asciirock writes "RJ, a University of Southampton grad student in the UK has just put his final year project online. Audioscrobbler is a free plug-in for Linux XMMS and Windows Winamp2. It tracks every tune you play, cross-references with others in the Audioscrobbler community and serves up recommendations. There's also msging, stats and user homepages. In other words... Firefly lives!"
by Anonymous Coward
on Sunday February 16 2003, @10:08AM (#5313815)
Listen to *one* Britney Spears track out of curiosity, get distracted by something to do in another room, forget that it's playing repeatedly for 3 hours in the meanwhile, get labelled a teen music sheep by the system and get recommendations for more degrading music. Arg!
Listen to *one* Britney Spears track out of curiosity...[]
What in the hell are you doing listening to Britney in the first place? It's like saying "I found this virus and decided to infect few of my computers.. you know, out of curiousity"
You might be tempted to say, "Oops, I did it again."
Ok, you can slap me now.
I must say though, my friends have always been more reliable at alerting me to things I might like than any program, and they're more, well, friendly too. I've never yet seen any automated system that didn't, sooner or later, get messed up. Which would be ok, except I've never known one to make particularly valuable recommendations when it *wasn't* messed up either.
Hey, you watched a car racing movie. Here's some other car racing movies you might like to watch.
Well, I *know* that you moron. I'm into car racing movies. Get it?
Recommendations are nice, but what I want is a tie in to Fast Track. I want a list of DATs that I can plug in to Kazaalite and download based on what I play.
i want to steer clear of filesharing to avoid getting savaged by the RIAA etc.. but i'm planning on exposing the data so other people can write "unofficial" addons to do stuff like this, which hopefully won't get me in trouble:)
Great.... Firefly.com has a patent for this sort of thing, and now Microsoft has it (Microsoft bought them). Is this another case of something getting off the ground and then squashed because of lawyers?
Software patents are not (currently) valid in the UK.
Correct. They are currently valid nowhere in Europe, although there are unfortunately plans to change that.
However, this doesn't prevent an unethical company [teles.de] to sneak just such a patent past a sleepy patent office clerk, and once it's on the books, they can bully whoever they want [www.dtag.de] with it. True enough, eventually the judge will decide in favor of the defendant, but until that date the defendant has to cope with a number of hassles (lawyer's costs, and more importantly: injunction to force him to change his software, so as not to use the disputed features, etc.). Just let's hope nobody patents the light switch!
Argh, why's there no support for Winamp 3? Now i have to choose between having my Media Library, or goin old-skool, but bein able to use this cool plug-in
It's probably because Winamp 3 is a bloated, nasty mess. I've tried Winamp 3 many times, and I always remove it to go back to Winamp 2. Most software reviewers tend to agree. Winamp 2 is pretty close to the perfect MP3 player.
Yeah, shame they don't add just the lightweight cross-fader to winamp2. I know there's plugins that do this for winamp2, but the ones I tried tend to make winamp2 suddenly more bloated and nasty (cause they have to work by buffering the next song, so you don't get the ability to switch between songs quickly and skip ahead easily anymore because there's a five-second delay - and if you decrease the delay the fades aren't so nice anymore).
Winamp3 is an attempt at completely rewriting Winamp so it's cross-platform (there is a linux version). Also, the component/plugin architecture is much better - Winamp3 has MUCH more development potential. However, it was released much too early by management (even the developers agree with this). Version 3.1 should be much improved. Perhaps calling it version 3 is a misnomer - Winamp 2 is still very much under active development, and a new version will be out soon.
If Winamp 2 is perfect for you, why is there any reason to try anything different? (heheh, there will be soon;))
Winamp3 is an attempt at completely rewriting Winamp so it's cross-platform (there is a linux version).
Ahem. There was a "Linux Alpha" version of winamp3 released almost two years ago now, which was completely and totally unusable, as was the alleged MacOS version. There has not been a linux release since, and it is not even currently obvious how to download the linux version. (Understandable, since it was not in any way useful.)
If you visited the site you would know that a WinAmp3 version is in progress. Besides, WinAmp 2 is a near perfect MP3 player, WinAmp 3 sucks the llama's ass.
