save the commas

Ok, tags are separated by commas, right? So if you want to have the tags “a” and “b” on some object you enter the tags as “a,b”, yeah? Right, well, who’s looking out for the commas?  One minute they’re getting entered, the next they’re nowhere at all.  Pity the comma!

Is music a substitutable good?

Ad-Supported Music Central: The Times is a Great Textbook

In any industry the low-cost producer of substitutable goods will always win (whether recorded music is substitutable is open to debate but I would argue that it is, sicne listeners have virtually unlimited choice in what to spend their time listening to). It seems like the recorded music business is just beginning to learn this fundamental principle of business.

What is substitutable about music is choice *before the fact.*

Once you have come to know and love a piece of music, only that one will do. That is why there are classics. There is only one Kind of Blue, and people who know and love it will never have a substitute.

But before you meet Kind of Blue, you have a universe of choices. Your listening journey doesn’t spiral inwards to an inevitable meeting with Kind of Blue. Drop into a path starting at Opsound, for example, and you will end up loving other musics.

LA gig thursday

I’ll be playing three waltzes, a galop, a polka, and a march on Thursday the 26th in Silverlake.  It’ll be a low-key scene in a small room with no cover charge.

Club Fluffer at The Hyperion Tavern
1941 Hyperion Ave.
LA 90027

Cross street is Lyric – across and a little up from Casita del Campo – look for the barber pole outside.

Madame Pamita
Tequila Mockingbird
The Recovering Catholics
Miss Lady La Diva
Lucas Gonze <– opening slot, so 9pm

guitar lessons as the transmission of culture

guitar teacherPer NPR

Thousands of guitar students lost a valuable resource last week. The most popular guitar teacher on YouTube saw his more than 100 videos yanked from the site. The reason: a music company accused him of copyright  infringement for an instructional video on how to play a Rolling Stones song.

Culture relies on shared references. Sharing requires copying. When a new guitarist copies the way that a skilled guitarist plays a well-known song, culture is being transmitted from one generation to the next.

When a music publisher prevents musicians from learning a song, they are destroying the value of the song. There’s no reason to learn the Smoke on the Water riff except that everybody else knows it, and cultural ubiquity isn’t possible unless learning is absolutely free and unencumbered. Notice that the song in the original quote is by the Rolling Stones, a band that couldn’t matter less if it weren’t part of pop culture canon.

One result of copyright extremism will be the disappearance of cultural icons like the Rolling Stones. They haven’t contributed anything fresh to the culture for close to forty years, and without third parties reusing their old work in ways that make it fresh they hardly exist. In terms of 2007 pop culture, all those covers of “Paint it black” *are* “Paint it black.”

This is why I am resurrecting 150-year old songs and posting them, along with sheet music, on my blog — it’s possible for those songs to be used as source material for new work.

But I suppose that this is needless worry. Waves of takedowns for items like free but unauthorized guitar lessons are usually part of licensing business deals. Nobody bothers to ask for the takedowns unless they have a competing commercial product for which they have paid to license the source materials. If unlicensed guitar lessons featuring Rolling Stones songs are being knocked down, it probably means that licensed ones are coming up behind them.

pumpkins vs beasties

Check out the Smashing Pumpkins’ Myspace. That entire huge graphic at the top of the page is a link to buy the new album on iTunes.

On iTunes, not Amazon. It’s digital or bust.

Compare to the Beastie Boys’ Myspace, which is one giant link to their blog. Both of these bands are Gen X legacy acts with a current release that they badly need to get traction, and both are doing something creative which is in tune with the times.

My bet is on the Beasties. A blog is about an ongoing relationship to their fans, while an album release — even on the iTunes store — is an old-school one-shot / fire-and-forget relationship with the fans. The Pumpkins’ access to the fans is cool and distant, while the Beasties’ is warm and intimate.

Part of the problem with CD sales, in other words, is that the art form itself is dying. The problem is partly with the CD as a distribution vehicle, but it’s also with the album as a medium.

independent web publishing for musicians

Music Publishing and the Web: Back to Basics | MiDNiGHT PARKiNG

Going back to the most basic questions:

a) How do musicians publish their work and support themselves?
b) How do listeners find new music?

