Are Formats What Napster Really Needs?

Adam Curry writes: "This article has taken me a combined 20 years of broadcast and computer experience to compile and I couldn't be more excited about the possibilities the Internet can bring now that we have witnessed the cultural change from the traditional broadcast models to the Peer to Peer networking model technologies such as Napster and Gnutella have shown us." Whether or not it convinces you that Gnutella needs formats, this piece also offers a bit of knowledge-in-passing about the music biz that may interest readers putting together audio streams at home or for friends.

Re:Formats: Just what we -do- want

I think you missed the point of what he stated completely and entirely. He's not talking about being forcefed by predefined playlists or whatnot. This is not an advertising related thing he's talking about.

What he stated was that there should be more definable (less broad, more ways of mixing various ones) genres with subtags and what not, and a large database of them online. Then, as the amount of music climbs skyhigh from heavy usage on Napster and proceeds to fork itself into the database, it gets to the point where you can go to the database, state "I'd like to see stuff that's kind of very similar to this music" and it can cut it up across genre and msicellaneous information lines, which gives you.. More music by more bands you haven't heard of that are in a similar style! Imagine that.

IMO this could be a major improvement over Napster, who's main purpose right now is finding mp3s of bands you know. Browsing other people's repositories can sometimes help to learn about more bands, but isn't as effective as it could be. (Although searching for 'remix' is amusing for hours on end.)

Anything to increase the diversity of music is a good thing, and that is what he wants. It won't remove anything as it is, it's a secondary layer atop it that nets like styles together, mixable by however you choose.

Re:Formats: Just what we -don't- want

.
Mr. Curry claims that "you may hear Madonna one time and Jennifer Lopez the next", but what if I wanted to listen to Madonna and got Lopez instead? I want to choose the songs I listen to; I don't want some impersonal AI with no inkling of my emotions to try to decide what I'll like.

The concept is that you will hear Madonna one time and Janis Light the next song. Who is Janis Light? I have no idea, but the DJ is betting that if you like Madonna, you'll like Janis Light.

If anything, Napster has the flaw of making people download the same few artists over and over again. Rather than having a DJ tie in similar artists by "feel" (like Nick Cave, the Cruxshadows, and Black 47, three compeletely different genres, but many of the same people like all three), Napster simply says: "Enter the artist you are searching for". Unless you know the artist, it dosen't help.

Now, having *said* this, Napster *does* have formats... you can view the listed shared files, and (aside from the people who download everything for the hell of it), you can guess that if you like 80% of a person's collection, you might like at least some of the remaining 20%.

Does Gnutella do anything like this?

--
Evan

Re:I think Gnutella needs more organization

Closing the source is hardly the solution. The only way to prevent abuse is for the system to require a trusted cryptographic public key for access. Then the people who are initially in control of the system can start signing other keys, and building up a web of trust. They can be extremely liberal about key signing, because the trust can be revoked at any time, or the key can be shitlisted altogether.

It sounds draconian but it is the only way to have an orderly and useful file trading service. Think BBS. When you joined a BBS Back in The Day, you didn't just get full access to everything right away. You got perhaps a few minutes a week with limited downloads and no messaging. Then maybe if you where a good user, you would get chatting priviledges. And after that, perhaps file uploading priviledges. If you were a fuckup, your account was permanently revoked.

Regarding Gnutella's particular scalability problems, I'm afraid there may be no solution aside from a complete redesign.

Define a "Format"

The problem with putting music into formats/generes is subjectiveness. The categories any particular system has also helps determine what "format" a song belongs to.

For example, I could have a "vocal," "choral," "madrigal," "classical," "baroque," "religious," song, depending on what service I looked for information about it. Likewise, metal fans would be hard pressed to mark many albums or artists as "light rock" or "heavy metal." In the end, the individual reviewer is the sole judge of if a song matches a format. Often I find myself disagreeing with them.

Amazing

This article has taken me a combined 20 years of broadcast and computer experience to compile
I don't doubt it. I tried running it through gcc and got more error messages than I can count.
--Shoeboy

Re:Format "problems"

I did a stint as an intern and then a board op in the programming department of a fairly big radio station in these parts (consistantly #1-2 in the books) and learned a lot about radio from picking the hell out of the brain of a kickass PD during that time (damn, do I miss that gig :)). Basically, there are two magic numbers for radio stations: 1) how many people they can get to tag the station at all, and, more importantly, 2) how many people they can get to listen for at least 15 minutes. That 15 minute one is the big whammy, as that is what most of a book is based on. That's why they use teasers, contests, etc to nab listeners for just a few more minutes, because once a listener breaks the 15 minute mark, its a win for the station.