While it might seem cool at the first sight to produce such a tool which creates recommandations from other people's playlists, it's in fact counterproductive at pratical applications.
What we have here is a stabilizing feedback loop,
songs often heard will be heard more often. This can be described by the following simple equation (h(t) - hear rate):
dh(t)/dt = h(t) * c + sin(h(t)) * phi(dt,H(t))
where the last term is a stochastic diffusion corrector which models connection drops etc.
This means that after a 3c/pi annealing time new injected songs (c1,...,ck) have no chance to be heard at all, because the system reenforces to old songs. The only possibility to get something new into the playlists, is to get an external stimulation at e.g. t0: phi(c-h(t0). Such a high current can be only injected be a very strong source covering a large part of the system.
In simple words: after some iterations an equlibrium is reached and all new song turning up in the recommendations are the top 24 played at MTV.
In fact, you are just replaying the shitty MTV mainstream taste.
I can't think that this is very good, first you don't need a computer program to recommend the MTV top 30 when you have a TV and secondly you only get boring mainstream stuff and nothing like exciting french chansons or so.
Someone finally proved mathmatically why clear channel is wrong! Quick, send this to the FCC, oh wait, they wont be able to understand it.
I think, however, that a correction factor is missing, you should have a phi(c - (p*h(t0)) where p is a correction factor for how open a person is to new music. I know for me that it doesnt take much for me to try something new, but for many people over the age of about 18, they know what they like, and arent going to progress with the change in popular music. This is why old people like oldies.
I have absolutely no clue whether those equations are correct (I don't even want to have a clue, really), but the tendency you describe certainly would exist. However, since people most likely won't base their hearing habits solely on such a database, I doubt equilibium will ever be reached - this is basically saying the "external stimulation" you refer to will always be quite strong. Nevertheless, it's certainly correct that such a system makes is unlikely for rare songs to be put into the main circulation - of course, since the developers are probably aware of this, they can counteract, for instance by adding modifiers for songs newly introduced into the system, or by allowing for user-moderated boosts to certain songs.
There is no real reason why that would be a feedback loop: You're listening to a set A of songs. So, you recieve a reccomendation from someone listening partly to A, partly to another set (all the songs he heard which are not in A), B. You exercise your own taste (which is not included in your text at all), and integrate part of B. (You might also give up a few over-played songs of A) Now you have new recommendations...
There's absolutely no reason why this should gravitate towards the MTV play list: it'll gravitate towards "music you like and music people who like that, likes"
I'm also not sure where that equation comes from. There's absolutely nothing which allows you to derive math from the situation. A person recieves a recommendation, and may choose to take it or not. He may listen to part of the song, decide to remove it, and the program will disregard that song.
You cannot write an equation to tell what that person is going to do...
"I'm also not sure where that equation comes from. There's absolutely nothing which allows you to derive math from the situation... You cannot write an equation to tell what that person is going to do..."
Hence the stochastic part of the equation--it is kind of a "fudge factor" to take guesses.
What the original poster confused was ranking and searching. A tool that ranks the songs and plays those more frequently that you play more frequently can be bad, depending on implementation, and cause you to continuously play through the same playlist of < 20 songs. This is particularly true of the self-reinforcing designs where it counts when it plays it as well as when you select it.
Mathematics can also be used to tell what *people* will do, rather than any one individual, through the techniques of social modeling, group theory, and other methods.
That being said: I haven't checked his equation for correctness, though the formatting is standard.
I used to work for a software company that did this sort of thing. While you can't predict exactly what an individual is going to do from an equation, it always amazed me just how good a hit-rate you can get when you test against a larger group.
The mathematics behind it is actually fairly straightforward & there's a lot you can do to bias the results so the most commonly listened to tracks don't always appear at the top of the list of recommendations.
The harder part is analysing the data-set & computing the recommendations in a reasonable time frame. Particularly since the quality tends to increase with larger & richer data-sets.
Is this phenomenon true for other suggestion-based systems, such as IMDB, Amazon, et al?
I'm always skeptical of suggestion based systems that make simple inferences (eg, if you like X you'll like Y) because they never suggest anything I like.