To answer the first, I would argue that any semi-professional musician that wishes to build a career off their work should establish their own unique website. It is more work but the benefits are tremendous – the artist maintains complete control over their work and can determine exactly how their work is published, distributed, licensed, and sold (even setting up their own storefront). The idea should be that this site is the *exclusive* method for updates, downloads, and sales of music.

Rather than use a social network or third party publisher, I own the domain where I do my music and writing. This creates a big problem in attracting people to listen or read, but it means that I own the upside when it happens.

It also means that I can do a decent job of keeping up with updates, since I don’t have to keep up with multiple social networks.

The Beastie Boys’ Myspace is just a pointer to, and is an unusually lively and well-maintained artist site. Which goes to show that:

  • There’s a good reason why the Beasties have done so well over the years — they get the point.
  • Artist sites have to be living things with a regular flow of updates. You are better off with one well-maintained site than many badly maintained social network profiles.

On the other hand, having many badly maintained social network profiles is a better approach than a mersh band site with a pretty but sterile environment. A band’s site which is just a department store window display is the only thing worse than an abandoned Myspace.

I don’t mean to imply that going it alone is always the right thing. There are no aggregators for musician blogs. A musician blog is an inward thing — it’s about your own music, not anybody else’s — so the few musician blogs that do exist have no reason to link to one another, and the network as a whole is going to have a hard time getting cohesion. Music listeners are technically unsophisticated, so relying on RSS subscriptions for stickiness is a bad idea.

There are problems, in other words. But good things are usually harder than normal. It’s worthwhile.

Ella Waltz 06032007

This post is a recording of the composition Ella Waltz by D.E. Jannon, which was published in 1854.

MP3: Lucas Gonze — Ella Waltz

Ogg Vorbis: Lucas Gonze — Ella Waltz

It is the third of a set of three waltzes by D.E. Jannon. I have also blogged recordings of Amy Waltz and Carrie Waltz. I don’t consider the series finished because I want to redo the Amy one, but who knows whether I’ll really come up with a better version in the end. It takes a ton of practice and a lot of trial and error with the arrangement to make one of these recordings, and I have other tunes that I want to move on to.

As I was learning the 3 waltzes I made up a back story for them. In my imagination they are named after D.E. Jannon’s three daughters. They are ordered from oldest to youngest. Amy is a teenager, Ella is a little kid, Carrie is in-between. Amy is going through a phase where she is hustling all the time and in a hurry to get away from her parents. Ella has been falling down, dropping things, running into stuff, and generally being accident prone. Carrie is moderate in all things.

The original writing on this tune had dead spots, places where the writing was thin or weak and needed fixing, so I rewrote many of the parts. My version isn’t as simple as the original, which is a loss, but it sounds better.

By the way, I got the name of this tune slightly wrong while I was working, and even though I corrected it in the end some of the metadata and file names are wrong. Right: Ella. Wrong: Emma.

These recordings are released under the terms of the Creative Commons Attribution-ShareAlike 3.0 license per my boilerplate licensing statement.

Scratch post for working out song posting process

This post contains my notes and scratch materials while I’m working on the song publishing process. It has to be public because I need to see the final public URI as part of the work. Keeping notes like this puts me in a position to write a program which automates the whole thing. It is vital that you only read the rest if you are an obsessive-compulsive open media software developer.

Licensing Steps

First, I’m need a stable URI for a blog entry before I publish it, which I need for Creative Commons metadata in a song file. I did that by setting the “post slug” in the WordPress blog entry editor. A final public URI can be gotten by appending that slug (e.g. “testingfixeduri”) to the blog URI root (e.g. “”). The final result in this case is

Second, I need to set up the song file metadata in Audacity before I export to MP3 and Ogg. I plug in the above post URI to the fields WOAF and CONTACT, and plug in the URI of the license I am using ( in the fields WCOP and LICENSE. Here’s what the metadata editor looked like when things were all ready:

Audacity Metadata Tags

It’s not clear to me that all of those tags made sense in context, though, since some were for MP3 and some were for Ogg, and there was no way to distinguish MP3 and Ogg tags in the editor.

