Two unfortunate bits of new today:
Several years ago, I heard that the authenticity of Jackson Pollack’s paintings could be determined empirically through a fractal dimension. I really wanted this to be true. Sadly, this paper makes the original and subsequent work on this topic look sloppy at best. Science News has an article on it as well, most specifically that the test can be fooled, and comes up negative on Pollacks paintings that analyzed later. Oh well.
Now art and empiricism do go hand in hand in book authorship determination, which can’t irrefutably distinguish Bacon from Shakespeare, but could probably differentiate either from me. Read this pdf for an over view and this one for a technical comparison between methods.
In other mildly disappointing news, Trent Reznor has forayed into alternative methods of selling imaginary property, and failed miserably. The story goes: 27/36 tracks are free at low bitrate. All tracks at high bitrate cost $5US. That’s like a 4-CD set for $5. Too bad Piratebay is a-buzz with imaginary losses of sales. The reader may have previously heard of NIN releasing music Open-Source-stiley. This was a success. Good job Trent, clever adaptation to new tech. Why did this new experiment fail? The Reg blames crappy webhosting. This is part true. If you assume the fee is for the providing service, like CD sales, it only takes 15 minutes to download the file from Pirate bay, the ease of download from the source should be proportional to the cost, which it aint. They estimate $2million a week loss, but this presumes that every downloader is a lost customer, which is stupid. Sorry Trent, right bus-line, wrong bus-stop. The economic ice-caps are melting. The recording industry might not have gills, but they’re growing flippers.
BTW, I hate to nit-pick, but why did the arXiv start accepting papers in M$word format? It looks so unprofessional. If you can’t use LaTeX or some other real typesetter, you shouldn’t be working in predictive science.