But I can buy into *complex* suggestion based systems that do a more in-depth job of matching preferences. For example, knowing that I like Tangering Dream, a simple system may suggest Brian Eno or Kraftwerk. But my personal playlist may go from Tangerine Dream, to the Replacements, to Miles Davis, to Richard Thompson, to the Velvet Underground.
Someone who also listens to those same artists might also have suggestions that appeal to me since it better reflects the complexity of my taste versus simple comparisons.
There are easy ways around this: you can age songs (guaranteeing that new songs get priority) or you can use web-of-trust ratings to see what high-trusted peers recommend currently.
BTW, I know this as my newsbot [memigo.com] does the same sortof filtering for news articles, and although there is some feedback reinforcement for recent news, it does work suprisingly well.
If I understand correctly, this system strikes me as somewhat similar to how google ranks pages. the google system obviously works... I have a feeling this will work too.
the feedback only breaks things down if users limit their selections to received recommendations. since many people continually update their collection, we have enough input to avoid "the one giant recommended playlist." most people search out new music.
So, in words, you're claiming, that the change of the hear rate (h) is mainly proportional to the amount one currently listens to the song?
Don't know about you. But for me, listening to a song usually peaks after a certain time, and then declines.
So, more something along:
h(t) = c * e^(k*w) * cos(w*t) ; k > 2*w
The problem with MTV is, that they're targeting the largest audience (hence mainstream). A large audience is less flexible in accepting new music.
The program (without reading it, due to/.ing) seam to cluster the usership based on their preferences, thus creating smaller communities, which are more suspectible to new influences.
I'm getting this plug-in and I'm going to test it because it sounds great, but it sounds creapishly like certain other pieces of software and licensing clauses.
Think about it; it profiles your music taste and make recommandations. That's what spyware does (or says it does.)
I don't doubt that this piece of software is completely innocent (it being made by a student,) but who knows when someone makes a "new and improved Audioscrobbler." That really profiles you and stores this information for resale and profit without you really knowing it. Sure you might prefere targeted music adwertising, but be warned such advertising would only come from a preselected, narrow artist pool.
Now, I'm using Audioscrobbler, but if it ever becomes mainstream I would be careful using any commercial equivalent (or even a commercial Audioscrobbler.)
Think about it; it profiles your music taste and make recommandations. That's what spyware does (or says it does.)
I don't know about anyone else, but I can't see how a system could possibly make intelligent recommendations without profiling me. If I happen to like listening to (say) Britney Spears, Metallica, and Herbie Hancock, I'd like to see what other people who also do the same are listening to. I DON'T want to know that people who listen to Britney Spears is likely to also listen to N'Sync.
To me, the value-added here is precisely in the profiling.
Everything that send out your info to the net isn't "spyware". That's a ridiculous term. If it's useful, then it's certainly not "spyware". Would you call SETI@Home spyware? How about email?
What I really don't get, and sounds a bit fishy to me, is the whole username/profiling thing.
I mean, this can be done without it: Have an anonymous user handle on that site. No email, no nothing. (sure, they can have your IP. They can have mine, too, if they want, it's a dynamic one)
Whenever you hear a song, it sends the info: a user who heard (set of songs) decided to hear (new song), and of course heuristics of how much any song is heard, bla bla bla.
The server keeps this huge database. When you want recommendations (downloaded every 15 min? or something) your program asks what the database recommends for someone who listened to (the set of songs you listened to). You're not giving away an email, no personal info, just an anonymous username (created automatically, or something. There is alot of 'or something's here)
There's no real reason for the server to know who you are or what you like for this to work. Perfect profiling is also not nessecary, in my view, but that's a different issue altogether
by Anonymous Coward
on Sunday February 16 2003, @10:25AM (#5313873)
MoodLogic [moodlogic.net] provides this service, at a cost. It works based on paid subscriptions, or submitted data, as tokens. It only does mp3/wma (despite some of us asking for ogg support), but it seems to work pretty well. I believe it's currently only for Windows, and has a WinAmp plugin, but I may be wrong.
For just a second, just one second there I thought this was talking about the Firefly TV show and somehow it was being brought back. sigh time to wander about aimlessly again.....
