Slashdot Mirror
The Nielsen Family Is Dead
An anonymous reader writes "An article at Wired walks us through how the so-called Nielsen Family, responsible for deciding which shows were good and which were flops since the '70s, isn't the be-all, end-all of TV popularity anymore. Quoting: 'Over the years, the Nielsen rating has been tweaked, but it still serves one fundamental purpose: to gauge how many people are watching a given show on a conventional television set. But that's not how we watch any more. Hulu, Netflix, Apple TV, Amazon Prime, Roku, iTunes, smartphone, tablet—none of these platforms or devices are reflected in the Nielsen rating. (In February Nielsen announced that this fall it would finally begin including Internet streaming to TV sets in its ratings.) And the TV experience doesn't stop when the episode ends. We watch with tablets on our laps so we can look up an actor's IMDb page. We tweet about the latest plot twist (discreetly, to avoid spoilers). We fill up the comments section of our favorite online recappers. We kibitz with Facebook friends about Hannah Horvath's latest paramour. We start Tumblrs devoted to Downton decor. We're engaging with a show even if we aren't watching it, but none of this behavior factors into Nielsen's calculation of its impact.'"
I look at the pirate sites to see what's popular on TV. That's a truer reflection of what the general public wants to watch, because the seed and leech count isn't some complex proprietary formula. While fakes do pop up, with companies trying to poison the peer population to discourage downloading, the protocol is self-correcting and it is really just further evidence of its popularity. It represents an intentional and willful effort to watch these shows, not just a casual interest because it feels less lonely than leaving the TV on to blare commercials while you do something else. If you want to know which shows are popular, not just locally, but internationally then torrent sites are really the best measure of a show's actual popularity. And it's not limited to TV either; A movie's true popularity is also reflected in the download count, moreso than an imdb rating.
You can't trust for-profit organizations to give fair an unbiased numbers -- for enough money, they're only too happy to rig the system. There's companies whose sole reason for existance is to push books onto the New York Times' best seller lists. Because sales data and other information is all kept hidden behind a wall of corporate proprietary data, it's possible to rig the system.
The pirates... you can't rig the system. Either it's popular, or it isn't. No games, no bullshit.
#fuckbeta #iamslashdot #dicemustdie
These days Amazon, Hulu, Netflix, etc. know not only if you watched a show, but how many times you watched it, where you paused, which parts you re-watched, etc. (FYI, hook up Wireshark or Firefox's web console and see for yourself what information is being logged!)
The quantity and quality of the data is better than ever. As more people switch from broadcast and cable to online streaming, why would you need a random sample like the "Nielsen family"? It seems so blunt compared to the accurate real-time data that streaming services can provide.
There's no -1 for "I don't get it."
You can rig it. There's little to stop you from downloading something many times.
BitTorrent Trackers don't count downloads - they count the number of active seeders and leechers.
While you are connected to the tracker and downloading a file, your session will add 1 to the leech count. When the download completes, the leech count is decremented 1. If you delete/move this file and initiate another download, this will simply temporarily add 1 to the leech count again.
Sure it is game-able. You could have multiple clients on different ports simultaneously downloading and hope that the tracker doesn't amalgamate all connections from the same IP to count as one. This, however, would be very taxing on your connection not to mention extremely resource intensive, especially with the wide adoption of crypto in BitTorrent.
Even if this was actively rigged, would it be any worse off than the Nielsen system? Can we honestly believe that it is not completely rigged? The participants are known ahead of time. Is it not possible that they be chosen according a particular agenda or otherwise enticed.
[Rent This Space]
Nielsen has one and only purpose - to help price ad-time buys. Shows on bittorrent have had the ads stripped out. The people watching those versions might as well not exist for all that Nielsen's customers care.
When information is power, privacy is freedom.
At first I thought this was another American gun massacre story.
I'm not watching TV since 2005. Such a waste of time.
Unlike slashdot
That's dumb. The main value of Nielsen is not that it measures popularity. It measures viewership of the shows (and thus ads) that air on TV.
Ad rates are tied to Nielsen ratings. And this has not changed. So the Nielsen family is as relevant as ever.
Sure, Hulu stats matter too, to fix the price of ads that air on Hulu.
It's amazing how many people think that somehow because they don't watch TV that it is not longer relevant what TV ratings are.
http://lkml.org/lkml/2005/8/20/95
The demographic that gets/views torrents is skewed towards the technologically minded.
Then explain to me the popularity of Jersey Shore on the Pirate Bay.
Most popular Jersey Shore episode on TPB: s06e13 with 104 seeds and 4 leeches.
Most popular Big Bang Theory: s06e19 with 19202 seeds and 538 leeches.
There... "popularity" of Jersey Shore on TPB explained.
Slashdot social media options: AIM, ICQ, Yahoo, Jabber and Mobile Text. Why no MySpace?
There... "popularity" of Jersey Shore on TPB explained.
Last airing of Jersey Shore: December 7, 2012
Last airing of Big Bang: Today.
