Slashdot Mirror
The RIAA Hit List - A Pattern Emerges?
Desus writes "Slyck News seems to have found a pattern in just what files the RIAA is searching on to find offenders. It seems the RIAA is targeting a wide reach of music, including Hip Hop, R&B, Rap, Rock, Pop and Country songs. Artists such as Ludacris, Michael Jackson, NAS, Busta Rhymes, Keith Sweat and Musiq were very common throughout the subpoenas. They've even created a helpful chart showing exactly what artists and songs seem to get one flagged." Update: 07/31 13:12 GMT by H : Here's another source for the chart.
http://perljam.net/misc/p2p/
Most popular:
Busta Rhymes Pass the Courvoisier (12)
Avril Lavigne Losing Grip (8)
Avril Lavigne Complicated (6)
Incubus Nice to Know You (6)
Marvin Gaye Lets Get It On (6)
Musiq Halfcrazy (6)
Tracy Chapman Fast Car (6)
-ted
KSpread will open it just fine.... I don't know what you're trying to use.
If you don't like Excel, you can grab the files in HTML format (25.2 KB) or in OpenOffice.org Spreadsheet format (10.4 KB).
The roots of education are bitter, but the fruit is sweet.
--Aristotle
THe article claims that from 50 total subpoenas being checked, they can deduce overall proportions of artist representation in the subpoenas, which is, frankly, a load of crock; with a sample size that small, margin of error would be enormous.
Ah, one of the great statistical fallacies... "sample sizes must be large to be valid". Not entirely correct.
Assuming a distribution, and reasonably random sampling, a sample of 50 would be plenty for single-digit accuracy, by my BOTE calculation. The problem is, what distribution shall we choose? Song preferences are clearly not Gaussian; personally, I'd guess Zipf.
But that's only a guess; not knowing the distribution is a complete stopper, and it can only be answered with extensive surveying of lots of data, which isn't about to happen for this study. It's not the sample size preventing good statistics, though, it's lack of knowledge of the distribution, which is a completely different matter. (Actually, it's a bigger problem, requiring much more data to be collected to answer the question, well beyond merely scanning the sued people.)
- Why didn't he use a third column for the count?
- Does the absence of "(X)" mean "one appearance" or "zero appearances?"
- Why use Excel for something so trivial, rather than HTML, RTF, or even ASCII?
- Is you insist on delivering the data in Excel format, why not deliver it organized in a useful manner?
On a side note, opening it in OpenOffice and saving it right back out to OpenOffice format results in a file 1/3 the size of the excel file.Napster-to-go says "Fill and refill your compatible MP3 player", which is a lie. It's not MP3. It's WMA with DRM.
Just chatted with my investigator friend at the RIAA again. He told me they've got this whole operation outsourced to online investigators (not sure exactly what that means) and law firms. They're budgeting the effort as a simple cost of doing business. They do in fact have patterns, schedules, etc. This is just going to keep going until a group finds a common defense and can start making this more costly for them. Otherwise he said that internally it's clear they're following this road as long as they can.
He also mentioned that they're now paying for staff at ISP's. Basically with the Verizon case everyone is ready to roll and RIAA finishes them off by offering to pay for the staff increases needed to fullfil the subpeonas.
Personally I haven't bothered downloading music since shortly after the Napster demise, but this stuff is bullshit. I really hope the folks getting targeted can band together with some sort of tenable defense and start making this more expensive for them. During the Napster case I was told by this same guy that RIAA was getting short on funding and the labels weren't willing to cough up extra cash for the case. It sounds crazy, but maybe enough individuals could eventually team up, get all cases into a single jurisdiction, and try to start bleeding them again. They're big, but there funds are not limited. Certainly a long shot, though, and expensive for everyone involved.
I am happy to present my results in the form of a new spreadsheet, a CSV file and a GIF formatted graph. I am too hungover, and too rotten a statistician, to draw any conclusions. Enjoy.
I used to have a better sig than this, but I got tired of it
how the hell you got an insightful I have no idea.
Small bands THRIVE on p2p sharing of their music.. Hell Every one of them that I ask give me permission to use their music in movies or ad's without anything but a copy of what we used it in.
they know that the only way to make it is to get people listening to their music, the radio stations are owned by the record companies and therefore wont play them (Don't even try to tell me they are not... I watched the payola go down for 2 years when I was in radio and friends today tell me it's worse now..) and they make their real money on venues and shows. EVERY one of them tell me they sell their CD's at the shows only... because they can't get them sold anywhere else as the stores don't want them.... even the small record shops won't let them put a small amount on their shelves at cost.
P2P sharing of music is the best thing to ever happen to a small band.
Do not look at laser with remaining good eye.