Slashdot Mirror
Is the iPod Shuffle Playing Favorites?
marksilverman writes "Steven Levy at Newsweek is reporting that his iPod Shuffle seems to favor certain songs. Is Apple receiving kickbacks to promote certain artists? Apple denies it, of course, and Levy had the good sense to ask a mathmatician and a cryptographer who explained that it's probably just humans finding patterns where there are none." Less neurotically, both CNet and PCWorld have discussions of the Shuffle's interior spaces.
I have heard YMCA an inordinate amount of times on my ipod shuffle. I wonder what that means.
...look for patterns, to at least internally provide an explanation. (whether it's true or accurate is irrelevant.) go to craps table - unless you are neurotic, everyone develops a pattern to how they roll the dice. no reasons, no explanations. we are just made to take emotional comfort in attributing some pattern, real or otherwise.
If i Hear Fiona APPLE one more time, I'll kill myself.
Levy had the good sense to ask a mathmatician and a cryptographer who explained that it's probably just humans finding patterns where there are none.
Didn't stop him from submitting it to Slashdot though. No facts? Great, put it on the front page!
"Levy had the good sense to ask a mathmatician and a cryptographer who explained..."
Why in the hell would you waste your time finding and possibly paying someone for such common sense? Maybe you have it set to play based on ratings which can be set in iTunes, dunno, but seriously use your head.
... A bad random number generator. It happens.
Maybe they just have a weak random number algorithm? Or as the blurb says, maybe they have a strong one and it's just all in people's heads.
Dare any of ya to come up with a way Apple could do the whole "kickbacks" thing and actually make the implementation work. It's just journalists wording things special to slant the facts and try to get a rise out of people.
It's not happening..
From the article:
I explained this phenomenon to Temple University prof John Allen Paulos, an expert in applying mathematical theory to everyday life. His conclusion: it's entirely possible that nothing at all is amiss with the shuffle function
The slashdot article??
Is the iPod Shuffle Playing Favorites?
and
Apple denies it, of course
Enough with the inflammatory headlines!
Quidquid latine dictum sit, altum viditur
This post brought to you by Teenage Angst and Caffeine!
Elmo knows where you live!
I think slashdot favours certain articles and repeats them over and over again.
Clearly we need to lock an ipod in a room with the same scientists who discovered revolutionary psychic patterns in that little black box a couple of weeks back on slashdot.
Perhaps Ipod will predict when Hewy Lewis and the news will make a mainstream comeback?
You're absolutely right!@ There's an array inside the iPod shuffle of about 150 artists that will take precedence over all other artists and will play songs by those artists 3 times more than all others because because Apple recieved $100,000 per to make it happen.
Those of you that think that sounds completely plausable, please step to the left. Everyone else please step to the right.
Everyone standing on the left, please drink the magic juice we're distributing because you've been selected to come with us to the great beyond where you will experience another plane of being.
Everyone on the right, enjoy the rest of your life because you enjoy logic and reason.
I just wasted your mod points! HA!
Computers can't generate true random numbers (ok, at least I don't know of any current methods) but only pseudo random numbers. There's a precise mathematical description that gets you from one number to the next.
Who knows, maybe Apple uses the meta data for a song to determine the random order (anyone hack it yet and finding the algo?) and some people just get "lucky" like prof John Allen Paulos explained in the article. You might happen to flip 6 heads in a row (despite being a 1 in 64 chance of it happening) and you might get 6 songs from an artist in a row.
Sounds to me that it's a conspiracy theory at best.
:wq
If you have any audiobook or spoken word that have proper metadata, they will never be selected in shuffle mode. While this can be handy for not falling onto a 20 minute chapter of a book randomly, it also makes it a bit more difficult to create cut-ups, or experiment with random spoken words when you want to: you must retag the tracks.
This is also useful to take long tracks out of random selection. A friend retagged Pink Floyd's 23 minute long 'Echos' as a book after getting pissed off that his Shuffle always seemed to select it.
The summary suggest that Apple may be playing favorites, citing an article that concludes pretty definitely that they are not...
Please correct me if I'm wrong, but you don't get this on a normal iPod because (1) there's far more songs to choose from and (2) because the iPod builds a playlist randomly when you set it to shuffle (this is the part I'm not completely sure about, but I've let mine play out for the four days or so worth of music I have, and it eventually gets to the end and stops playing, just as it does when playing from a playlist).
With the iPod Shuffle, there's fewer songs, and it's more likely that the user would want songs repeated, so it seems logical for it to choose the next song one at a time, rather than building a list. Thus, it appears to be playing certain songs more often.
Either way, just hit the "next" button if you don't want to listen to the current song.
The beginning of iPayola?
On macslash.org, there was an Ask MacSlash about iTunes somehow figuring out what songs sounded good together using some crazy sonic algorithm. The guy was asking if it was plausible that iTunes analyzed the songs in the playlist to spit out the best mix possible.
The basic consensus in the discussion was either "dude, your entire playlist is songs you like, of course it's gonna be a good mix.," or, the option mentioned above about humans looking for patterns.
Although, throughout my history of having large (over 1000 song) playlists, I've found that no matter what mp3 player I used (hardware, software, or otherwise), there always seemed to be certain bands or artists that would get play more often. I've had weeks at a time where I'd hear Snoop Dogg's Lodi Dodi, Iron Maiden's Quest for Fire or In Flames' Clayman nearly every time I picked up my iPod
...spike
Ewwwwww, coconut...
When looking at the candies through the side of the glass jar, the first thing you notice is that the distribution of red and green candies doesn't look evenly distributed at all. Instead, there are lots of areas where many red candies are adjacent, and lots of areas where many green candies are adjacent.
For a long time, many people thought there must be some kind of static electrical effect present that was causing candies of the same color to tend to stick together. Eventually, however, some statisticians did the math and found that there was no such effect at play -- in a completely random system, such "blobs" of like colors are inevitable. Indeed, a jar of candy with no such blobs would be a bit suspect -- what are the chances of the red and green candies always pairing up so that no groupings occur?
To put it another way: it's all in your heads, guys.
I don't care if it's 90,000 hectares. That lake was not my doing.
The odds of the iPod playing a completely even distribution of every song, with no detectable 'pattern', is MUCH smaller than it playing a list that has a detectable 'pattern'. Think about it this way.
