The other day David Weinberger approached me with an idea: what if he bought album downloads from me in bulk so that he could give them away? It was interesting but I was sceptical – I'm already giving away the music for free, why would anyone care?

So we agreed on a bulk price, I rigged up a special download link he could distribute and he twittered and blogged it:

I'm trying an experiment with a business model I like to call a reverse referral fee. Here's how it works…

You click on a link that lets you download a copy of Brad Sucks' latest album, Out of It. The album of wonderful music is yours for free in every sense. (Share it! Please!) But, I'm going to pay Brad for each copy downloaded, at a bulk rate he and I have agreed on.

To my surprise it got a fair amount of attention (aka free word-of-mouth advertising). Many people thanked him for buying them the album, I got a lot of mentions on Twitter that I wouldn't have ordinarily. The 50 copies were all downloaded within about an hour, but it's pretty clear more than 50 people got introduced to my music. Plus David paid me so I made out like gangbusters.

What is there to learn from this? I'm not sure. It's clear that the reaction was much larger than if David had said “go download Brad's free album, it's free and anyone can go get it whenever”. Saying that money was changing hands on behalf of their download definitely got people's attention and created a small viral chain reaction.

David thinks this could be a viable option for super-patrons and that I should offer it as an option. What do you think?

I want a Twitter-style service for freeform, taggable, time-stamped data. My intended purpose would be to log real-world items. Like I'd IM/text/email the Data Twitter service with:

bike ride tags: bike, ride, exercise

writing a blog post tags: status, writing

phone call from annoying dude tags: annoying dude, phone

basketball tags: basketball, sports, exercise

These would all be time-stamped as I add them. A duration may also be useful.

The data could then be queried, searched, accessed via API from other web sites & apps, read via RSS and imported into spreadsheets.

Potential applications:

• import specific tags (like 'exercise') for a certain time period into a spreadsheet for analysis
• power cross-site social widgets from the data feed (most recently listened to songs, task completion, current status)
• use as a remote control for triggering events in software monitoring the feeds (bittorrent clients, server administration)

It's super nerdy, by nature deals with private instead of social data and likely wouldn't scale, but I still want it.

Here's a picture I took when I was sorting socks a few months ago:

Look at all these different bastards! What for? Can the human race not agree on a black sock style?

What I'd like is a standardized black sock specification. So I could always buy replacement or additional socks that match the ones I already own. Please: open source community, W3C, Creative Commons -- somebody help make sense of this important issue.

Scott has some great ideas for Music 2.0 services. I've been meaning to write up mine for a while as I've learned I'm too busy to launch any new sites.

My big idea lately is one I've been meaning to pitch to Magnatune:

Independent record labels should provide hook-ups to their artists for services such as graphic design, manufacturing, merchandise, booking, bio-writing, press kits, photography, advertising, PR and more. These are all things artists will pay for but it's hard to weed through the scams and overpriced poor quality services out there.

The record label would find quality professionals to perform these services at a reasonable price. Record label makes itself valuable to the artists by simplifying their lives, record label takes a cut or referral fee from each service transaction, artist has access to quality resources to improve their career, sells more music, everybody's happy.

LiveAPI is an open-source third-party Ableton Live API, so Ableton Live is now hackable via Python. Create Digital Music has a write-up -- very cool.

Here's my idea: let your audience control cowbell levels of your set via text message. Make millions off of cowbell volume fights. Retire.

A while back I thought it would be cool to install one of those LED belt buckles in my guitar. And now my dream has finally become reality:

It was easier to do than I thought. Take one slightly abused guitar:

Take one LED belt buckle:

Flip the LED belt buckle over and there's a piece jutting out that's supposed to go into your belt:

Drill a hole in your guitar (not too big or the buckle will flop around):

Push the belt buckle into the hole.

Video:

It's pretty sturdy but I'll probably tie it to my guitar strap hook to make sure it stays attached to the guitar if it falls out. Also I may install a switch so I don't have to pull it out to turn it on and off.

John Buckman has the details on putting Magnatune streams into your Second Life land. This is a great idea. Users get royalty free music for their lands, Magnatune gets promotion. Very cool. 

Something I was thinking of: is there an easy way to stream podcasts into Second Life lands? That'd be another huge source of royalty free audio and would be pretty interesting. Virtual radio stations.

I still don't have a Wii (out of stock everywhere and also I haven't tried very hard) but that hasn't stopped me from fantasizing about things to do with the accelerometer in the Wii controller.