RJ is already having hosting trouble, slashdotting him is not going to help audioscrobbler.
At least it happened before it moved to the new hosting service. Now only his university is going to be pissed, and he has a server to move to in the near future.
I think that associations, in computing, is a great idea for user interface.
A program like this (lets disregard the Big Brother for one second, and look at computer+user alone) tells you what songs it thinks you'll like, based on what you've heard before.
It could also tell you what songs you'd like to hear NEXT, based on order of songs you had before, and make these easier to access on the playlist (like, on the recommendation list. I'm getting out of hand aren't I?)
The whole idea of associating user actions can be great. Suppose you work on a project. Slowly, the computer (the brand-new GPLed Associator program) associates a certain directory, where all the files are, with the files themselves, your favorite editor, the compiler for that language, and certain sites you visited researching for it. via some UI, it'll make all these accessible when 'triggered' - when it is pretty sure you're working on the project right now, or going to.
In some sense (in a small amount of cases), the computer will be 'one step ahead of you' - holding the line when you're just about to ask it to call...
"All the source code will appear on the site soon (GPL)."
Why not today?
Don't be embarrassed about it being crappy code, all code is crappy in the early stages.:) Put the GPL LICENSE file in the root directory, and follow its directions for adding notification to your source files. Then tar it up and call it 0.01.
Put it up and keep putting it up as you update it. If you think you might have security issues, best that you open the code now before your user base gets any bigger -- let people review it and send you suggestions. If you don't think you have security issues, you have no reason not to release it.
For a project that demands community participation, a promise of GPL code in the future is worthless. What's valuable is the code itself.
Licenses, releases, security feedback, other feedback... this is all part of doing a project like this. It's something that isn't normally taught in a university, but if you really want to run a project that depends on its community, this is not extra-credit, this is a prerequisite.
I can't find any licensing anywhere that tells me the terms under which their collected information will be used.
As far as I can tell from prowling over the site's FAQs and other documents, the student who put this together might collect a ton of data about your personal listening habits for a year and then (A) get bored with it and shut the project down without releasing that data back to the community who might want to actually keep the recommendation-system running, or (B) sell it all to marketers who promptly turn it into a paid service.
We've learned from CDDB what happens when users volunteer to build something that isn't Free: if it becomes popular enough to do any good, someone will buy it and shut out the very people who built it.
The creator has a good idea but needs to think it through before he'll get my participation.
I've been talking with robert from musicbrainz (audioscrobbler will be using TRM technology soon) and ben from agentarts. I'm gonna be using some agentarts data, and i will make all my data available freely when i've implemented the TRM system to sort out badly named songs.
i also want to syndicate the data (xml/rss) so ppl can stick live info on their blogs/websites.. this wont happen till i move servers tho.
i'm not gonna run off and give the data to the riaa or start emailing you crappy adverts. its a uni project that's about half way thru. the project will run and run tho- i'm not gonna shut it down.
Should i ever get border of it (unlikely) there are plenty of ppl that will take over. i'll just slap it on sourceforge.
Forgive me, but I've never used the system, just reading slashdot:). Anyway, what I was wondering, does it only collect the titles, or do you have some way of rating them too?
Otherwise it seems to me there could easily become a self-feeding loop, song gets recommended to people, they play it, and even if they don't like it, it'll get recommended to other people with similar tastes.
Having some kind of rating system also makes a alot more sense when recommending, otherwise you might be recommended the same music (that you dont like) time and time again. Loves Britney Spears, hates Nsync is a lot more info than just Loves Britney Spears.
FireFly was a rating, preference-matching, and suggestion system developed at the MIT Media Lab long before anyone had really heard of Joss Whedon.:-)
There were a couple research versions of the multidimensional matching system run out of the Media Lab (one for music, then an expanded one for music, movies, and books, as I recall). FireFly was the name used for the spinoff company. It went through a brief period of excitment during the internet boom, then (iirc) was purchased by some large corporation or other. (I have a friend who worked on the research project.)
What's wrong with the Winamp/XMMS interface? It's fairly intuitive and simple. XMMS itself is reliable and well-written to boot. Even on my ancient K6 mailserver/jukebox, it uses almost no CPU and almost no memory.
