Lucas Gonze blog

Media type:
application/json; schema=dublin_core

Null data URI with this type:
data:application/json;schema=dublin_core,%7B%7D

Link response header with null data URI:
Link: data:application/json;schema=dublin_core,%7B%7D; rel="meta"

Link response header with artist and song title:
data:application/json; schema=dublin_core,{Creator%3A"Michael%20Jackson"%2CTitle%3D"Billie%20Jean"}

Using a data URI instead an HTTP URI is more robust, simpler and faster. I had to invent the media type, though. application/json; schema=dublin_core is a neologism.

Also, having two semicolons in there (you need one for the media type parameter to the data URI and one for the rel attribute) might be a bug.

1. • lucasgonze clears throat
2. [6:21pm] lucasgonze: I have this thought –
3. [6:22pm] lucasgonze: apply http://www.w3.org/wiki/LinkHeader to audio files on the web, as a method to annotate pcm data.
4. [6:22pm] lucasgonze: as an alternative to id3
5. [6:22pm] lucasgonze: so let’s say there is http://example.com/example.wav
6. [6:22pm] lucasgonze: wav has no metadata support
7. [6:23pm] lucasgonze: GET on http://example.com/example.wav would return a response header pointing to a metadata file
8. [6:23pm] lucasgonze: Link: example-metadata.json; rel=meta
9. [6:25pm] lucasgonze: GET http://example.com/stairwaytoheaven.wav
10. [6:25pm] lucasgonze: Link: http://musicbrainz.org/ws/2/recording/ccfdd180-22e1-49b9-bd81-b5fcf2c6474e?inc=artist-credits%2Breleases; rel=meta
11. [6:35pm] lucasgonze: use CORS on the metadata resource http://www.w3.org/TR/cors/ to enable cross site requests from ajax
12. [6:36pm] lucasgonze: then a javascript-based audio renderer in the browser can access metadata for any file
13. [6:37pm] lucasgonze: the metadata technology is factored out of the codec.
14. [6:37pm] lucasgonze: the http request becomes the envelope for waveform data and metadata together.
15. [6:37pm] lucasgonze: metadata technology and audio signal processing technology can evolve independently

See also the Web Audio API and jsmad.

Scandinavian Music Company Set to Take Over By Reinventing the Mixtape.

X5 focuses on back catalogs of classical music and creating custom compilations with titles like “The 99 Darkest Pieces of Classical Music” or “The 50 Most Essential Pieces of Classical Music” which, since being released in 2008, has made more than $2 million worldwide. Essentially, they buy up a truckload of song licenses at low-rates, package them into winning compilations and resell at a moderate markup. X5 has released more than 8,000 of these thematic albums — some by composer, mood, holiday, etc. — with most falling under the “classical” genre.

In 2010, X5 was the number two classical label in the U.S. with a 20% market share, and had 13 #1 Billboard Classical albums — more than any other label, save for Universal Music Group (with whom X5 is currently in talks).

The company has been able to make all that money through some simple tricks: The albums are inexpensive, the artwork is simple but striking, X5 distributes through all major music sites — iTunes, Amazon, Spotify, Rhapsody, etc. — and designs albums with a kind of “music SEO” in mind. “Think of the person that types ‘classical’ into the iTunes search box,” says Scott Ambrose Reilly, X5′s new U.S.-based CEO. “That’s the kind of person we’re trying to sell to.”

Steve Jobs inspired me by building his business on quality. Before the rebirth of Apple the dominant designer in the computer industry was Microsoft, which specializes in willfully shoddy workmanship. They inspired others to do knowingly bad work and made it harder to develop a business consensus for good work. Microsoft’s world was cynical, depressing and demoralizing. Apple under Jobs countered this. It showed that cynicism towards your own workmanship is not necessary.

But when you buy an Apple product you more or less give up your right to repair it. You can’t install the software of your choosing, you can’t sell your software to other owners without Apple’s blessing. You give up your right to modify and improve a possession, which negates your own property rights. Apple is such a powerful corporation that they can sell you stuff without your getting true ownership. This is undemocratic. It increases the power of the biggest and most powerful corporations. Even Microsoft at its peak was more open. This makes me sad and a little more cynical.

Daring Fireball sez:

Show me something real, I say. Look at Amazon. Everyone knows they’re building a tablet. What have they said, though? Nothing. What have they shown? Nothing. When will they say something? When it’s done. What will they show? Something real.

Turntable.fm users pay more attention to the page than Pandora users, because TTFM is much more interactive. Ads displayed in the page will come to their attention. That’s not true for Pandora users, which runs on autopilot.

Pandora relies on audio ads. The rates for these are low, on the order of $2-$3. Display advertising commands higher rates, on the order of $8. So Turntable can earn more per user than Pandora.

Does this make Turntable a better business than Pandora? Not necessarily. The more attention something requires, the fewer users it gets. Pandora’s lack of interactivity will gain it more users.

Turntable will have higher profit margins, but Pandora will have more volume. If Pandora’s volume is big enough it will have better return on investment.

However the base rates that both Pandora and Turntable pay are quite high. Pandora’s low-margin/high-volume strategy may be too close to the bone. It may not be able to pay the bills for each user. Right now that’s certainly true – Pandora loses money.

TTFM is seeking licensing deals so that it can offer on-demand features. These deals would raise its royalty bills. It may earn more per user, but it will also spend more per user.