First I came across this collection of videos of the 20 Greatest Guitar Solos. My main thought was "a little guitar solo goes a long way".

Then I watched this Wii Wiimote Ableton Live Controller video.

So here's my Wii guitar idea:

1. Strap the Wiimote to the headstock of your guitar.

2. Run your guitar through some effects, such as say, a flanger or a filter.

3. Have the Wiimote control various expressive parameters of the effects.

This way, when you pull this move:

Not only would you be looking dead sexy, you'd also be making expressive changes in your guitar sound based on the movement of the accelerometer in the Wiimote. Everybody wins!

So I bought a Behringer FCB1010, it's a (cheap as hell) twelve switch, two expression MIDI foot pedal. I'm thinking I can control my laptop with it while I play guitar and sing. 

Anyway, the entire experience has been great but not because of the device itself. The manual is terrible, the factory settings don't work with Ableton Live out of the box and programming it via the foot pedals is tedious.

One of the greatest things about buying gear for me as a total nerd is when there's a thriving user community around whatever I just bought. And there's a great one around the FCB1010. Here's some of what I've found:

• There's a great Yahoo Group full of resources for it. Photos, utilities, tutorials, patches, sysex dumps, hardware mods and more.
• Hackers have made $10 replacement firmware chip for the FCB1010 that adds a lot of great functionality such as tempo tap, stomp box mode, not needing to put the device into "sysex receive mode" to transmit patches to it and more.
• There's an excellent home-brew PC Editor for programming the device.
• A great FAQ.
• Plenty of tutorials for getting the FCB working with Live.

As I was describing this, particularly the replacement firmware, a friend of mine wondered why Behringer doesn't open source their firmware. They make their money on the hardware (unlike video game consoles for instance) and are also widely criticized for ripping off other company's designs, manufacturing them cheaply and selling them at a fraction the price.

Seems to me they'd have nothing to lose by open sourcing their firmware. Hackers could add all the functionality they want, people would buy their products with the intention of tricking them out, it's free R&D that they can fold into future devices, and their nerd karma would go through the freaking roof.

Just as I was about to ditch del.icio.us for Google bookmarks, del.icio.us beefed up their social networking. So now I want to stay with it rather than move to lonely ol' Google Bookmarks. But Google Bookmarks has some great features, so I'm torn. I wish there was a service that would synchronize all my public data between all these services. They all want my data and I'm more than happy to give it to them. Why can't I do this? This ties into my previous complaint that there's no way to import my Wordpress blog entries into MySpace. A synchronizing service would also help a lot of smaller services get the amount of data they need to develop and compete on a feature level. Ho hum.

BitTorrent + Donations = Viva la revolution? is an interesting question on Ask Metafilter about providing a service that makes it easy to share music and donate to the artists (which I was actually just talking to someone about the other day.) Metafilter user scarabic suggested that the Internet doesn't need any help with the trading but to stick to the  facilitating donations aspect. Which made my wheels turn and I posted this:

The idea is decent but I think scarabic might be right. Why not find a way to hook donations into all the trading activity that's going on right now?

It would be interesting if someone would build a verified directory of where to donate money to artists. Have an open API so torrent trackers and other music trading sites could implement it as a feel-good gesture.

That way if someone's on some torrent tracker and is grabbing my album they could have a little link that says "Hey, like the album? Click here to donate to the artist."

I think that'd be neat and have the potential to catch on and also potentially put a nice spin on all the trading that's going on if it were to actually work in artists' benefit.

Not sure if it's actually doable though. If a big paypal directory's not possible, you could turn yourself into a collection agency, hold and distribute the money or something. (Then turn evil and crook all the musicians out of their cash of course.)

I haven't thought much about the technical aspects of it but I think it's an interesting idea.

Matt posts about his TV Time Capsule idea and this is actually something that I've been wanting to build into junklog. Being able to log and discuss episodes of TV. But I have my own thing someone should build because I don't have time. It would be an open episode guide service with some sort of API. My idea is that it would be like Wikipedia but for episode guide (and movie listings, and so on) data in a structured format. This would make it possible for services like junklog to identify unique episodes and group them. I think there are a ton of tv-based services that could exist if this data were available.

But maintaining a database of that size and keeping it up to date seems virtually impossible for just one guy (me) to do, so some sort of open wiki-style episode guide would do the trick.