Forget all the bogus attempts to move your antique business model online.
They're bogus for a reason.
What you want to do is license this technology from its creators, and build a mechanism to sell digital copies of the recommended tunes
No, it most certainly isn't what they want to do.
The RIAA represents the recording industry, not the music industry. Their entire existance relies on tying music to physical objects. Doing as you suggest would be simply hastening their own demise.
A recording session used to cost huge $$$ - but due to advances in technology, now costs relatively litte (it's possible to build your own recording studio for a few thousand dollars.) The internet has started a similar revolution with regards to distribution.
The RIAA knows this, and they know that it dooms them - with cheap recording and distribution, the artists no longer need them (and their lop-sided contracts). The problem is that the artists don't know this yet.
The recoding industry's whole "this is theft" mantra is basically a smoke screen to prevent artists from finding out they have an alternative to being a slave to a record label.
It tracks every tune you play? (Score:5, Interesting)
Re:It tracks every tune you play? (Score:5, Funny)
What in the hell are you doing listening to Britney in the first place? It's like saying "I found this virus and decided to infect few of my computers.. you know, out of curiousity"
Parent
What's even worse, if you do it a second time. . . (Score:2)
Ok, you can slap me now.
I must say though, my friends have always been more reliable at alerting me to things I might like than any program, and they're more, well, friendly too. I've never yet seen any automated system that didn't, sooner or later, get messed up. Which would be ok, except I've never known one to make particularly valuable recommendations when it *wasn't* messed up either.
Hey, you watched a car racing movie. Here's some other car racing movies you might like to watch.
Well, I *know* that you moron. I'm into car racing movies. Get it?
KFG
Go further! (Score:5, Interesting)
Re:Go further! (Score:5, Informative)
i want to steer clear of filesharing to avoid getting savaged by the RIAA etc.. but i'm planning on exposing the data so other people can write "unofficial" addons to do stuff like this, which hopefully won't get me in trouble
RJ
Parent
Nice user profiling (Score:3, Interesting)
oh swell.... (Score:5, Informative)
Eech.
Re:oh swell.... (Score:4, Informative)
Parent
Re:oh swell.... (Score:3, Interesting)
Correct. They are currently valid nowhere in Europe, although there are unfortunately plans to change that.
However, this doesn't prevent an unethical company [teles.de] to sneak just such a patent past a sleepy patent office clerk, and once it's on the books, they can bully whoever they want [www.dtag.de] with it. True enough, eventually the judge will decide in favor of the defendant, but until that date the defendant has to cope with a number of hassles (lawyer's costs, and more importantly: injunction to force him to change his software, so as not to use the disputed features, etc.). Just let's hope nobody patents the light switch!
Winamp 3? (Score:2, Interesting)
Re:Winamp 3? (Score:5, Insightful)
Parent
Re:Winamp 3? (Score:2)
Daniel
Re:Winamp 3? (Score:4, Informative)
If Winamp 2 is perfect for you, why is there any reason to try anything different? (heheh, there will be soon
Yes I am a Winamp3 apologist
Parent
Cross-platform my ass. (Score:3, Informative)
Ahem. There was a "Linux Alpha" version of winamp3 released almost two years ago now, which was completely and totally unusable, as was the alleged MacOS version. There has not been a linux release since, and it is not even currently obvious how to download the linux version. (Understandable, since it was not in any way useful.)
Re:Winamp 3? (Score:2)
Group think, bad taste and braindamage. (Score:3, Interesting)
What we have here is a stabilizing feedback loop, songs often heard will be heard more often. This can be described by the following simple equation (h(t) - hear rate):
dh(t)/dt = h(t) * c + sin(h(t)) * phi(dt,H(t)) where the last term is a stochastic diffusion corrector which models connection drops etc. This means that after a 3c/pi annealing time new injected songs (c1,...,ck) have no chance to be heard at all, because the system reenforces to old songs. The only possibility to get something new into the playlists, is to get an external stimulation at e.g. t0: phi(c-h(t0). Such a high current can be only injected be a very strong source covering a large part of the system.
In simple words: after some iterations an equlibrium is reached and all new song turning up in the recommendations are the top 24 played at MTV.
