A thought has been bouncing around my mind for quite a number of years now regarding the accumulation of knowledge and media. Basically there are two separate but related thoughts here. The first relates to keeping on top of new science that's published relevant to my work, and the second relates to trying to keep on top of culturally-relevant media such as video games, comics, music, movies, books, etc.
As a neuroscientist I spend a decent amount of time just trying to stay on top of new research. A few weeks ago I wrote a post about how I use a mix of PubMed, RSS readers, and email to keep on top of the very specific topics and researchers related to my work. Although these tricks save me time, I spend probably an hour a week just sorting through new publications. That doesn't include the time it takes to skim and read them, which is (or... should be) much more.
According to this paper by Karl Friston published last year in Science, there are roughly 80,000 neuroimaging papers published every year in fMRI, PET, SPECT, EEG, and MEG. Let's assume this is a decent metric for the number of actual cognitive and systems neuroscience papers published every year relevant to my interests, though this is only a very rough measure because I tend to ignore many fMRI studies and over-emphasize single-unit, LFP, EEG, MEG, and intracranial studies.
Now, going back to mass media...
You know what? Initially I was doing this off-hand, but I think I'm going to cut this post short and break it into a few parts.
I've actually gone and run some simulations now, and I'm going to post the results here later as I put finishing touches on the model, but it looks like if we make some modest assumptions that 0.1% of media are worthwhile, culturally-relevant, and interesting enough to your own tastes to consume, it would take 72.3 years to get through those media. Here I make the assumption that one could get through everything (every video game, book, movie, etc.) in an average of 10 hours each.
Of course, there are a lot of assumptions here, but I'm going to keep fiddling with the model, start using real data, and see what happens.