Hours per Week?

[Corey McGlone]

Perhaps this is the wrong place to post this, but this was the closest forum I could think of.
I've been noticing a handful of references (authors, websites, etc.) that remark about the balance between productivity and work hours. For example, not knowing much about XP, I believe most XPers believe that you should never work more than 40 hours in a week.
I also heard a recent story (although this was just by word of mouth - I have no idea of the accuracy of the story) that a company was requiring employess to take a 30 minute nap in the afternoon and then saw a great jump in productivity.
Have there been studies done on this? I would imagine that there have been. Does anyone know where I could find more information about this? There seem to be many companies that still believe that the longer you sit at your desk, the more you'll get done. I don't necessarily believe this is true, but I'm interested in learning more.
Any help or comments are greatly appreciated.
Corey

[Reid M. Pinchback]

I'm not familiar with the nap study. Back a half dozen years ago when I went through a slug of project management training, I heard similar things about studies of how productivity dropped dramatically after 40-50 hours per week, and how turnover rates increased if excessive working hours continued for very long.
I don't know if any study has made a distinction between total working hours (e.g. at office plus pounding on the laptop on the train or at home watching tv, or being tied to a pager all the time) versus in-office hours. I've seen articles that mentioned that the growth in unofficial hours is probably more evident in North America, and would be a source of increased stress because of the reduced downtime.

[Ilja Preuss]

I can't provide a study, but browsing http://c2.com/cgi/wiki?FortyHourWeek might be interesting.

[Frank Carver]

I'm not sure about the nap example. Maybe it's just because I'm generally tired these days (too much work and two young children ) but if I sleep anytime except at night I wake up really disoriented and it takes me hours to get straight again.
I have seen great benefits from keeping working hours under control, though. One thing that I think is especially important, and only implicitly adressed in XP, is that when you come to the end of a days work, you should stop at a point where things are finished and tidy. Any work still "on your mind" greatly reduces the benefit of the time away. See my Golden Rules of Stress Free Programming for a more detailed discussion.

[Matthew Phillips]

Great article, Frank. Another reason to test early is the opportunity to see a dumb machine do what you told it occurs earlier and more frequently.

[Reid M. Pinchback]

Done repeatedly while zoning in front of the monitor, it is probably a good stress-reliever too. You can stare at the little green bar saying "the gods may be crazy, but at least this works..."

[Ilja Preuss]

A colleague worked long hours yesterday to fix a critical bug - the software should get released today. Some minutes ago she realized that she was so tired at the end of the day that she confused "commit to repository" with "replace with latest from repository"...

[Mark Herschberg]

Doing my best not to hijack this thread...
I don't know about any studies for long work hours. http://www.ilocis.org/sample1.html looks interesting. http://www.businessweek.com/smallbiz/0006/wf000626.htm correlates it with fewer hours of sleep, and that has been shown to reduce productivity.
Now for the hijacking... Peopleware actually sites studies, unlike most books out there. The include an IBM study which showed that workers need space and privacy to be productive. Those open offices which have become so popular because they improve communication, don't. They only keep overhead costs down, at the cost of productivity.
--Mark

[Reid M. Pinchback]

I was just talking with somebody at work about this today, the open space issue. Two-person offices are fine, but I've never found it easy to concentrate in cubicles, even if the cubicles themselves were spacious and had decently high walls. Something about the space just shut my brain down. A cubicled co-worker of mine is finding the same thing to be true for her, she needs to work at home on any problem that requires real thought.

[Frank Carver]

I find the topic of cubicles interesting. It's one of the few areas of development where I can't emphasize with Dilbert, as I've never worked anwhere which used cubicles.
I've worked in open plan offices with shoulder-high privacy screens. I've worked in offices with a load of flat-topped desks where you stare at the mess of wires and fluff begind someone else's PC. I've worked in labs where there is so much equipment on the benches there's hardly any place for a keyboard, let alone any paperwork. Currently I work at a desk with my back to an open corridor, so anyone passing by can see my screen. I've got used to all of them.
To me the biggest difference is simply whether or not you are co-located with people working on the same project. Overspill from other peoples work can be useful and even sometimes synergistic if you are all working on the same thing. Overspill from unrelated projects can be immensely irritating.
Oh yes. And teach people to turn off their *%#! mobile phones. If I hear another stupid ringtone from a phone someone has left behind at their desk I think I'll turn into the Incredible Hulk