Third, I export from Audacity to MP3 and Ogg.

Fourth, I need content-derived identifiers to publish as “verification metadata.” The way I do that from the shell on my iBook is like this:

lucas-gonzes-ibook-g4:~/Music/release/emmawaltz-06jun2007 lucas$ md5sum *
05387f19c7e084d324a92f8a7031a35d  EllaWaltz_06032007_final.ogg
6426f327e71665272c263cd42607b210  lucasgonze_EllaWaltz_06032007.mp3

According to the instructions on the Creative Commons wiki, I turn that into HTML by using this template code:

   <span about="urn:sha1:MSMBC5VEUDLTC26UT5W7GZBAKZHCY2MD">
        Example_Song.mp3 is licensed under
        <a rel="license" href="">CC BY 3.0</a>

Given that I’m using the MD5 hash algorithm instead of SHA1, and that my own songs have their own hashes and licenses, I believe this is what I end up with:

   <span about="urn:md5:6426f327e71665272c263cd42607b210">
        lucasgonze_EllaWaltz_06032007.mp3 is licensed under
        <a rel="license" href="">CC BY 3.0</a>
   <span about="urn:md5:05387f19c7e084d324a92f8a7031a35d">
        EllaWaltz_06032007_final.ogg is licensed under
        <a rel="license" href="">CC BY 3.0</a>

On a practical level, though, this is gibberish, because I made up the md5 URN type and because I know of no consumers for this verification metadata. I am pretty sure that search engines and other bots will interpret this to mean that the embedding web page is under a Creative Commons Sharealike license, which is true only by accident and not relevant to the license on my song files. And that’s in the best case situation where I managed to correctly interpret the instructions from Creative Commons.

Note to crypto-weenies: I used MD5 instead of SHA1 because I happen to have a program to compute md5 hashes on this computer and I don’t happen to have a program to compute sha1 hashes. Please don’t carp at me — MD5 is secure enough for this application.

The last thing I need to be done with copyright is a textual explanation for the benefit of human readers, which is probably the only aspect of this rigamarole that will ever be used in practice. This is it:

These recordings are released under the terms of the <a href=””>Creative Commons Attribution-ShareAlike 3.0</a> license per <a href=””>my boilerplate licensing statement</a>.

All together, this licensing jazz is a royal pain in the ass and most of it is not productive.

Gather media files

I make a new directory to hold all the files. This feels similar to packaging up open source software.

In Audacity I export to MP3 and Ogg Vorbis. I don’t export to a lossless format like Ogg FLAC in order to preserve the ability to sell lossless files for commercial uses.

I grab an image file of the sheet music.

I take a picture in the mirror to use as album art, then connect my camera to my laptop, import the picture into iPhoto, and export from iPhoto as PNG.

I grab a copy of the original sheet music from the Library of Congress web site. (I have previously checked the LOC copyright claims to confirm that this is OK).

Lastly, I upload these files and save the final public URL for each.

Create blog entry HTML

This is what I’ll need in the HTML:

   1. Composition metadata:
         1. Song title
         2. Song composer
         3. Publication date
   2. Links to all media files
   3. Link to source web site for composition
   4. Media player to render the song in-place
   5. Annotation about the music or recording
   6. Inline image for sheet music, linked to source web site

Here’s a template to fill in with all that, using square brackets ‘[‘ and ‘]’ to mark where I need to fill in items:

    <h1>[song title] [recording date]</h1>    <p>This post is a recording of the composition <q><a href="[Link to source web site for composition]">[song title]</a></q> by [song composer], which was published [publication date].</p>
<p>[Annotation about the music or recording]</p>
<p><a href="[Link to source web site for composition]"><img src="[Inline image for sheet music]" alt="[Link to source web site for composition]" /></a></p>
      <p>MP3: <a href="[]" title="[song title]" ><img src="[album art]" alt="" />Lucas Gonze -- [song title]</a></p>
      <p>Ogg Vorbis: <a href="[]" title="[song title]" >Lucas Gonze -- [song title]</a></p>
      <p>[Media player to render the song in-place]</p>

Update July 11:

Here is PHP code for filling out the template HTML.