You wouldn't believe the insane amount of research that goes into each and every goddamn song that gets spins. There was this massive book in the PD's office that had 2-3 pages for damn near every single song in the archive that contained interest levels of every age/sex group for that song, so that they could determine what adds to throw into the playlist to boost a certain demographic. The amount of time and money devoted to testing music is enormous! And radio stations still use their own testing to supplement the millions of dollars worth of research already available. The station I worked at had a weekly listening session where they'd bring in a few people, stick some cans on them and play cuts of new and older music to get an idea of what people wanted to hear, what they are burnt out on, what they should reintroduce into the playlist, etc. Every night, they play a couple new tunes being considered for the playlist and solicit responses from the listeners. We even periodically kept track of all requests made throughout the day as another tool for determining interest.

While it may be true that the record companies push off certain pap on to the stations to get spins, radio stations are not tools of the RIAA, as they make their money off of advertising revenue based on the almighty ratings, so they can't afford to blindly play whatever gets thrown on their plate. You may think that radio sucks because they play the same crap every three to four hours, but that's because you only see it from our unique perspective of high volume music listeners. The majority of people have only a passing interest in listening to radio, such as on the way to and from work or at lunchtime; hence the reason that the magic number of minutes for a listener to peg in the ratings being only 15. And, while I hate to say it myself, they love that shit, and while you may want to argue with millions of dollars of market research, the fact is that radio stations have very little influence over what a person really wants to hear; shit, if they could actually do that, they sure as shit wouldn't bother paying the promotions department or the on-air personalities to try and maintain listener loyalty so that they could squeeze a few more minutes out of them; they'd just force you to listen to whatever they have on hand, as so many people love to assert, and save the tens of thousands of dollars spent trying to figure out how to keep people listening in the first place.

Whether you agree with their methods or not, their is a damn good reason why radio stations play what they play and use the formats that they do: because they work. For those of us who listen to a whole LOT of music, it begins to seem pretty boring after awhile, but we aren't the real target anyway; Joe Sixpack who spends around 45 minutes a day listening to the radio is, as there's a whole hell of a lot of him than there are us. Hope that helps at least give you an idea of why the current state of affairs in the industry exists and why independant stations who eschew format don't have great reputations of longevity.

Deo

Slashing Back

Aw hell. I wrote a big ol' novel, because I have so many thoughts on the topic - as a fellow geek and radio nut.

I think Curry has gotten it right, but at the same time, I bemoan the loss of real, old-fashioned, free broadcasting.

I loved doing college radio, because every show was basically sharing songs and getting excited about music. Now that we have the net, we have the ability for a LOT of people to share songs with us.

But more likely to evolve is some sort of XML-based jug of songs with other stuff mixed in. (But not NEWS, Curry; when we want news we go to the news channels. Does anyone else get irritated that, when a news channel has "traffic on the 2s", you know that you have to wait as long as nine minutes - an eternity for our supposedly-connected age.)

What's missing is a human touch. I don't want a channel that programs ballads every 25 minutes. I want a human, who programs ballads because s/he feels like shit, who programs love songs because s/he's in love. I want someone who appreciates the same key changes I do.

Before radio was taken over by big business, it did nothing less than spark a cultural revolution. Now radio is a soundtrack to a bad movie. Maybe the *real* reason that Napster et al have become so popular is because our ESTHETIC/ARTISTIC NEEDS are not being met by the corporate world.
--

Format "problems"

A couple of things Adam Curry doesn't address (or chooses to ignore) in his article.

(1) Stations sell advertising to businesses for cash. This cash is what keeps the station going, especially if the format isn't the most popular in a given area. More popular radio stations can charge more because more people are listening and can potentially be influenced by the advertising.

(2) Stations don't play songs that aren't popular (at least not for long). If the song doesn't have an appreciable fan-base, it doesn't get played nearly as often as the latest pablum from boy-bands or teenage-blond-singer-of-the-week.