Songs 1, 2, 3, 4, 5.
And you want to play 5 songs. To have it play each song once and only once would be 5!. However, this is out of the number of choices total, which is 5^5.
In other words, it's 120/3125 that the distribution would be even. As opposed to 3005/3125 that it wouldn't.
anyone know if it's a standard USB mass storage device (i.e. can I mount it in Linux and XP sp1 computers w/o itunes)? And how long's that battery suppose to last (and does it cost more than the unit to get replaced)? TIA.
Hi! I make Firefox Plug-ins. Check 'em out @ https://addons.mozilla.org/en-US/firefox/addon/youtube-mp3-podcaster/
Seriously. There have got to be better submissions in the queue than, "I think my iPod shuffle is preferring certain songs over others. It's a conspiracy!" Come on.
...but when I put my 20GB 4G iPod on Shuffle play mode... there are some songs it just seems to NEVER play.
If you're listening to songs on your iPod, you've already bought them. Either through iTunes, ripping a CD, or maybe you just ripped it off from Kazaa, but no matter what they're not looking at any more sales.
LK
"Hi. This is my friend, Jack Shit, and you don't know him." - Lord Kano
winamp has done the same thing to me... I have 230 some odd CD's on my pc and if I queue them up in winamp, the same songs will tend to play during the beginning of listening... eventually, I just scroll randomly to find something different. Seriously, there is something to this - maybe a faulty algorithm, but there is something. After all, 230*(~13songs) is too many to be coincidence.
Not just slashdot, whats up with CNN reporting each new ipod release as headline news. I like Apple but I hate advertising posing as news, or movies, or slashdot stories...
"And Tonight at 6, can bees think?
A new study reveals that, no they cannot."
I have been hearing a lot of the same stuff... first it was a bunch of 3 Doors Down, and now it keeps picking Alice in Chains!
Oh wait, mine's not an iPod shuffle. Never mind.
Bears don't normally eat things that talk and move backwards.
People are really good at spotting patterns, even when there isn't one - which is why you have to step back and find out what is instead of setting out to prove something.
The iPod shuffle decoder chip can decode WMA files. For some mysterious reason however, this feature is not enabled on the iPod shuffle, even though such a feature would be an obvious benefit to those who would like to purchase music online from someone other than Apple.
but i just bought a mini, and i favor it quite a bit :)
Reminds me of a Dilbert comic:
Accounting Troll: "Over here we have our random number generator"
Number Generator Troll: "Nine Nine Nine Nine Nine Nine"
Dilbert: "Are you sure that's random?"
Accounting Troll: "That's the problem with randomness: you can never be sure"
...now mostly in shuffle mode. The oddest thing I've noticed is that sometimes XMMS has a tendency to play two songs in a row by the same exact artist. What's more is that sometimes it will play a song by a particular band and the very next song will be a song from a solo project by a member of that band! But I don't think it has anything to do with XMMS.org setting up a "two play" feature or anything. I think it has more to do with how my music is filed and the algorithm that they use to randomize the music. For example, here's what happened today:
;P
Track 1: Japan - Visions of China
Track 2: David Sylvian (former vocalist for Japan) - Wave
Track 1: The Thompson Twins (TT) - We Are Detectives
Track 2: Babble (New project by former TT members) - Just Like You
Track 1: Thomas Dolby - Airwaves
Track 2: Ryuichi Sakamoto (feat T. Dolby) - Field Work
Track 1: Depeche Mode - Shake the Disease
Track 2: Depeche Mode - Knocking on Death's Door (from a soundtrack that is unrelated to DM otherwise)
This happens all the time, but I still don't think it's XMMS doing anything intentional. Usually when this happens, I've loaded my 'all.m3u' playlist which contains all 7000+ songs I have stored on my server. The songs are all stored in artist folders first, and then albums within the artist folders. In total, I have about 346 artist folders. The only thing I can figure is that XMMS must use some kind of folder proximity algorithm which is succeptible to the way I've organized my music and therefore less random. Either that, or it's all quantum physics and my observation of the music quanta (ie. listening to the waves) is altering the other quanta through entanglement and I'm mental energy is shifting bits in my PC which cause XMMS to play "two fers" almost 80% of the time.
-"...bad old ideas look confusingly fresh when they are packaged as technology" - Jaron Lanier (Digital Maoism on Edge.o
Holy Christ, this story is a MONTH OLD! Anyone who gave a shit read it four weeks ago, and anyone who didn't hear about it when it was ALL OVER THE INTERNET back then isn't going to read Slashdot anyway.
I concur. There HAVE to be better submissions in the queue than month-old crap that wasn't even all that interesting the first time around.
Ah well. At least it's not a dupe...
p
In Korea, long hair is for old people!
One of the articles about the Shuffle internals estimates their price to have about 40% markup over hardware costs. If that's true, is it the R&D department we're financing, or just paying for the privilege of a fashionable item?
http://zero-to-enterprise.blogspot.com/
With a little bit of Logic and Reason, you could make your own music.
Then, you can offer apple $100,000.00 and see if they'll add you to their magic artist array!
Of course, you might have to pay out to the other artists on it, too, since you'd be decreasing their play frequency by 2/3 of a percent...
SELECT quote.text AS sig FROM quote NATURAL JOIN attribute WHERE attribute.description = 'witty';
0 rows returned
If i Hear Fiona APPLE one more time, I'll kill myself.
Perhaps you could remove it from your iPod Shuffle loadout?
pants
It's quite common for random processes (like coin tosses) to get unlikely results here and there, like runs of six heads in a row. Over a very long time, it evens out, but it's hard for us to envision that. "We often interpret and impose patterns on random processes," he says, adding that this might be expected in the case of music, which evokes strong emotions. Paul Kocher, president of Cryptography Research, puts it another way: "Our brains aren't wired to understand randomness."
Which explains why so many people insist on losing money at casinos. Especially roulette. Yes I know its a long quote, for a short comment.. but what can a guy do?
There are some quality random number generators on the Internet like Random.org, HotBits, and Lavarnd. But to be technical, their numbers come from background radio noise, radioactive decay, and lava lamps (er, "Lava Lite lamps"), meaning they don't truly generate it on the chip.
