A very simple question:
Does anybody know of any good references that scientifically prove the effects that listening to various types of music while coding might have on a developer?
Workplaces typically provide some sort of reference library to help their developers be more productive. Would it also be conducive to productivity if some began offering their developers music service subscriptions?
I think it depends on what type of developing you're doing. I find that if I'm working on a complex piece of code that requires a great deal of concentration, then music tends to be a distraction. However, if I'm writing more mundane code (like CRUD stored procedures, etc.) then music doesn't seem to have a negative impact.
I think a lot of parallels can be drawn between this and listening to music while studying. Here's an article for reference. The crux of that article in my opinion is this statement:
Research has shown that, with practice, people can improve how often and when to shift focus to other tasks most efficiently, and they can sharpen their ability to visually scan between windows open on a computer screen. But decades of experiments on adults have proved that performance suffers when people try to multi-task.
So my personal opinion is that if you want to be at your very best when writing a complex piece of code, then it's best to remove as many distractions as possible and focus solely on the task at hand. I guess another word for this is "tunnel-vision".
Others may disagree, which is fine and encouraged :).
In all seriousness (re: my joke in the comments), Google results are filled with papers and discussions on the topic, but I doubt there's any "scientific proof" of anything. Maybe start by looking at it from a different angle? Try finding information on surgeons using background music in operating rooms. I imagine more real studies would be behind something like that, but the concept is the same.
The clamor and din does not impair playing tetris:
http://www.springerlink.com/content/t703451w2753u344/
My take away from that article is that if software developers work in loud environments, they want us to play tetris.
However, if you are writing code, it clear quiet is better:
http://ieeexplore.ieee.org/xpl/freeabs_all.jsp?tp=&arnumber=362626&isnumber=8308
Well I gave it a whirl and tried to find some references because I know the study of music on work and exercise performance. Unfortunately I didn't find much. You'll probably want to look into google scholar or even better a research database if your company/school provides one.
Here's a paper on the relation between programming and being a musician:
http://www.springerlink.com/content/1r05quulnv27num2/
Here's a paper on the distraction of music on cognitive tasks which argues the point against the mental disposition of the worker i.e. whats the effect against a type a and type b personality.
http://www3.interscience.wiley.com/journal/11970/abstract
Personally I believe this data would be better if done through a self experimentation method. It's not scientifically exact but is more honed toward your unique personality and vatious environmental variables that affect your day-to-day. Keep a log of day-to-day variables like music used, sleep, exercise, diet and so on until you have a decent ammount of control data. From there you can start by changin one variable at a time to see the effect.
Probably not what your looking for but I didn't know this wasn't what you weren't looking for...
http://webworkerdaily.com/2010/07/12/can-listening-to-music-boost-your-productivity/ seems to indicate music could have a positive effect.
I don't know if you can really argue for premium music subscriptions, however. There is plenty of free music available on the web.
McConnell's Software Project Survival Guide has two interesting surveys, one of which is related. The first is "effectiveness of a good developer vs a mediocre one", which is about 10:1.
The second - the much more relevant - is the "effectiveness of a developer in a quiet environment vs a noisy/distracting one". 10:1 again. I'd suspect the effect of listening to music the developer is comfortable with has a similar effect as a cone of silence; that is, a developer listening to music in a not-otherwise-silent room may have a productivity boost of up to 10x.
Even a quiet office has phones, faxes, talking, coffee machines, lifts dinging, swinging doors etc and few programmers rate a private office :-)
Music allows you to tune out these random outside factors: you don't listen to it as such