I wish Turntable wouldn’t do that. Why not just run a healthy business? Why go to the moon then kill yourself reaching for the stars?

You’re paying for a purchase with cash and the amount is not a whole number like $1 but a decimal number like $.01. How can you minimize the number of coins in your pocket after the transaction?

If you hand over $1 you’ll get back three quarters, two dimes, and four pennies, for a score of +9 coins. But if you hand over a penny ($.01) you won’t get any new coins back and you will dispose of one pre-existing coin, for a score of -1 coin. (*Assumptions listed below).

Low scores are better than high in this game.

The more change you start with, the more coins you have to choose from in forming your payment, the more optimal coin combination you can pick, and the lower your final score. And vice versa: the less change you start with, the higher your final score.

Let’s say you start off with a $infinity bill and no coins. On your first transaction your score will be high. As time goes on you will accumulate coins. Once you accumulate enough coins your scores will be low. As time goes on your count will reach a steady state and stop increasing.

Let’s say you have a change jar on your bedside table. When putting on your pajamas you put your pocket change in the jar. The next morning when you get dressed you do not return that change to your pocket, so that you always leave home with no coins. You will get the highest possible score over time.

This is because you will usually accumulate the maximum number of coins before hitting the steady state where you have the optimal number. Your score for each transaction will be high until you hit the point where you have so many coins that you can usually get a low score. The number of coins in your change jar will be the sum of those day scores.

Let’s say you have no change jar on your bedside table. Each day you leave home with the exact same change as the day before. You will end up with the lowest possible score over time. This is because there will only be one day (your very first day, the one in which you leave home with no change) when you have high scores. After that day you will have the optimal amount in your pocket and your scores will not increase.

*Assumptions: the clerk always gives the least amount of coins in return, for example a nickel instead of five pennies; ignore whole dollars for the sake of simplicity (but if you can’t do that, treat bills as just another form of coin, like a quarter, dime or nickel); American currency, because I’m American; the clerk insists on being paid the exact amount and won’t let you slide on a penny; a quarter = .25, a dime = $.10, a nickel = $.05, no other coins to be discussed.

There will be no apocalypse in the music industry. Copyright will remain.

The majors are the last man standing. They’re shaky, it’s true, but they’re a lot stronger than Napster, Aimster, Grokster, etc etc. Hundreds of internet companies have gotten into the ring and been knocked down. In the meantime the labels keep doing what they do.

There is still plenty of friction in accessing hit songs. Anybody with a DIY project can access anything, but no commercial vendor can directly help them. For the convenience and service that only commercial sources can provide, the sources must negotiate licenses (like Spotify), work within the DMCA non-interactive guidelines (like Pandora), or offer technology rather than content (like the MP3 Tunes music locker).

You can argue that copyright is fucked up. But that is an overly literal understanding. Copyright is how government regulation is implemented. Government will regulate the arts one way or another. It is currently doing it with a fiction of property rights on expressions. The fiction is managed and executed by the courts and from time to time tweaked by legislators. No court will invalidate copyright. The contrary: courts implement copyright.

If you have a problem with copyright, you’re fighting an imaginary enemy. Copyright is not an entity. Your problem is a human to human conflict like any other.

Regulations related to expressive works are a tangled mess of kudzu, no doubt. But the problem is not that regulation exists, it is that a lot of money is riding on the regulation. Copyright is the right to hire a lawyer.

No on-demand streaming subscription service has enough customers to matter to the labels. The labels would starve if they had to live on the revenues. So the existing services all had to accept the deals they could get without any real leverage, and these came with brutal – barely survivable – terms.

Once some service has enough customers for the revenues to matter to the major labels, the deals will be renegotiated. $200 million dollars split between all the stakeholders in the recording industry would do the job. At $10/month that’s 20 million subscribers, which happens to be about the size of Sirus/XM.

And then things will get started for real. For now we’re still in the prelude.

Hype Machine has a cool new project out – “Fast Forward.” It’s a riff on Shuffler.fm.

Part of the fun of Shuffler is the explosion of visual design as you go from site to site. FF keeps that going, but overlays its own graphics on the source page in a way that’s sexy as hell.

The original blog page is a screen shot rather than an inline frame. It looks just like the page and when you click on the image it opens the actual live page. The original page doesn’t get a bump in its page view counter, which takes away some of the fun of Shuffler for the blogger. But the overall thing loads faster and more reliably, and the blog page can’t bust out of the frame to interrupt your playlist.

A key difference with Shuffler is that FF doesn’t play the whole song, just a snippet. It’s not supposed to be a radio experience. You don’t run FF in the background and then check back when something gets you fired up.

What you do instead is use it actively to collect things to listen to. You surf around by skipping through the playlist or letting the samples play to the end. When you come across something interesting you add it to your favorites, and eventually you leave FF, go over to your main Hypem favorites, and listen to the songs in full length.

Tip: you have to be logged in for the experience to make sense. Otherwise playback and favoriting don’t work.

I love the supersexy graphics and I like the integration with all the rest of the Hypemachine universe. But it felt like too much upfront investment to have to go through a hunting and gathering phase before actually listening.

But maybe I’m just not used to the new flow yet. It’s supposed to be a different thing, not a clone. So I won’t know until later how it works for me.

Links:
Blog post about it – http://blog.hypem.com/2011/07/fast-forward/
The thing itself – http://hypem.com/fast-forward/launch