This sig is certified free of self-referential humour!
If you download tunes from your own mp3 collection onto the IPod shuffle and call it good, it doesn't really matter what Apple does.
people need to understand that "even" distributions are not random. clusters exist when there is true randomness. learned this little snippet of knowledge from an episode of numb3rs when an algorithm was being devised to locate a serial rapist turned murderer based on his seemingly random series of attacks. so i guess people would prefer listening to a single song from a randomly selected artist from their collection, instead of a random song.
http://www.sledgehammercomputers.com
...but doesn't shuffle favor your higher rated songs?
...I could be way off base here, I don't remember reading this anywhere in particular, I just thought that was the way it worked. Can anyone confirm/deny?
So, I know its us the loyal readership that is supposed to do the editing essentially (which makes me think we should have a moderation-style system for voting on which stories in the queue should go live (editors, you reading this?!!) but hey!) but when our favorite web tool has spell-checking built in, and you can search up to 32 words at a time.... I mean you might as well just cut and paste.
What's really interesting, that even though this is a geek site, we can't even get mathematician right. Even more silly is if you check the link above from google, you'll see that the first two results (of 3) are also from right here at the dot.
All I'm saying is: if it's power to the nerd masses, let's do it. Many posts so far are already complaining about the story. Not to mention it's from MSNBC. Not to mention I've already read it, because it's from almost four weeks ago.
I realize I am off-topic and complaining, but I wanted to see if we couldn't get a discussion going about a smarter, more democratic way to elect submissions to go live.
So before dismissing thousands of people, I'd entertain the idea that Apple's engineers simply stuffed up. It wouldn't be the first bug that slips through QA testing.
Have you tried the iPod shuffle lately? We hear it plays favorites...like trolls do on Slashdot! *sheepish grin*
So while you may be hearing "BURN, MTI, BURN!" on your Zen, we here at MTI keep hearing that same ol' tune "When MTI Comes Marching Home!!"
I expect the same thing is happening here. People have some songs they like (or dislike) more than others, and so they note when they are being played.
Sure, apple is spending R&D costs on creating a way to get their audio player to promote songs that users have already bought. Thats makes a ton of sense.
I think I've noticed how Slashdot tends to repeat articles on Google when on shuf^H^H^H^Htimothy...
I have two songs on my shuffle, but I would swear it plays this one tunes like 50% of the time. And it is a U2 song. So obviously they are trying to hypnotize you into buying a black iPod. Duh.
.. will be the day i give in to that music app for the rest of my life (well, let's say works as one would expect) I currently have over 14,000 songs in my library, and I've used every version of winamp along with the last two years of iTunes and I've yet to find one with a decent Random Number Generator. I mean, if I ave 100 songs of Pixies and I choose library->random why are 3 of the first 13 songs Pixies? For the last five years, that's been my biggest groipe about any music app. i mean, c'mon. I've given up a lng time ago... It's both a science and an art. Read up on RNG before you implement the most generic shuffler you can. these don't cut it. i'd rather shuffle potatoes all night. choose a random open computer on the net and extract their desktop color and use that! do something original. shuffle has been dead for a long time and it's time to revive it! (even though shuffle is noew creating a new to "look at" music.)
Umm, why are we talking about this "conspiracy" when the last link talks about things like an FM tuner and an LCD controller in the Shuffle? Isn't that much more interesting? ;)
It IS an interesting story. And there are facts.
1. People think there's patterns on the iPod shuffle
2. Apple says there isn't
3. The math guy says there probably isn'
4. People tend to look for patterns
Don't take life so seriously. No one makes it out alive.
If someone will buy me a Shuffle, I'll test it out and see if it's true. Then I'll report back.
Less neurotically, both CNet and PCWorld have discussions of the Shuffle's interior spaces.
For a hands-on discussion of the Shuffle's interior space, including lines like "If you remove the wheel like I did in the picture you will never get it back in", see An iPod shuffle on the inside.... Sure beats C|net's stock photo's.
this article came out on January 31.. and it has no basis - why on earth do we have a statement like "Is apple getting commissions?" Is slashdot trying to go lower than the National Enquirer?
"When the only tool you own is a hammer, every problem begins to resemble a nail." - Abraham Maslow (1908-1970)
than pseudo-random?
This reminds me of the birthday paradox:
"The birthday paradox states that if there are 23 people in a room then there is a slightly more than 50:50 chance that at least two of them will have the same birthday. For 60 or more people, the probability is greater than 99%. This is not a paradox in the sense of it leading to a logical contradiction; it is a paradox in the sense that it is a mathematical truth that contradicts common intuition. Most people estimate that the chance is much lower than 50:50."
Applied here, suppose you have 365 songs. How many random selections must be played before you have about a 50:50 chance of hearing a repeat? Just 23 songs.
What most people want is not random selection, but random order.
Im not saying that Apple is pushing anything, by my iTunes loves the band Hole. Now I have two Hole albums out of 37 days of Music in iTunes. But if I use random on my entire library ever hour or two I hear... you guessed it Hole. Only about one in four hundred songs should be Hole if it really is random, but with iTunes random its more like one in forty. Stupid iTunes, I mean really couldn't it have picked a better band?
Even stranger is it use to love this one song, Closet Romantic by Damon Albarn from the TrainSpotting Soundtrack. My iTunes use to play this one song very often on random, always in the first say eight hour of randomness. This all stopped about a year ago when I added another 50 so albums so my best guess is that song was on some algorithm sweet spot.
I would really like to see how iTunes random really works though, as I might be over analyzing, but it seems to throw in alot of double-shots.
nope
I just chalk it up to a bad shuffle mechanism.
In fact, I already had Party Shuffle before Apple added it, and I like my verion better. I have a smart playlist set up as:
Match all of the following conditions:
My rating is greater than 3 stars
Last played is not in the last two weeks
Genre does not contain audiobook
Genre does not contain comedy
Genre does not contain television
Genre does not contain theme
Limit to seventy minutes selected by random (so I can burn a CD for the car quickly if I choose).
Live Updating.
The only downside is I can only skip tracks until I reach the bottom of the list, as new tracks don't fill in until a track is played and taken off the list for not meeting the "last played in" stipulation.