In fact, you are just replaying the shitty MTV mainstream taste.
I can't think that this is very good, first you don't need a computer program to recommend the MTV top 30 when you have a TV and secondly you only get boring mainstream stuff and nothing like exciting french chansons or so.
Re:Group think, bad taste and braindamage. (Score:4, Funny)
I think, however, that a correction factor is missing, you should have a phi(c - (p*h(t0)) where p is a correction factor for how open a person is to new music. I know for me that it doesnt take much for me to try something new, but for many people over the age of about 18, they know what they like, and arent going to progress with the change in popular music. This is why old people like oldies.
Parent
Re:Group think, bad taste and braindamage. (Score:2)
Nevertheless, it's certainly correct that such a system makes is unlikely for rare songs to be put into the main circulation - of course, since the developers are probably aware of this, they can counteract, for instance by adding modifiers for songs newly introduced into the system, or by allowing for user-moderated boosts to certain songs.
Re:Group think, bad taste and braindamage. (Score:5, Interesting)
You're listening to a set A of songs. So, you recieve a reccomendation from someone listening partly to A, partly to another set (all the songs he heard which are not in A), B.
You exercise your own taste (which is not included in your text at all), and integrate part of B. (You might also give up a few over-played songs of A)
Now you have new recommendations...
There's absolutely no reason why this should gravitate towards the MTV play list: it'll gravitate towards "music you like and music people who like that, likes"
I'm also not sure where that equation comes from. There's absolutely nothing which allows you to derive math from the situation.
A person recieves a recommendation, and may choose to take it or not. He may listen to part of the song, decide to remove it, and the program will disregard that song.
You cannot write an equation to tell what that person is going to do...
Parent
Re: Math (slightly OT) (Score:4, Informative)
Hence the stochastic part of the equation--it is kind of a "fudge factor" to take guesses.
What the original poster confused was ranking and searching. A tool that ranks the songs and plays those more frequently that you play more frequently can be bad, depending on implementation, and cause you to continuously play through the same playlist of < 20 songs. This is particularly true of the self-reinforcing designs where it counts when it plays it as well as when you select it.
Mathematics can also be used to tell what *people* will do, rather than any one individual, through the techniques of social modeling, group theory, and other methods.
That being said: I haven't checked his equation for correctness, though the formatting is standard.
Parent
Re:Group think, bad taste and braindamage. (Score:3, Interesting)
The mathematics behind it is actually fairly straightforward & there's a lot you can do to bias the results so the most commonly listened to tracks don't always appear at the top of the list of recommendations.
The harder part is analysing the data-set & computing the recommendations in a reasonable time frame. Particularly since the quality tends to increase with larger & richer data-sets.
Re:Group think, bad taste and braindamage. (Score:3, Interesting)
I'm always skeptical of suggestion based systems that make simple inferences (eg, if you like X you'll like Y) because they never suggest anything I like.
But I can buy into *complex* suggestion based systems that do a more in-depth job of matching preferences. For example, knowing that I like Tangering Dream, a simple system may suggest Brian Eno or Kraftwerk. But my personal playlist may go from Tangerine Dream, to the Replacements, to Miles Davis, to Richard Thompson, to the Velvet Underground.
Someone who also listens to those same artists might also have suggestions that appeal to me since it better reflects the complexity of my taste versus simple comparisons.
Re:Group think, bad taste and braindamage. (Score:4, Informative)
BTW, I know this as my newsbot [memigo.com] does the same sortof filtering for news articles, and although there is some feedback reinforcement for recent news, it does work suprisingly well.
Parent
Re:Group think, bad taste and braindamage. (Score:3, Interesting)
the feedback only breaks things down if users limit their selections to received recommendations. since many people continually update their collection, we have enough input to avoid "the one giant recommended playlist." most people search out new music.
Re:Group think, bad taste and braindamage. (Score:3, Interesting)
Don't know about you. But for me, listening to a song usually peaks after a certain time, and then declines.
So, more something along:The problem with MTV is, that they're targeting the largest audience (hence mainstream). A large audience is less flexible in accepting new music.
