, simply cannot wait until tomorrow. If I were evaluating one of my subordinates I'm not going to look at how many hours they worked per se but I am going to want to know if they've used their time efficiently and if they've been willing to take some extra time when it was needed. I am not so impressed by people who will leave at their usual time while we're in the midst of a crisis.
[ February 20, 2007: Message edited by: Jason Cox ]

[Jeanne Boyarsky]

"The problem is, it's been pretty well established that in the IT world, hours worked have virtually no relationship to productivity whatsoever."
There's a correlation relationship, even if it hasn't been proven. It's just a complicated one. It certainly isn't linear. I think we all agree that someone working 40 hours is more productive than the same person working 20. However is someone working 62 hours more productive than someone working 60. [edited this paragraph because I re-read Mark's post and realize he wasn't asserting no relationship]

"I was able to save a couple of hours and use them to invest time in myself."
Rohit makes an important point. Investing time in oneself (whether at work or at home) is critical to sustaining an efficient pace. Otherwise you solve problems in inefficient ways. For example, you can remove comments from a String representing an HTML file in one line of code and a few minutes using regular expressions. If you have been so busy all the time and didn't know Java 1.4 had regular expressions or have time to learn them, you could easily waste 30 minutes (or more) writing the loop. And it would be more complex logic, making it less likely to work on the first shot - wasting even more time.

Jason: Amen! I couldn't agree with your post more. I want to frame it . I still think the 40 number is arbitrary. Each person has a limit though. It's important to know where it is as an individual!
[ February 20, 2007: Message edited by: Jeanne Boyarsky ]

[Tim Holloway]

I'm not just spouting anecdotes. There are people who get paid to analyze productivity and to distinguish real productivity from apparent productivity. One of them is Ed Yourdon. Without digging back through my library, I believe it was Yourdon who made the observation that "when programmers are productive, they are very very productive, but when they are bad, they are horrible." I'm sure I've mangled the quote, but you get the idea.

Actually, I've often wondered if other professions are as intensely analyzed as IT. From the beginning days of my career, I've read studies of the programmer psyche - mostly in COMPUTERWORLD, but IIRC, also in various ACM and IEEE pubs as well. I've seen stats from projects I worked on myself. Latterly, I've read stuff in the Wall Street Journal, but admittedly that's not a place where hard numbers show up often.

Almost everyone is peddling an agenda, so I take things with a grain of salt. On the other hand, when people like Fred Brookes, Ed Yourdon, James Martin, et al. consistently come up with similar observations, it does make a strong case. There's a lot at stake. There have been some spectacularly massive - and expensive - software project failures in the last half-century. Things like the failed project to overhaul the California Driver's License system. Like the week-long shutdown of the Hershey company when their SAP cutover went awry not all that long ago. People are willing to pay good money to find out what happened, why, and how to ensure it doesn't happen again. And people who get paid that money don't keep getting paid unless they can make a pretty good showing of why.

If I was absolutely forced to, I suppose I could go back and rummage through a lot of that material and present it formally with hard numbers (where available).

Before I do, however, who among us is willing to state that their functional code output is truly a linear function of time? That's the whole basis of the more hours = more productivity assertion, isn't it? There are cases where this is true. For myself, I have a toolset that I can predict with certainty that given no unusual requirements, I can - given "n" database tables, produce web pages (everything from ORM infrastructure to CSS) to do CRUD functions on them at the rate of about 4xn hours. But that's not something I'm often called to do. What is normally required of me is to research and apply new technologies. That's not a predictable function. It's the kind of thing where the ultimate answer usually pops out spontaneously in the shower or at 4 in the morning.

And, of course, my best days are the ones where I produce negative LOC. When I take something sloppy and hacked out and find evil ways to collapse it into something smaller, faster, more reliable and more flexible. That, in fact, is one of the things that makes the whole job worthwhile.

Ergo, when I maintain that the "hamburger grinder" model is fallacious, I base it on 3 sets of data:

1. My own personal experiences and observations over the last several decades as a professional.

2. The collaborating reports of peers - some of whom have made similar observations in various forums here on the JavaRanch.

3. Published case studies and similar documentation by respected figures in the industry.

To play Devil's Advocate, I will admit that I can probably collaborate that many, if not most of the successful IT projects involved people working absolutely insane hours. I probably don't have to go any farther than Tracy Kidder's "Soul of a New Machine". However, to counter that, I'll offer the statement (I think it was Fred Brookes) that the challenge of a good IT manager is less in motivating the staff on a good project than in assuring that nothing de-motivates them. Putting in long hours on a project, I understand. Something worth doing well is worth obsessing over. I do it my own lazy self.

However, the question that started this thread related to the idea of working staff overtime as a routine process. A coerced process. That I can't justify, if for no other reason than the one I gave earlier. You can't handle special cases if you're already at 100% capacity.