(3) What the heck does this mean?

Don't expect any major radio networks to really jump on the bandwagon of "Formats" the way Curry outlined it. Radio stations do not want the listeners to be able to cut out the ads, or modify the playlists by 'subtracting' artists from the playlist. Yes, there will be web-casts, but they
will be what the radio station wants to play.

That's for major radio networks. Independent web-radios can play whatever the heck they want, for the most part, because they aren't dependent on advertising. Heck, it might be nothing other then the DJ ranting on local stuff for hours on end. But the prolificacy of web-radio means that if you search long enough, you can find whatever you want, be it "All-Metallica, All The Time", or 24-Hour Reggae...

Now, assuming that Napster doesn't die the Good Death due to Mssr. Ulrich and Co. any time soon, I do see Napster (and Gnutella and others) updating their programs by using "Formats" and ID3 tags to seperate the music out to make it easier to locate what you want. Napster will always have the inherent drawback of listing the .mp3s as whatever the user has them at, but if the ID3 tags are properly used, as least potential "shoppers" could restrict their searches to the type of music they want. (Of course, adapting a really good SQL database could work too...)

All in all, while using "Formats" would definitely be beneficial to the music browsing process, it is ultimately dependent on the User, as they are the ones who need to set the ID3 tags correctly.

Kierthos

Similar stuff - "formats" on the 'net

A lot of people don't seem to realize that he's not proposing created-by-the-industry-and-forced-down-your-throa t formats. He's saying that now it's feasible to create-your-own format.

In fact, it's already been done, although not for peer-to-peer stuff. Check out www.launch.com (I think.) Unfortunately, last I checked, it was Winblows-only. (Windows Media Audio for their codec.), but if you occasionally boot to Windows, launch.com is a good example of a broadcast version of what he's talking about. Basically, you pop in a list of radio stations and genres that you like. Then launch.com starts broadcasting music. If you think a song totally sucks, you tell the system never to play that song again. The system will remember that, and the chances of a similar song playing will be reduced. Rate a song high, and the system will try to increase the chances of similar songs being played.

Unfortunately, it takes an hour or two of learning before it starts putting out mostly good music. (I think my rather varied music taste confused it... Not fine-grained enough...) Note: This is an hour or two of time, even skipping songs that suck.

Check it out, play with it, and think how much cooler it would be if it had a Napster-sized music collection, including all the nifty esoteric stuff that Napster has and launch.com does not.

Another similar approach - Go to mp3.com. Go to the site of an artist you like. (Assuming that you know of an artist or two there that you like.) Then check out the "other artists we like" links.

BTW, if you like the Rocket Arena 3 soundtrack, most of those artists are on mp3.com. If you liked the stuff from Silent Warrior, Upbeat Depression, or Masada, there's a lot more good stuff on their pages.

I don't fit in a box

As I got around filling up a fair library of MP3's (mostly from my own CD collection due to low bw here) I found it more and more difficult to even listen to the broadcast radio choices here in the LA area. What I discovered about my own musical tastes is that I just don't fit in any of the boxes that are put together out there, nor would I even be able to put together a box of content for myself. My tastes in music are to varied for that.

For example, if you were to categorize the styles of music that I have in my own playlist it would include rock, blues, jazz, modern swing, 40's pop, modern pop, a couple of country tunes, some classical, comedy, a smattering of movie quote wav's, new wave, and a time range covering every decade since 1940 to present. How do you box that into some pre-packaged product for the masses?

The answer is you can't, nor should we allow them to take what amounts to a marketing scheme and apply it to the Internet. What this article is really about is the guideline for re-tunneling the flow of art through old style funnels. Prince hit the nail on the head when he compared a music consumer to those that love the art form. Old world media doesn't know how to deal with the latter, so here is a proposed blueprint for creating more of the former.

I am kind of interested in this

Where I live radio is in a dismal state. There are only a few 'formats' and they are lame. Playing the same tunes over and over. Different mixes, but same stuff. I am almost never surprised. (Go KNRK!) Napster has been a lot of fun because of the shared songs feature. I do just what the other post says above. Look for people that like what I do, then grab some new stuff to see if I like it also. This has opened me up to lots of music that I would not have been listening to otherwise.