The program (without reading it, due to
But ... (Score:5, Insightful)
Think about it; it profiles your music taste and make recommandations. That's what spyware does (or says it does.)
I don't doubt that this piece of software is completely innocent (it being made by a student,) but who knows when someone makes a "new and improved Audioscrobbler." That really profiles you and stores this information for resale and profit without you really knowing it. Sure you might prefere targeted music adwertising, but be warned such advertising would only come from a preselected, narrow artist pool.
Now, I'm using Audioscrobbler, but if it ever becomes mainstream I would be careful using any commercial equivalent (or even a commercial Audioscrobbler.)
Re:But ... (Score:5, Insightful)
I don't know about anyone else, but I can't see how a system could possibly make intelligent recommendations without profiling me. If I happen to like listening to (say) Britney Spears, Metallica, and Herbie Hancock, I'd like to see what other people who also do the same are listening to. I DON'T want to know that people who listen to Britney Spears is likely to also listen to N'Sync.
To me, the value-added here is precisely in the profiling.
Parent
Re:But ... (Score:4, Insightful)
Parent
Re:But ... (Score:4, Interesting)
I mean, this can be done without it:
Have an anonymous user handle on that site. No email, no nothing. (sure, they can have your IP. They can have mine, too, if they want, it's a dynamic one)
Whenever you hear a song, it sends the info: a user who heard (set of songs) decided to hear (new song), and of course heuristics of how much any song is heard, bla bla bla.
The server keeps this huge database. When you want recommendations (downloaded every 15 min? or something) your program asks what the database recommends for someone who listened to (the set of songs you listened to). You're not giving away an email, no personal info, just an anonymous username (created automatically, or something. There is alot of 'or something's here)
There's no real reason for the server to know who you are or what you like for this to work.
Perfect profiling is also not nessecary, in my view, but that's a different issue altogether
Parent
MoodLogic anyone? (Score:3, Informative)
Damit! (Score:2, Funny)
Ideas. (Score:2, Interesting)
And also maybe keep track of the amount of times i play it..
That way it could find songs that i like, and i could have a category by which to order my songs when i can't decide what to listen to..
Re:Ideas. (Score:5, Informative)
And with smart playlists, you can have 'most heard', 'never heard', top or bottom rated....
Easy!
Mark
Parent
Why submit this to ./? (Score:2, Informative)
At least it happened before it moved to the new hosting service. Now only his university is going to be pissed, and he has a server to move to in the near future.
Beyond the music realm.. (Score:4, Interesting)
A program like this (lets disregard the Big Brother for one second, and look at computer+user alone) tells you what songs it thinks you'll like, based on what you've heard before.
It could also tell you what songs you'd like to hear NEXT, based on order of songs you had before, and make these easier to access on the playlist (like, on the recommendation list. I'm getting out of hand aren't I?)
The whole idea of associating user actions can be great. Suppose you work on a project. Slowly, the computer (the brand-new GPLed Associator program) associates a certain directory, where all the files are, with the files themselves, your favorite editor, the compiler for that language, and certain sites you visited researching for it.
via some UI, it'll make all these accessible when 'triggered' - when it is pretty sure you're working on the project right now, or going to.
In some sense (in a small amount of cases), the computer will be 'one step ahead of you' - holding the line when you're just about to ask it to call...
I don't get it.. (Score:2)
"remember firefly" (Score:2)
I build Audioscrobbler (Score:5, Informative)
I was kinda hoping NOT to get slashdotted for a few days - i'm moving to a bigger better server soon. :P
The site is currently hosted by my Uni, no wonder the webserver ground to a halt..Oh well at least i dont have to pay for the bandwidth used at uni :)
The site's gonna be pretty slow for a few days, but please bookmark it and revisit soon- should have much more bandwidth and a faster server..
I could do with some help developing the XMMS plugin and the winamp 3 plugin. All the source code will appear on the site soon (GPL).
RJ
Re:I build Audioscrobbler (Score:5, Interesting)
Why not today?
Don't be embarrassed about it being crappy code, all code is crappy in the early stages. :) Put the GPL LICENSE file in the root directory, and follow its directions for adding notification to your source files. Then tar it up and call it 0.01.