He gives one good example of why Itunes/Shuffle is screwed up. He mentions that he has a libary of 3000 songs, and the ITunes has a feature to choose random songs from your libary to put in the IPod. He says that 3 songs from the same album were selected from the same album.
I did some Math and programming so I could calculate every possible combination of songs, and every possible combination of selecting 3 songs from _That_ album(assuming of 10 songs). It worked out to be .5% chance that the ITunes would select Three songs from THAT album. Take in account he might have almost 30 albums. You can easily see how it might be very likely that the computer could completely randomly choose 3 songs from the same album a large portion of the time.
I think the sad fact is people dont want random, they want an algorithms that select music in some equally distributed way every time.
mnewberg.com
I noticed that too.
I have a 1GB iPod shuffle, currently it contains roughly 100 songs.
My iPod shuffle "prefers" playing BAD RELIGION and SLAYER.
I don't know how or why.
Maybe you guys can help me figure this out.
here's the breakdown:
100 tracks (roughly)
40 BAD RELIGION's song
40 SLAYER's song
20 of other bands
THIS IS REALLY FRUSTRATING that the iPod shuffle prefers playing SLAYER and BAD RELIGION.
HELP!!!!!!!
(***casm mode ends)
Maybe the editors have a *go with me on this*
:-D
Slashdot Shuffle (TM) system for the articles they post.
It would explain the dupes and the lack of quality control with one theory.
I keed, I keed
Will infinite Shuffles produce produce pirate stories?
rrrrr! RRR! rrrrr! rrrrr! rrrr! rrrrr!
RRR! rr! AVAST, RRRIAA! rrrr!
-gko
First, there is the possibility that Apple screwed up the shuffling algorithm -- although not entirely likely. If you ask an introductory programmer to write some code to shuffle an array, you'll most likely get something like this:
for i in range(array_length):
j = random() % array_length
temp = array[i]
array[i] = array[j]
array[j] = temp
This code does NOT produce all permutations with equal probability! Instead, you must use the following code:
for i in range(array_length):
j = i + (random() % (array_length - i))
temp = array[i]
array[i] = array[j]
array[j] = temp
}
This was cribbed from c2 -- see the full article text here for a more informative discussion.
Second, I see a lot of people saying "I have a 20GB iPod -- and I swear sometimes it just NEVER plays this one song." Okay, let's assume that a 20GB iPod holds 5000 mp3 files. What's the probability that you play 5000 songs in shuffle mode, and never hear a particular song?
It's the probability that 5000 times in a row, you hear some other song -- that is, one of the 4999 other songs. Calculating, we get:
(4999/5000)^5000 = 0.3678.
So we have a 36% probability of this happening -- which is not a negligible amount! This will further be compounded by two things: First, you have no way of recalling exactly it has been since you heard a particular song -- if your favorite song was played 1000 songs earlier, it probably feels like 2000. If it feels like 2000, it's probably 4000. Because it's a favorite song, your mind will exaggerate the amount. It's like if you crave nicotine, it can feel like days since you've had a cigarette when it's only been hours. Second, you probably have a lot of songs you would call a "favorite" -- with each having a 36% chance of not being played over the course of 5000 plays, your mind will probably register that at least one of them is "feeling neglected."
Probability is a strange and beautiful thing. Don't expect your average audiophile to understand it. (And I'm not claiming to understand it either, beyond a very cursory level.)
- shadowmatter
Geez, I put one showtune on my iPod, and now it only plays Liza, Barbara, and 70s era Disco!
So the reason all the people around me are total geeks and all the hot chicks group elsewhere is pure randomness?
phew! I was worried it might be something else.
For less than $50 in parts (most of them can sampled for free from friendly parts manufacturers, you can easily hack an iPod shuffle... can it run Linux? Yes. A shot of the reassembly.
Apple products seem somewhat overpriced to me. I recently bought a tiny 20GB Archons Gmini for about $200 and am very happy with it.
Excellent sound, lots of space and low price. My kind of deal.
It had a particular fondness for Steely Dan, whose songs always seemed to pop up two or three times in the first hour of play.
I can disprove his theory right there. I listen to my iPod all the time, but mine has NEVER EVER played a Steely Dan song. Not even once. I don't even have any Steely Dan songs. So there, QED.
& I wish I knew the password to your heart . . . &
My iPod Shuffle Thinks I'm Gay!
As far as I can see, the majority of MP3 players take a "play random(maxNumberOfTracks)" attitude to the next track they play. All the ones I have tried, both embedded hardware types (like flash players) and software for your desktop seem to do this. You end up hearing a track again before you have heard all the others.
The only way to get a random list that does not repeat before everything is played is to scamble the order of the play list and play sequentially through that scambled list. You also get a working previous that can hop through all the songs you've listened to if you want to listen to the song you heard 5 tracks earlier again.
This is not a problem of a bad shuffle algorithm. The problem here is thousands of years old. It is human nature. People see patterns where there are none. People generate theories based on these non-existent patterns. This is how people concluded that the sun orbits the Earth: "Oh, look, the sun looks like it is orbiting the Earth! Therefore it must be!"
The problem is, if you just put together a playlist with a bunch of artists and play it, it is entirely likely that someone will be played three times in the first hour. And in this guy's case, that someone was Steely Dan. So what does that prove?
About 500 years ago, we invented something called the "scientific method". Although it is taught to most people in both science and history classes, few seem to understand it.
The scientific method says that you cannot use past observations to make a conclusion. You must develop a specific test to prove or disprove your hypothesis. You must show before you perform the test that the outcome of the test is relevant to your hypothesis. You can then perform the test and use the results to back your conclusion.
The scientific method could very easily be applied here. What this guy needs to do is start with the prediction that Steely Dan or whoever will be played three times in the first hour. He must use statistics to compute the probability of this happenning in a purely random shuffle, and should show that the chances are less than 1% (this is a pretty straight-forward use of standard statistical methods). Then he should run the experiment and see what happens.
My guess? Steely Dan will not play three times in the first hour.
Without a proper scientific experiment proving this guy's theory, there is no story here.
Mmmm... pickle.
Old Europe FB!!!
> You're absolutely right!@ There's an array inside the iPod shuffle of about 150 artists that will take precedence over all other artists.