<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN"
$songtitle = "Ella Waltz";
$recordingdate = "July 3, 2007";
$linktosourcewebsiteforcomposition = "";
$songcomposer = "D.E. Jannon";
$publicationdate = "in 1854";
$inlineimageforsheetmusic = "";
$linktoogg = "";
$linktomp3 = "";
$inlineimageforsheetmusic = "";
$albumart = "";

$annotationaboutthemusicorrecording =<<<END
<p>It is the third of a set of three waltzes by D.E. Jannon, though I don't consider the series finished because I want to redo the first one.</p>

<p>When I play this I imagine that the 3 waltzes are named after the composer's daughters.  In my imagination I think of this third one as being for the youngest, who is going through a phase where she falls down a lot, drops things, and otherwise has a lot of accidents.  I usually make up little stories along these lines about songs.</p>

<p>By the way, I got the name slightly wrong while I was working, and even though I corrected it in the end some of the metadata and file names are wrong.  The wrong name is "Emma Waltz."  The right name is "Ella Waltz."</p>


print <<<END
<h1>${songtitle} ${recordingdate}</h1>
<p>This post is a recording of the composition <q><a href="${linktosourcewebsiteforcomposition}">${songtitle}</a></q> by ${songcomposer}, which was published ${publicationdate}.</p>
<p>MP3: <a href="${linktomp3}" title="${songtitle}" ><img src="$albumart" alt="" width="200" />Lucas Gonze -- ${songtitle}</a></p>
<p>Ogg Vorbis: <a href="$linktoogg" titl!e="${songtitle}" >Lucas Gonze -- ${songtitle}</a></p>
<p><a href="${linktosourcewebsiteforcomposition}"><img src="${inlineimageforsheetmusic}" alt="${linktosourcewebsiteforcomposition}" /></a></p>
<p>These recordings are released under the terms of the <a href="">Creative Commons Attribution-ShareAlike 3.0</a> license per <a href="">my boilerplate licensing statement</a>.</p>

digital locker progress report

On June 26 I blogged that I was checking out the MP3tunes locker service, and that it felt like there was something worthwhile there.

What I liked was being able to access the music files I keep on my home PC from work, as well as having a backup. I could hack together my own solution to enable this, but it would be a fair amount of hassle and potential expense to come up with enough disk space.

The down side became apparent pretty quickly, though — disk allocation is all or nothing.

MP3tunes had a 1GB cap on the size of my locker unless I bought their premium services. I wasn’t going to pay, and 1GB is too small a portion of my personal cache to be useful. I fiddled around with what little music that was, but it just made no sense. I do maintain a 4GB subset of the total for listening on my portable, but I’m not going to do that for an even smaller 1 gig subset that I don’t care about much in the first place.

So I didn’t come back to MP3tunes after the first couple days.

The most recent development is that I got this email:

Congratulations! Your 1 GB Oboe music locker has been upgraded to UNLIMITED STORAGE.

Yes, we said unlimited storage for all of your music – FREE! (no catch, no credit card, completely free)

That’s clearly the right thing for me, so I’m now going to go back and upload the rest of my cache.

The deal with whether you get unlimited storage or not is a little whacky. At sign-up time you are in a lottery to see whether you will get unlimited or limited storage:


Unlimited Storage is available for a select number of sign-ups each day. Sign up for Oboe Free and start syncing to be first in line. Or get unlimited space now – sign up now for Oboe Premium.

And now I have learned that after you accept a limited size locker you automatically get upgraded to unlimited, meaning that there is no size limit as long as you can wait and are willing to put up with initial inconvenience.

I don’t think they care about the disk space, I think that they care about price discrimination. If they can get somebody to be a premium customer, that’s clearly more profitable than an ad-supported customer, so they put on the screws. But the underlying product is almost exactly the same thing.

What I read from this is that the business is not so great at selling ads at this point. They’re probably generating little or no profit on the ad-supported business. Otherwise they would remove all limits on locker size. This would maximize the number of users and make their business model more scalable.