Put it up and keep putting it up as you update it. If you think you might have security issues, best that you open the code now before your user base gets any bigger -- let people review it and send you suggestions. If you don't think you have security issues, you have no reason not to release it.
For a project that demands community participation, a promise of GPL code in the future is worthless. What's valuable is the code itself.
Licenses, releases, security feedback, other feedback... this is all part of doing a project like this. It's something that isn't normally taught in a university, but if you really want to run a project that depends on its community, this is not extra-credit, this is a prerequisite.
Just my opinion :)
Parent
Re:I build Audioscrobbler (Score:5, Informative)
I'm not sure if i'm allowed to (university regulations) put the code up as GPL until i hand in the final project on May 8th this year.
I will find out tomorrow when i go to uni, and post an article on audioscrobbler.com explaining the status.
RJ
Parent
"Free" but apparently not Free (Score:5, Insightful)
As far as I can tell from prowling over the site's FAQs and other documents, the student who put this together might collect a ton of data about your personal listening habits for a year and then (A) get bored with it and shut the project down without releasing that data back to the community who might want to actually keep the recommendation-system running, or (B) sell it all to marketers who promptly turn it into a paid service.
We've learned from CDDB what happens when users volunteer to build something that isn't Free: if it becomes popular enough to do any good, someone will buy it and shut out the very people who built it.
The creator has a good idea but needs to think it through before he'll get my participation.
firefly (Score:5, Funny)
Seriously - I thought this was going to be about how the music you then upload helps you have adventures in Reaver territory.
Ok! Ok, I'm moving on...
All data to be made available (Score:5, Informative)
i also want to syndicate the data (xml/rss) so ppl can stick live info on their blogs/websites.. this wont happen till i move servers tho.
i'm not gonna run off and give the data to the riaa or start emailing you crappy adverts. its a uni project that's about half way thru. the project will run and run tho- i'm not gonna shut it down.
Should i ever get border of it (unlikely) there are plenty of ppl that will take over. i'll just slap it on sourceforge.
RJ
Now I can finish that poem (Score:4, Funny)
Hmm... cobbler
Can you rate what you listen to? (Score:3, Insightful)
Otherwise it seems to me there could easily become a self-feeding loop, song gets recommended to people, they play it, and even if they don't like it, it'll get recommended to other people with similar tastes.
Having some kind of rating system also makes a alot more sense when recommending, otherwise you might be recommended the same music (that you dont like) time and time again. Loves Britney Spears, hates Nsync is a lot more info than just Loves Britney Spears.
Kjella
Re:I don't get it (Score:5, Informative)
There were a couple research versions of the multidimensional matching system run out of the Media Lab (one for music, then an expanded one for music, movies, and books, as I recall). FireFly was the name used for the spinoff company. It went through a brief period of excitment during the internet boom, then (iirc) was purchased by some large corporation or other. (I have a friend who worked on the research project.)
Parent
Re:Next to be subpoenaed... RJ (Score:4, Interesting)
Don't worry, they won't. The RIAA has no power in the UK and none over the government.
Part of the fun of being British these days is the RIAA can't bribe - sorry, fund - polititions in Westminister nearly as easily as in Washington.
Parent
Re:Next to be subpoenaed... RJ (Score:3, Funny)
Re:linamp (Score:2)
Nope, will never fly.. (Score:3, Insightful)
They're bogus for a reason.
What you want to do is license this technology from its creators, and build a mechanism to sell digital copies of the recommended tunes
No, it most certainly isn't what they want to do.
The RIAA represents the recording industry, not the music industry. Their entire existance relies on tying music to physical objects. Doing as you suggest would be simply hastening their own demise.
A recording session used to cost huge $$$ - but due to advances in technology, now costs relatively litte (it's possible to build your own recording studio for a few thousand dollars.) The internet has started a similar revolution with regards to distribution.
The RIAA knows this, and they know that it dooms them - with cheap recording and distribution, the artists no longer need them (and their lop-sided contracts). The problem is that the artists don't know this yet.
The recoding industry's whole "this is theft" mantra is basically a smoke screen to prevent artists from finding out they have an alternative to being a slave to a record label.