what are difficulties that prevent us to measure and compare between pair and solo programming ? is it difficult to compare their productivity? if yes , why ?
A:
Perhaps, but it is certainly possible and there are many academic papers on the subject.
Justin Ethier
2010-10-17 03:31:16
A:
I think that the first difficulty would be establishing your measure of productivity. Lines of code is a bad one, for instance, but you could use number of issues fixed, or perhaps a metric which is based on how fast tasks are completed relative to the time that was estimated for them to be complete.
I've heard somewhere that teams measuring pair programming found roughly a 40% drop in productivity (no idea how they measured it) for the first few months, after which the numbers began a steady increase.
RyanHennig
2010-10-17 03:32:27
A:
In the past I have found this issue to be more of a trust issue than anything else. You can not count lines of code to determine who has produced the most, but if working with someone else make your delivery time short and reduce stress I would said it more productive. Also watch out for perfectionism as it will make hard for you to see how others perform.
redhatlab
2010-10-17 03:35:51