Shared mp3 pools also have this same effect. Everybody puts some in, and you listen on shuffle play. Refresh the directory every once in a while, and you get new songs. Kind of like a group radio station. This also is way better than radio.

Maybe the building of formats has value. Just reaching into the bag of free music tends to be a lot of trouble if you are looking for something new. It takes your attention, and you have to sift through the crap to get to the gems. This has always been the reasoning behind the formats in the first place. Radio stations build an identity by the songs they play. The more diverse the list, the less they are able to focus on a particular audience.

Right now promoters have to basically go to the radio stations, and get them to put songs on their play lists. This works in a fashion, but what if you are in a lame market? You don't get to hear new stuff that is relevant to your lifestyle and interests. (could be you that is lame also, and the same problem still applies :) )

People are always going to buy CD's. The quality of mp3 and the effort to play/move/encode them will not be worth it for a lot of people. Popping in a CD and pressing play is the way to go. Most people want easy. I don't think that will change.

This means that the 'free' music on Napster actually comes at a price. This price is the time and attention required to actually get complete quality copies of songs you like. The other price is that you lose album continuity. (There still are artists out there that know what this is!) Even though the number of tunes is high on Napster, the depth still has quite a ways to go. Not everybody puts up an entire album, so that means that most of the hits are going to be on the mainstream singles with the occasional remix. Fine for people on a budget that want to fill out their collection with a few extra tunes that are not worth a whole CD, or for those that really just are backfilling old songs. What about those wanting something new?

This format thing could easily be a service that people would pay for. You get the songs for 'free' but you also get some ads, and some new stuff that you don't ask for, but that might fit. Maybe you will buy some more CD's because you are hearing stuff you like, not stuff they want you to buy.

I have always liked the surprise factor radio can have. You are listening to one of the better stations that will throw in random stuff, and all of the sudden there is a great tune! Subscription mp3 format services that are personalized have the potential to fill this void.

With this sort of system, new content would have value a few times over. Just as it does now, only more so. You sell it to the subscribers, then based on their response, move it to the radio, and promote the CD. Let the subscribers get the CD a little sooner so word of mouth works in favor of new sales. Then finally as things wind down it cycles in there with everything else and things work the way they do now. Some people buy 'em some don't. Eventually the case gets notched, and they sell at $3.99 or something.

They can still market the cd. It has ease of use, cover art, album continuity (spelling czars back off!), and sonic quality in its favor.

Formats: Just what we -don't- want

Having read Adam Curry's proposal, I have to say that this is exactly what we don't want to see happen to MP3 -- driving it into the same sorry state as American radio.

Mr. Curry claims that "you may hear Madonna one time and Jennifer Lopez the next", but what if I wanted to listen to Madonna and got Lopez instead? I want to choose the songs I listen to; I don't want some impersonal AI with no inkling of my emotions to try to decide what I'll like. Curry says that this would be "my radio station", but I don't want a radio station -- I want a collection of my favorite songs that I can sort through at will.

The whole concept behind the digital music revolution has been to empower consumers to be able to listen to the music they want, whenever they want to -- and not just be forcefed the Britney Spears / Blink-182 / Metallica / Moby drivel that record label fat cats want you to pay for. How can you expect a system like the one Mr. Curry proposes not to be abused? Napster is already under substantial pressure from the recording industry to stop distributing free music-on-demand.

Implementing "formats" in Napster might finally make Shawn Fanning and Jon Johansen rich, but it would be a huge step backwards for the digital music revolution. Let's not turn MP3s files into "My Radio Station" -- let's keep them in the independent state they should be in.

A (music) portal... How new!

There are radio stations on the web.
They do the "format" thing. That's not
what downloading is about. Downloading
is inherently offline oriented, i.e.
I download now and play back later when
I am offline.
P2P is only big now because it offers
lack of commercial involvment. I upload
you download and no commercials or
banners are involved. You don't know me,
I don't know you, nobody gets spammed.
In reality, this will degrade quickly but
formats would be worse from the start.
Also, many if not most go to Napster and
the like to find obscure songs from obscure
(often unique) artists. What format would
the great Russian band Aquarium fit in?
The nail in the coffin though is that
people have experimented with formats,
channels, boxes etc. Most have failed
because in this day and age, we need
EXACTLY what we want, not just close enough
topicwise.