Back in late 2001, I wrote a simple program which learns which songs I press "Next(b)" before it completes. Finally after 8 weeks, I realized that I listen to- Eminem and other rap in the morning
- Pop music later into the afternoon
- Rock was for the 5-7 pm slots
- After 10 , it was usually playing Enigma and instrumentals
Was quite different on a weekend with no music on saturdays and often slow Elvis songs on sunday afternoonsQuidquid latine dictum sit, altum videtur
Yvan eth nioj, baby!
It would be cool if it didn't suck.
Then why the hell is he keeping it in his i-pod anyway?
Just imagine how they are laughing now while reading this article...
It make their day more exiciting. The Weekly World News is never going out of business. Batboy will forever be on the loose. Jesus is coming back real soon now! Bigfoot was just seen, again, but the photo is grainy. A chupacabras ate a baby. Again!
Your talk of facts and implementation has no sway with these people, especially if they have an anti-Apple grudge to begin with.
People would rather talk about big evil conspiracies than just admit to boring old talk about badly written randomizing code or a poor random number generator.
This conspiracy would be funny if it didnt again expose the shocking lack of skepticism and critical thinking skills of so-called educated people.
Even more depressing, the last Weekly World News sports the title "Osama is dead." Again, wish fulfillment and conspiracies. Its just pathetic.
i do believe they pay kickbacks...
record companys been paying radio stations to play certain songs too...
This is old and it's not news.
I give this the LAME article award.
iTunes Party Shuffle does not appear to be entirely random... My one never picks songs that have never been played - the play count must be at least 1 to get onto the party shuffle - I'm sure there are other criteria too.
For a while I had my MP3 playlist running in Winamp while I did my writing. The songs seemed to play as a soundtrack to whatever I was writing at the time. And then there's the fact that Green Day's Basket Case kept playing as the third song whenever I put it on random. It was rather creepy.
~Kyrthira Phelan~
reason would suggest
that given the nature of the business
regardless of it's overall distribution around a mean
as reported recently in slashdot
seeking outside confirmation
and why is it do you think that humans might be wired to observe patterns ?
Personal facts next time, please.
The easiest wait is just to generate a 32- or 64-bit PR number instead of the 15-bit ones rand() returns. That should reduce the number of "favourings" on the lower ID songs.
As an added bonues, it wouldn't take a couple of seconds to change song in your 100,000 song playlist because it has to sort it every time...
iPod shuffle owners are more likely to be credulous orange sunglass wearing gimps.
This is easily explained by the fact that, to be an iPod shuffle owner, one needs have looked at the range of available MP3 players and then purchased an iPod shuffle.
Margins on the ipod shuffle: Apple is not making a 40% profit margin on the iPod shuffle. That would mean that the distributor and the reseller are getting zero margin. I don't think the shuffle is that much of a loss leader for any reseller to approach it that way. Chances are if someone is walking into a store to buy it, they are not planning to buy anything else.
..., you start the player it plays songs 1 46 32 45 and then you stop it. Start it again, it goes back to 1 and goes through the same sequence again 1 46 32 45, then it is repeating. Even in a device like the shuffle it should still be possible to improve the variance of the start point (say using a RTC chip).
Randomness: Repetition is possible if the seed for the pseudo-random generator does not change. If the generator starts from the same point all the time, then the first few songs in the queue are the most likely to be played. For example if the sequence is 1 46 32 45 87
I particularly like the "iPod Shuffle Internals" articles.
1. The iPod has two chips and a battery. One chip is memory, the other is for everything else.
2. It has a battery, that might be hard to replace.
3. The memory, battery and CPU costs about $59 [512MB]. The article concludes that each iPod Shuffle costs Apple $59, because as we all know, labor and manufacturing facilities and management are free. It is rare that labor and manufacturing costs make a significant dent in costs. ( -- sarcasm)
My library has around 1500 songs. I used to play them all shuffled. After some time I find I listen to some songs more often, especially on the iPod. Possibly bad random algorithm, possibly bad human brain. Now I'm using a bunch of smart playlists to some what workaround this problem.
1. Create smart playlists of "100 most played songs", "100 least played songs", "100 least recently played songs", and "songs added in the last 3 months".
2. Create another smart playlist that includes all songs from the above playlists (union the above lists). Let's name this "Special".
3. Create a smart playlist that is "100 most recently played songs".
4. Create a smart playlist that are songs in "Special" but not in the "100 most recently played songs" ("Special" minus "100 most recent").
Now use Party Shuffle or your iPod to play the playlist created in step 4 and enjoy. They are either new songs, your favorites, or long lost gems. Plus they are not played in the last few hours.
Make sure you check "Live Update". Change the conditions to suit your needs.
People don't want true randomness. If the machine picked each song to play randomly and independently, then it might play the same song twice in a row; and it might never play a song.
Instead, what people seem to want is to hear all of their music, in an unpredictable order. And that's a random shuffle. (Hence the name!) Each individual track selection depends on the previous ones.
That's also fairly easy to do too, of course. But most of the simple algorithms will assume that the set of songs is fixed... It's much harder to keep an even shuffle when adding or removing songs from the set. Maybe this is one cause of the 'problem'?
Ceterum censeo subscriptionem esse delendam.
Maybe songs keep repeating themselves because they're the one's we secretely want to hear? Ie. the randomness of the iPod Shuffle is being compromised by our sub-conscious? What would Roger D. Nelson say? [http://noosphere.princeton.edu]
So has anyone linked their iPod with a lava lamp random number generator (e.g. LavaRnd.org) to see if it picks better random songs?
I've noticed that iTunes must generate a "random" sequence once per startup of iTunes. If after listening for a while you go back and explicitly listen to one of the songs you already heard and then let it continue randomly then it replays the same song order as before. They must seed the random generator once at startup instead of using the clock to seed each time they go to select the next song. So while the playlist is generated randomly, its the same randomness every time until a restart :)
If I can predict the next 5+ "random" songs, it's definitely a pattern and not just in my head.
I have seen this happen (I've got 180 songs on my shuffle and noticed the repeat after about 30-40 songs).
But I don't think it's a conspiracy, more likely a bad random seed or an algorithm that needs tweaking.
iTunes has an option "choose higher rated songs more often" which can have an effect. I left this off, but I'm wondering if it might always think it is on?
