[btilly]

I find it sad that this article leaves out the (admittedly unpublished) research on the 10x phenomena, which is described in Peopleware. They consistently found 10x differences between experienced programmers on the same task, with the same tools.

I find that particular research important for three reasons.

1. The 10 fold difference was consistently demonstrated many times.

2. Their research was conducted with professional programmers, in their workplace.

3. They were able to estimate the impact of a number of different factors, and they found that the single biggest impacts were workplace factors. (Ambient noise, dedicated office space, availability of white boards, etc.)

Admittedly they are unable to tease out why the connection exists between the workplace and productivity. Do productive programmers choose good workplaces? Or do good workplaces make programmers more productive? Probably a bit of both.

BUT if you're an employer, go read that book. Then consider your office plan. It is a lot easier to pay for a good workplace for an effective small team than to pay for a less effective larger team. Office space is probably not where you want to scrimp.

[gruseom]

Much as I like Peopleware, it isn't credible as research. The studies were never published, and the authors made stuff up, as one of them told me himself (I'll see if I can dig up the link where I wrote about it.)

Edit: https://news.ycombinator.com/item?id=1995716. It sounds mean and accusatory to say on an internet message board that they made stuff up. But actually, our brief exchange was charming. He said it with a twinkle in his eye.

[jacobian]

How, exactly, is one supposed to cite unpublished research?

[btilly]

The authors of the book were the ones who did the research for their consultancy. But you could easily cite Peopleware itself for the research.

[chii]

tbh, qualitative research like this has no predictive power. It's just a (few) case studies. The variables that are in these case studies cannot be controlled for, nor tested and/or predicted. All it gives is some example anecdotes. Unfortunately, a lot of social science research end up like this, because of the complexity of the problem. Thus, i wouldn't cite peopleware as research, but as anecdotal examples, or case studies.