[Mark Herschberg]

Originally posted by Tim Holloway:
I'm not just spouting anecdotes. There are people who get paid to analyze productivity and to distinguish real productivity from apparent productivity. One of them is Ed Yourdon. Without digging back through my library, I believe it was Yourdon who made the observation that "when programmers are productive, they are very very productive, but when they are bad, they are horrible." I'm sure I've mangled the quote, but you get the idea.

...

If I was absolutely forced to, I suppose I could go back and rummage through a lot of that material and present it formally with hard numbers (where available).

Before I do, however, who among us is willing to state that their functional code output is truly a linear function of time?

Yeah, with respect, it still sounds like BS to me. I'm familiar with Yourdon's writings (he and I have been trying to schedule a time to meet actually). But his observation is irrelevant to your claim. You claimed The problem is, it's been pretty well established that in the IT world, hours worked have virtually no relationship to productivity whatsoever. I'm asking you to substaniate your claim with real data from experts, not "I seem to recall someone once said."

To your question of

Originally posted by Tim Holloway:
Before I do, however, who among us is willing to state that their functional code output is truly a linear function of time? That's the whole basis of the more hours = more productivity assertion, isn't it?

This is also irrelevant. The only person who has made a claim here is you, and I'm asking that you back it up and not just spout your personal opinions and call them fact. (I you look at my past postings you'll see I say this to everyone who makes such claims in this forum.)

Originally posted by Tim Holloway:
1. My own personal experiences and observations over the last several decades as a professional.

2. The collaborating reports of peers - some of whom have made similar observations in various forums here on the JavaRanch.

3. Published case studies and similar documentation by respected figures in the industry.

Yes, I've seen all this anecdotal data. I've read Youdon, Lister, DeMarco, Kidder, Cosby, etc. I've talked to friends. I've interviewed hundreds of people (it might even be in the thousands by now). You know what I've concluded? There are many of incompetent people.

Yes, lots of people work inefficently. I'm not talking about spending 30 minutes Monday morning talking about your weekend with co-workers. I'm talking about people who stop working every 15 minutes to answer personal email. People who spend 30-60 minutes every day gossiping. People who show up unprepared for meetings and then have to do it again.

It's not just workers, many managers think more hours always equals more productivity. I don't. (To be clear I never claimed the above, but to those who claimed that there is norelationship, as Tim did, I provided the counter-example, disproving his overly general claim.) But many managers do and they misuse their team's time.

But that there are many bad managers and employees also doesn't prove your claim. Here's what I do think....

Baseline: A single employee can work a 40 hour work week

1) That can employee *can* increase productivity linearly up to anywhere of 50-75% but linearly increasing his/her hours that week.

2) In general (e.g. for the average employee), if an employee works more than about 50 hours a week for more than about 3 weeks, his efficency will go down.

3) Many people could be more efficent in their work day.

4) Eliminating *all* non-work related time (e.g. casual chats in the hallway) will make the average developer less efficent, as the day becomes overly focused on work thoughts only.

5) There exists exceptions (people) to all the above rules; e.g. there is no doubt someone out there who can work productively at the same efficency for 80 hours per week for months on end. Likewise there is someone who starts to become effiency after 30 hours of work per week.

6) These exceptions don't prove a universal rule. Nor does the average case.


Tim, my point is, I think it's not uncommon to see what you describe. I disagree that this is some universal law. I know a lot of people who do work effiently and who can work more than 40 hours per week. I am open to having my mind changed but your arguments thus far aren't convincing. Please provide the data you have.


--Mark

[Mark Herschberg]

Originally posted by Jeanne Boyarsky:
I still think the 40 number is arbitrary. Each person has a limit though. It's important to know where it is as an individual!

While I agree that everyone is different, the 40 hour work week is not arbitrary. If you look at the history of the labor movement in the US you'll see that there were a number of historical forces that resulted in the 40 hour work week.

I'm not claiming that we are evolutionarily designed for a 40 hour work week, but I am saying it wasn't simply someone picking a number out of a hat.

--Mark

[Jeanne Boyarsky]

Originally posted by Mark Herschberg:
While I agree that everyone is different, the 40 hour work week is not arbitrary.

Oops. Let me further qualify my sentence:

I still think the 40 number is arbitrary as the peak amount of time for employee productivity. There's going to be some point at which working more hours makes people less productive. I think this point varies by individual, so it's not some fixed number anyway. This number is not likely to be exactly 40.

[Mark Herschberg]

Originally posted by Jeanne Boyarsky:

Oops. Let me further qualify my sentence:

I still think the 40 number is arbitrary as the peak amount of time for employee productivity. There's going to be some point at which working more hours makes people less productive. I think this point varies by individual, so it's not some fixed number anyway. This number is not likely to be exactly 40.

That, I agree with. This is why when doing startups where project time can make to break the company, I go higher. My empirical data has generally been around 50 per week--but as I usually build my own teams *and* I work only in major metropolitan areas, there is a huge possibility of sample bias.

--Mark