The solution for me was the remove all the songs from my shuffle, and then "autofill" them back (from my playlist). That seemed to shuffle the songs up.
I haven't verified whether the new update (1.1) addresses the random song issue in any way.
Is Apple receiving kickbacks to promote certain artists? Apple denies it, of course.
Did marksilverman, a slashdot poster, kill a puppy to get an erection? marksilverman denies it, OF COURSE.
Reminds of the dilbert where he goes to the accounting department and he see the randum number generator troll.
....
Number generating troll: nine, nine, nine, nine, nine,
Dilbert: Are you sure its random numbers?
Acounting troll: That's the thing about random numbers. You can never be sure...
He, who dies with the most toys, wins
Random does not mean everything distributes itself out in some sort of ordered manner so that three songs from the same album won't appear together. In fact to stop this type of behaviour would require a less random algorithm. The behaviour of similar things seeming to be clustered some times but not others is a testament to the random algorithm working properly. In a perfectly random shuffle, it even possible -- although not very probable -- that all songs from all albums could be grouped together.
It's not about how we picture random, because people end up subconciously assigning some order to it while trying to make random suit their needs.
--Muzz
This same story was also covered on NPR... Here's a Link!
Gravity!... It's not just a good idea... It's the Law!
Depending on how the iPod Shuffle seeds its random number generator, it's entirely possible that Levy is frequently reseeding the generator with a small set of values in which case he would eventually perceive a non-random sequence. The RNG is working just fine but the seeder algorithm could be crap.
Going OT, what amazes me is how successful Apple has been selling products that'll end up in landfill within a few years. They did it with the original iPod and have gone over the edge with the Shuffle. The battery will die in a year or two and then the whole thing is worthless. This from people who claim to worry about the Kyoto treaty.
Randomness is too important to be left to chance! :-)
When all I had was a single disk CD player, "shuffle" was a useless activity. Turn the player on and hit shuffle, and the same disk would always play in the same "random" order.
Similar thing happened when I got a 5-disk CD player. "Shuffle" consistently produced inferior song orderings. For instance, shuffle might select most of the songs from the 3rd and 1st CDs and play mostly those before arriving at the least-played CDs and playing mostly those songs. When I put my 5 favorite CDs into the player and hit shuffle, I want a different sequence next time I turn the unit on, otherwise, what's the point?
Now with WinAmp/Xmms, I haven't confirmed this by extensive testing, but shuffle still seems to suck, and I'd rather hear a CD in the "intended" order rather than some software-selected random order.
I also listen to a streaming web radio station, and although the guy must have a 10,000 song archive, there are times when I hear the same song twice or even 3 times in a day. I have not asked, but my guess is again that it's a (pseudo) random shuffle.
At this point, I should probably define "sucks". Basically, I don't want to hear anything from the same CD successively, and I don't want shuffle to alternate between two CDs over several songs, and I'd like all CDs to be selected more or less uniformly... etc. There's a whole list of rules one could make to produce a "DJ-quality" playlist.
Anyway, I must admit that most of this is simply my own perception, since it hasn't really risen to a high enough priority to start chasing it down. What I'm wondering is whether other people have noticed this, and whether anyone has written any kind of shuffle software that does something "better" than "next_song_num = (int) (random_float()*number_of_songs)" for a shuffle algorithm. If not, then I think there's a great opportunity for someone to invent a shuffle heuristic, perhaps with some user-selectable tweaks, for incorporation into our favorite music players.
The iPod Shuffle does not randomly play songs. It shuffles the playlist...then plays the songs in the new order.
whereas a random play function could lead to a single track being played twice as often as another, shuffle precludes that.
hence the name, rather than "iPod Random".
--
Here's your technique:
The distribution of heads/tails calling in your school is NOT even, no doubt heads was called more often than tails.
Over a long run heads/tails comes out even, your friends lose more often as long as you have enough cash to last you out of any short runs of "bad luck"
The paradox is that in isolation each friend did not do wrong to chose "heads" for a single throw.
Sam
blog.sam.liddicott.com
Everyone I know with an iPod (including myself) have admitted that they have "moods." Some days my iPod likes to favor one artist, the next day a different one.
And that's good enough explanation for me.
I love stories like this. It reminds me that /. employees and most of its readers are really dumb, paranoid, little kids. You people really need help.
I understand why a record company would pay for a radio station to play their songs. The more times it is played, the more people will purchase the songs - at least that's the hope. Of course, with an iPod, you've already purchased the songs (or otherwise appropriated them in permanent form). There is no more money to be had.
The article is about the Shuffle function in ALL iPod models... in fact, the iPod Shuffle isn't mentioned very often if at all... it mostly (as far as I recall) talks about the bigger iPods.
http://www.npr.org/templates/story/story.php?story Id=4473363 [NPR.org]
No, no, no, something like
"Apple backing away from unbiased iPod tune selection!"
You click an OK button when a message box pops up. You focus a window when a task is complete, etc.
iTunes and my regular iPod allows me to rate songs with 1-5 stars though (don't know about shuffle, it doesn't have a display AFAIK so I guess not), I've always wondered if that mattered for how often a song comes up. I'd almost say yes, but as has been pointed out, humans try to find patterns. I guess I do too :-p
For a simple (e.g., no shuffling) random number generator that uses its previous output as the seed for the next one, to repeat once is to repeat forever. Of course, we rarely use the raw output of an RNG directly; usually you'd take a modulus to map it to a discrete set of possible values (e.g., 0..9). In that case you're just as likely to repeat as to get any other particular value.
This seems to me to be a version of the birthday paradox. The birthday paradox is the fact that if you have 23 random (in other words no twins) people in a room, there is over a .50 probability that two of the people will have the same birthday (that is day not considering the year).
So, if you have 365 songs on your Shuffle, and you play 23 songs, the chance that two of the songs will be the same is over 50%.
Or perhaps they were seeing a more prominent occurance.
I can't believe this is post worthy. I see this in MMORPGs -all the time-, non-technical people who see a pattern in something that is near random, and then complain about it.
It should be obvious to most smart people that this is pure crap, what I don't get is how it's been legitimized with an article here. Give me a break.
