I'd argue that this definition of "productive" is insufficient. Conveniently, the link to the study is broken (and to "vouchercloud.com", not BLS, the stated so

I'd argue that this definition of "productive" is insufficient. Conveniently, the link to the study is broken (and to "vouchercloud.com", not BLS, the stated source of the study), so it's hard to check the numbers. My point is, few people can walk up to a keyboard and program "productively" for three hours -- and then do the same thing the next day again indefinitely. Hidden in those productive hours is a lot of informal coordination, necessary for those three hours to be actually productive. By no means is it given that 8 hours is the optimal workday, nor that the way we generally structure work is the most efficient, but luckily, our field is especially receptive to radical experiments, so someone will surely try this any day now. (/s)