[Ilja Preuss]

Originally posted by Reid M. Pinchback:
I was just talking with somebody at work about this today, the open space issue. Two-person offices are fine, but I've never found it easy to concentrate in cubicles, even if the cubicles themselves were spacious and had decently high walls. Something about the space just shut my brain down. A cubicled co-worker of mine is finding the same thing to be true for her, she needs to work at home on any problem that requires real thought.

Yeah, with cubicles you get the worst from two worlds: They hinder communication without shielding noisiness effectively.
What many agile processes advocate is *not* putting *anyone* in the same room, but to provide one room *per team* (you could call it a "war-room approach"). It happens that you are typically not as distracted by noise of the same type you are producing yourself (I think this is mentioned in Peopleware, too). It is also my experience that pair programming makes me less vulnerable to distractions.
Also, it generally isn't recommended to totally forsake privacy - see http://c2.com/cgi/wiki?CaveAndCommons and http://www.xprogramming.com/xpmag/c3space.htm
[ July 04, 2002: Message edited by: Ilja Preuss ]

[Reid M. Pinchback]

Originally posted by Frank Carver:
Oh yes. And teach people to turn off their *%#! mobile phones. If I hear another stupid ringtone from a phone someone has left behind at their desk I think I'll turn into the Incredible Hulk

I hear you! We've got somebody that does that all the time; she has a phone that will ring once a minute to notify her if she has voicemail. She leaves it at her desk and then disappears for an hour. Not sure which is worse; that, or the perfume stench that hits you from 40 feet away.

[Ilja Preuss]

Originally posted by Reid M. Pinchback:

I hear you! We've got somebody that does that all the time; she has a phone that will ring once a minute to notify her if she has voicemail. She leaves it at her desk and then disappears for an hour.

What did she say when you talked to her about the problem?

[Reid M. Pinchback]

Originally posted by Ilja Preuss:

What did she say when you talked to her about the problem?

She bares her fangs when anybody dares talk to her about anything, even trivial stuff like this. We've all learned the hard way to work around her. The solution: close the office door and put on headphones. Thank god I don't work in a cubicle.

[Ilja Preuss]

put on headphones.

There is an interesting section in "Peopleware" about music in software development. IIRC there was an experiment where they gave a bunch of people programming assignments, half of them allowed to hear music of their choice, the other half working in a quiet environment.
The outcome was very interesting: There seemed to be no significant difference in time needed or bug count between the two groups. However, there was a way to shortcut the used algorithm (some of the steps were unmaking each other) - and the developers in the quiet environment were considerably better in spotting this shortcut.
It seems as if the irrational, creative half of the brain gets occupied by the music, whereas the rational, logical part still is able to concentrate on the programming part.
Interesting, isn't it?
[ July 09, 2002: Message edited by: Ilja Preuss ]

[Reid M. Pinchback]

I wonder what the effect of white noise would be? Lately I've gotten into using CDs of rain/weather sounds. Less distracting, particularly when compared to music with vocal tracks.

[Frank Carver]

Reid wrote: Lately I've gotten into using CDs of rain/weather sounds. Less distracting, particularly when compared to music with vocal tracks.
Neat idea. I've got some of those CDs around here somewhere which I used to use as background sounds for video work. I'll have to give it a try. I've tried music in the past, but it frags my concentration and I end up with a headache.
As an aside I did try an experiment once. One night it was unseasonably hot here, and neither I nor my wife could get to sleep. So I fished out a "thunderstorm builds in a rainforest" CD and played it. In a wierd psychological way it seemed to cool the room down more than the fan had!

[Ilja Preuss]

Originally posted by Frank Carver:
As an aside I did try an experiment once. One night it was unseasonably hot here, and neither I nor my wife could get to sleep. So I fished out a "thunderstorm builds in a rainforest" CD and played it. In a wierd psychological way it seemed to cool the room down more than the fan had!