Sigs are awesome huh?
That would only really be possible if you had pop music on your iPod. I'm sure none of you out there listen to that crap, do you?
"We shall party like the Greeks of old! You know the ones I mean." - HedonismBot
In theory, you could use Audioscrobbler to keep track of that information.
/wishes he could program in more than just ksh
You'd need to build something to act on the data collected, but it's an easy start.
I use the audioscrobbler data to realize I'm listening to one group too much and expand my collection. I only have a limited amount of space at work for music, so once I find myself listening to the same thing day in day out, I rip a couple more CD's and keep going. Otherwise I end up getting whole cd's stuck in my head and then I burn out on them.
I agree - a plugin that would make educated guesses about what music you'd like to listen to would be great. just load your whole list into it, and it, after training, would play appropriate music for that day/time of day. Would be very cool in fact.
We emerge from our mother's womb an unformatted diskette; our culture formats us. - Douglas Coupland
The skewing effect requires that RAND_MAX is not a multiple of num_songs, and is very tiny when RAND_MAX is very much larger than num_songs.
But to see how the skewing arises, imagine you have num_songs = 3 songs with id's 0, 1 and 2, with random() returning 0 to RAND_MAX = 9. Then for random() returning 0, 3, 6 or 9, song_id is 0, so there's a 40% chance of hearing song 0, which presumably is "In The Navy".
I have had a second generation POD (Apple iPod mp3/aac/wav/ALLAC player) for a while. I swear that if I listen to a song for 20 seconds and push the next arrow, a song by the same artist will come up. That is the only consistent pattern I have found in the shuffle but I have had some great luck after loading up the pod with Smart Playlists of my favorite songs.
Apple's users wouldn't be Apple's users if they weren't nutty as fruitcakes!
Most of them would be better served with a bottle of Paxil or Lexapro than an iPod.
Best Buy can have you arrested
Sound like a classic random seed generation problem to me. Just shuffle the array a couple of times before starting and everything should be fine.
Well, you're not Fiona Apple, and if you're not Fiona Apple, I don't give a rat's ass.
--
"Outlook not so good." That magic 8-ball knows everything! I'll ask about Exchange Server next.
a time-seed would (and probably does) work in itunes and the full-size ipods, tho.
maybe that's why the shuffle seems less "random" ?
iPod mini: Pay almost as much for 1/10 of the space!
iPod shuffle: Y'know the shuffle feature that comes with the iPod? Get only that and nothing else!
The new iPod photo I can understand. These, tho, I just don't get.
Well, I can see the problem right now. You've got 40 Bad Religion songs, when everyone knows there have only wrote one song. If you get rid of the 39 other copies, it should fix itself up right quick.
HTH
It was probably an illusion or a bug. But the idea that the reply "random means random" was made and believed means the engineers don't know shit about randomness or think little about the manager asking the quesion; the manager in fact knows shit about randomness; and of course that with a closed source system like this nobody can prove it either way. (Unless you actually happened to make a list of what it picks and analyze that but nobody seems to have done that either).
However another poster mentioned the black box experiment that supposedly has discovered our entangled global consciousness. iPods would be great for that! They could be another proof for mind over matter, and if you wirelessly network iPods given to a bunch of people around the world, you might get more volunteers to help precog earthquakes, missile strikes, etc.
The only problem is that marketing and secrecy are still causing way too much noise for us to be able yet to hear our heartbeats so to say, and tune in to the global mind.
iTunes keeps track of the number of times a song has been played. I put all of my music in, and haven't added a whole lot over the last several months. I usually play a random shuffle of about 250 songs while I work.
You would expect, from a truly random sampling, that all songs would be played roughly the same number of times. However, the play counters in iTunes show that some songs are played 2-3x as often as others.
There does not appear to be any sort of favoritism with respect to particular bands or albums. There does appear to be some correlation with the "my ratings" column. Most of the less-frequently played songs are at 3 stars, while most of the more-frequently played songs are at 4 stars.
(My random shuffle is an automatic playlist generated from all songs rated at 3 stars or higher.)
I suspected that there were other kinds of bias in the iTunes random shuffle, but I convinced myself otherwise. For example, if I started to play a song by Rush, and then tapped the "shuffle" button twice to re-shuffle, it always seemed to have another Rush song in the top 5. After a couple of quick experiments I convinced myself that I just had a lucky streak.
Is it a mathematician without the degree? Or perhaps some type of numerogoly wizard?
- sigs are for wimps.
Winamp's randomizer has always done this for me too. The answer should be in the header of each file though, unless your player is biased toward a certain string of bits. 10010101>01101010. YUM!
At DRMing time they could throw a extra value of say a 1-10 weight into it. Reader sees the extra value and promotes said file to a higher priority. Hello new payola.
Maybe that's the same magic smoke that the Shuffle parts use in order to work, and maybe it's similar to the smoke that the parts of my last motherboard had, but it escaped once, so that motherboard doesn't work right now because the lack of that smoke.
Because my Sony CD Jukebox also seems to select songs to play by its billboard popularity.
With the president of Sony being at the last keynote...
Hmmm...
The clustering illusion demonstrates the universal principle of entropy. No matter what random collection of objects you look at, the most common distribution will not be the most evenly dispersed one, but the most evenly dispersed disorganized one.
If we randomly distribute the integers from 1 to 100, it's going to look pretty strange if the first 50 are all even, and the second 50 are all odd. So will a distribution where all the primes come first, or any other detectable order. Entropy acts as a sort of 'inflating' principle which makes less-smoothly-dispersed collections more common than more-smoothly-dispersed ones.
Clustering is inevitable, especially when you take small selections from a large, frequently reshuffled population. In this case, we're talking about a situation equivalent to shuffling the numbers from 1 to 100 and taking the first 10, then shuffling again and taking another 10, etc. If we were to do that 10 times, and found that we'd collected exactly one copy of every value from 1 to 100, that would be very unlikely. If it happened consistently, something would be very wrong with our RNG.
Back in the days of Winamp 2.91 there was a bug that meant when Winamp started on shuffle it would always play the first track. I think the behaviour of shuffle was also tweaked by the legendary Ryan Geess to not repeat a track until 50% of the other songs had been played. This may well have had statistical side effects.
of the time I was trying to explain probability to my second grade daughter. I took a coin out and flipped it five times in a row and it came up heads every time.