I think I see a flaw in your cunning assessment. If you look for a single episode of Big Bang uploaded on or before the same date, you get about the same count: 128 seeds, 2 leeches, respectively. When Jersey Shore isn't on its off-season, those numbers will be a lot higher. But I can't fault you for not knowing that torrents of TV shows tend to be most popular when first released, and then quickly drop in both seed and leech count... I mean, it's not something the average person would know.
#fuckbeta #iamslashdot #dicemustdie
I think this article, and probably many of the comments, shows how little people understand of how the ratings and broadcast networks' business model works.
The "family" has been dead for years, ratings-wise. The only numbers that count -- at all -- are adults 18-49, and within that group women 18-34 are particularly valued. That's why singing and dancing competitions rule the airwaves. If you are under 18 or older than 49 your TV viewing habits do not matter to advertisers, they do not matter to networks. An 100 million kids could watch a prime time TV show, and it will still get canceled if not enough adults are watching.
Why? Because TV networks do not have viewers as customers, it's the advertisers that pay their bills. And the advertisers have decided that those are the only age ranges worth selling to, on prime-time TV.
Online, DVD sales, international sales do NOT bring any revenue whatsoever to TV networks, and no matter how popular a show is off of a US TV set, it is worthless if it does not have an high rating in the key demo. Unless -- and only, unless -- the Network is also the production company for that show. (but most are not) Production companies do make money from DVDs online purchases, rights and online ads -- so a company (such as amazon or Hulu) can bypass the Networks and produce successfully online, as is now happening.
I do disagree with the advertisers age ranges, and feel they could monetize the younger and older audiences as well. But I do also understand why they feel they can reach these audiences easily without any need to pay for expensive TV ads.
We are probably reaching a transition point in TV viewing anyway. A business model like the MLB.TV model is one that probably works best. A worldwide 24/7 online TV channel paid for by subscription and/or advertising. It provides full demographic info in real time, allows one-click purchasing to firms, and it allows for long-tail and niche programming too. That is a much better model for advertisers and viewers -- but not too good for the network middlemen, unless they jump on that bandwagon right now.
As an aside, similar is true for movies -- which have a totally different demographic (12-24 usually). Long, long gone are movies like "The Sand Pebbles". Why? Because adults do not go to the cinema in sufficient numbers to matter, unless they are taking their kids to see a kids movie. There is very little overlap between TV and movies in terms of significant audience. Movies are only for children, and TV is only for adults these days (and female adults mostly too, since men are easy targets through sports).
Good point. I guess one could connect to the tracker and announce that they are seeding/leeching and simply drop all incoming connections. This is highly probable, though the multitude of leechers coming from the same IP / IP range is sure to tip of the tracker maintainers and BitTorrent community quite promptly. The community will likely respond with a patch to count only 1 connection per IP and to disregard counts for inactive downloads/uploads.
Actually, just such a patch has been built into all mainstream bittorrent clients for years. Clients will only accept 1 connection at a time from a given IP address. And if the data provided is incorrect, after a certain number of bad chunks (defined in the client options, but typically around 3-5), it will be banned.
That said, some trolls did try to interfere with the downloading DVD screener copies this past year right around the time the Emmy's were being voted on by registering thousands of fake peers with the trackers, in a sort of DDoS. The reasoning is believed to be that if they could lower the effective download rate or otherwise make it take a long time to download the torrent, people would give up. Unfortunately for them, their cunning plan failed to consider that computers do not "give up"; After a few hours, all of their fake peers had been attempted (and banned by each other participating client), so while the length of time did increase for the torrent, it was not by an appreciable amount -- it doesn't take long to send 68 byte packets to a few thousand, or even ten thousand, unique IP addresses, and you don't need to get more than a handful of non-fakes to get your download up to full speed.
The other, more successful, method was to seed fake torrents with similar names and filesizes to the legitimate ones, thus forcing people to waste large amounts of bandwidth to get rick-rolled (proverbially speaking). The files would be corrupt, have severely distorted video and/or audio, or simply be a "Shame on you" advert repeated over and over. Very shortly after this, all the major torrent sites introduced the notion of "verified" torrents, and allowed anyone to rank a torrent, or otherwise flag it as crap. The practice has since stopped for the same reason spam e-mail usually doesn't make it through: A web of trust is a simple, yet powerful way, to sort the chaffe from the wheat.
So while there are ways to attack the bittorrent protocol, they are expensive and only result in a small loss of time and effort. For this reason, these attacks aren't common anymore, though less-technically minded groups (I'm looking at you, form-letter enforcement companies) perenially make the attempt thinking nobody's ever done it before. :)
#fuckbeta #iamslashdot #dicemustdie
the moment VHS and Beta came on the market and allowed us to tape our shows, that's when Neilsen became obsolete. That and the fact that there are certain types of shows which timeslot and popularity have no relationship. We are empowered to watch a TV on our own time by being able to record it. Neilsen ratings have been a failure for a long time, but then again, the entire sponsoring mechanism behind funding a show isn't all that good either.. Look at shows like Stargate SG-1 who didn't rely on sponsors.. lasted 10 seasons and spun off 2 other shows and direct to movie DVDs, etc..