"Excuse, me," I said, "I'm going out to buy a lottery ticket."
Post may contain irony: discontinue use if experiencing mood swings, nausea or elevated blood pressure.
Hm, clever idea.
Besides just the player understanding it, it would be nice for it to track it and upload when you connect.
An open source extension api for both the iPod & iTunes would be sweet, Apple.
iTunes' (and the iPod as well) shuffle function isn't just random. It seems to weigh its choices on play count. If I rip an album, it won't come up on shuffle for a while. Once I start playing songs off of it, it'll start coming up all the time. I had a Bad Religion kick a few months ago, and it seems to play Bad Religion every 4th or 5th song now (in a library of 6500).
I mod down pyramid schemes in sigs.
In Soviet Russia, the iPod shuffles YOU.
--Z
I almost hate to mention it, but Windows Media Player has several "smart" playlists that can do just this. I think that one of them is "Songs I listen to at night" and another one is "Songs I like but haven't listened to lately".
Do I have to hand in my geek membership card for admitting that I use WMP?
-Montag
So why do you typically get three Wilco songs in an hour while Aretha Franklin waits in the wings forever? Yep, mine plays "Less Than You Think" over and over.
But, I wanted socialized health insurance!
My shuffle will find this song and play it every 10th song or so. I will have the thing FULL of music, and for some reason it wants to play Karma Police by Radiohead more than any other song I have ever loaded onto it. Anyone else have this issue? And I know I may be just being a human and seeing a pattern where there is not one, but keep this in mind, nothing is truly random. No Algorithm is completely and truly random.
My Dension DMP3 player seems to play the same few songs (from hundreds of songs) each time I random-play a mix directory.
It's the rise of the machines I tells ya! They now have taste in music.
That's the thing about finite state machines. If they're ever in the same state, they'll produce the same results. Hence any software PRNG can and will repeat a given series, given enough time. The trick is to have enough bits of state that it doesn't enter the same state. If you have 32 bits of state, you'll get at most 2^32 unique results before repeating. If you have 1024 bits of state, you'll get at most 2^1024 unique results before repeating. Of course, it could be much less with a poor PRNG.
I've had my iPod shuffle for a couple of weeks now and have not heard any Steely Dan yet- Thank God!
Apple's decoder chip supports drm as well. seee r.htm
http://www.sigmatel.com/products/audio-decod
The Gmini seemed like an OK device to me, but from playing around with it at the store, I didn't like the interface for navigating it. It seemed to be a lot less intuitive and require a lot more clicking and work. I still prefer the iPod over most of the other HD devices just because I much prefer it's physical hardware interface and it's actual on-screen display and appearance. The size of the Gmini is impressive though. But I don't think Apple's devices are overpriced......... there are cheaper devices out there, but I haven't generally found them to be as well designed or engineered.
It wouldn't be a real random pick if it didn't sometimes bring two songs side by side. The problem is most people don't want a truely random shuffle.
Geez...whatever happened to your sense of humor, mods?
i always thought winamp was the worst as far as shuffling. it seemed to always play the same 15 songs out of a list of 3000. i always took it was crappy shuffling. anyway, there is a skip foward button, and i tend to use it a lot when i listen to shuffled music.
Surprising as it may seem, it is not easy to tell if "a run" -- sequence of outcomes -- is random or not, even if the sequence is numerical or easy to code with numbers. What are you going to compare it with ? There are different notions of randomness, and sometimes they conflict with each other. While random events are totally unrelated and unpredictable, collectively they make patterns, and the patterns differ. For example, the heights of students in a classroom -- where the students are chosen for grades, not height -- will follow the famous bell-shaped curve. The expected lives of wooden utility poles do not. They follow a Poisson curve. And the distribution of per capita incomes in the upper reaches of an economy follows a log-normal curve. And so on. To complicate matters further, actual runs observed in real life may fool us. Random processes -- like roulette wheels -- can occasionally spew out a sequence of outcomes which looks non-random, such as five sixes or seven reds in a row. Some random processes generate patterns called Markov chains or random walks. They will spend a great deal of time away from their expected values, as in winning streaks and losing streaks in games of chance. So your cumulative winnings [or losings] at roulette will very seldom be zero for a particular outcome in long run and may be a very large number ! Some deterministic processes -- such as pseudo-random number genetators -- are designed to generated sequences which always test random, by one or more tests, although sometimes they may fail one ! Moreover, because they are deterministic, all PRNG's have some hidden structure and will eventually repeat themselves, if the run is long enough. But they will not put out six fives in a row ! Other deterministic processes of the erratic kind [chaotic] -- such as hurricanes [cyclones] spend most of their time wandering randomly and only a little time chaotically. But not the equivalent kind of storm in the Aleutian Islands. They spend most of their time in a chaotic mode ! End.
Thanks much!
You might want to take a look at IMMS
If anything, hearing the same artists over and over again would get me fed up and cause me to delete their songs, or at least not get any more in future.
-- I say we dust off and nuke the whole site from orbit. It's the only way to be sure. - The Space Marine played by Micheal Biehn from Aliens
Cool post relating to the quote.
My Journal.
That when I allow iTunes to select my playlist to be dropped into my shuffle for the workday ahead, it tends to favor songs that I've either legitimately purchased, have rated well (or at all), or have played most often from within iTunes (the ratings thing being a feature, IIRC).
Now, when I put my own playlist into the shuffle (for those days when nothing less than Bane will get me through it all), when in shuffle mode, it skews towards the rated songs and then the shorter named songs. I've basically just gotten to the point of removing my ratings from all songs and then selecting my playlist song by song (250 doesn't take that long, anyway), and trying to stay away from full albums.
"How like you to drag your keyboard to a gun fight." - Aaron Bedard (BANE)
I've been using an iPod for over a year now, and conferring often with many friends who use iPods. We've all come to the conclusion that the iPod shuffle is not entirely random, that it most definately "learns" what songs you prefer: it picks tracks for shuffle based on playcount and last played. iTunes does the same thing, and the data that is used can be transferred from iTunes to an iPod and vice versa (when syncing is on).
All circuits busy.
You mean Life is not Random?