It started with a tweet from @AudioPornTwitz about a Little Boots remix contest. The Earthquake stems were posted online so I DM’d Fernando to see if he was interested in having a play with them for our follow up to Separate.

I think it’s safe to say that Fernando and I have had enormous fun working on the Little Boots stems.

Producing glitch files, uploading initial arrangements, hosting diva-style bitching sessions (only kidding) and developing sounds on the fly have involved most of the social networking and music collaboration sites at one point or another.

We’ve Twittered, Waved, IM’d, SoundClouded, Skyped, Bandcamped and DivShared. We even emailed once. (But we’ve still not met.)

There have been tasters posted to this blog, my blog and private shares to a very privileged few – thanks for the comments guys! And our conclusion from all of this?

Guitars Are Pianos.

So here it is. We proudly present Half-BIT’s take on Little Boots’ Earthquake – The GAP RMX:

Earthquake – GAP RMX by Half-BIT

Done reading? Let’s go.

Phase II Concept

Grab tweets that are either using #expanding09 and/or Barcelona and feed them into Max/MSP. Inside Max/MSP those tweets are going to be dissected and used to attribute several values to parameters that will generate cross effects that will be then applied to the original tracks. Each keyword uses a stereo channel.

So far so good right? But I was not really happy with this. Twitter is all about constraints (the 140c rule) and I started to think that a tweet that was more characters should have more weight on the whole databending process than a tweet that had less characters.

Development

Setting the two original tracks as targets I started to feed tweets (the infamous #soundgeek #flirtlines) into Max/MSP. The first objective was to get it to respond to those keywords and generate parameters (by associating to each keyword a random value that would be fed to the effects pool).

When this was working to my satisfaction I then gave the next step which is the way that the whole patch is going to work:

Each tweet that includes one or both of the selected keywords will not only trigger the effect pool. The amount of characters that each tweet has will be taken into account by attributing random values to each character on that tweet.

(On a offline test I fed a PDF file into Max/MSP: The original files were destroyed, literally. I knew I was on the rigth path).

Now you are asking: “Random? We all know random real random doesn’t exist! Or do you have a lava lamp?” Nope, no lava lamp here.

To generate those random values I will be using dynamic data as well: in this case the number of followers that the user, sending the tweet,  has.

What you can listen to below, are the 2 original tracks and what was produced by setting the patch to grab tonight’s trending topics #oldnames and Glee.

Next saturday the more you tweet #expanding09 or Barcelona the more effective this demonstration of using Social Media to create new media art will be.

Expanding09 by Half-BIT

Alka is to blame. He did this amazing track called “Separate” and asked around for remixes.

Kieron did a remix. Fernando did a remix but he was not totally happy with it. Kieron intervened. Half-BIT was born. This is the start of a new project.

And it starts like this.

Separate – STRBX RMX by Half-BIT

Well, Mr Fonseca and Mr James  require your help again. Our newly formed collaboration, Half-BIT will be providing an art sound installation at Netaudio’s EXPANDING 09 Festival on 14 November in Barcelona.

Half-BIT is all about collaboration. It’s our raison d’être. It’s how the whole project was kicked off. And we want it to be global. Our first track started life as a wonderful IDM piece from Alka’s “Dog Lost in the Woods” album. It was remixed by Fernando Fonseca, then remixed again by Kieron James. None of us have actually met. Alka is based in New Jersey, Fernando kicks around Barcelona and Kieron inhabits the far less urban-cool, though equally chic Cheshire, UK.

Just like Cosecha, Fernando has asked Kieron to seek parameters to create a new source file for him to glitch the hell out of for a sound installation on 14 November (when Kieron will be tucked up in front of Faithless at Manchester’s Warehouse Project). And we need YOU to help with those parameters.

This time, Kieron intends to use two of his favourite VSTs: Ohm Force’s Symptohm and BigSeq2 from Audio Damage. He’ll request some parameter values to create the sound (e.g. send  a number between 1 and 5000) and you’ll tweet back with your number. Just like Cosecha, Kieron’s not going to disclose which values are being requested, but unlike Cosecha, Kieron anticipate far fewer requests being made of the contributors. (Promise!)

Once Kieron has  handed our sound over to Fernando, he’s going to work even more collaborative magic to create a very special piece for Barcelona by using a max/msp patch that will gather live data from Twitter that mentions Barcelona and the Expanding’09 Festival that will mash the glitch out of the original sound in real time.

Play your part by adding a comment below with your Twitter username, or tweet us up with #halfbitbarcelona in your tweet.

We’ll keep you informed of progress via our favourite collaborative social networking wonder-drug: Twitter.

The following people / organisations have volunteered to respond to data requests.

@_alka
@BlueEydGlambert
@dust_bunnies
@fbish
@groovecriminals
@heuristicmusic
@HipGnosis23
@leeconnolly
@LucidAcid
@madebyrobot
@publicspaceslab
@pure_tone
@retorta
@ronsens

Sound 1 – Ohm Force Symptohm Output [30/10/2009]

Thank you to all the contributors for providing numbers today. Below you’ll find a SoundCloud player which previews the three bar loop your numbers have generated. Don’t get too excited just yet. There’s a second stage to the process where we mash this up using Audio Damage’s BigSeq2.

The table below illustrates how the sound was generated from the numbers supplied:

None of the contributors was aware of the parameter for which they were submitting a value. The request simply read (for example), “Please send me a number between 1 and 4000 (inclusive). Thanks.”

Phase 1

Here’s how the loop was generated:

The first four contributors provided the notes for the loop. I preselected C major scale and with 12 notes in an octave we have a pattern of E, E, C, D.

Parameter 5 governs whether the pattern should then repeat or reverse (0 for repeat, 1 for reverse). So the second four notes are D, C, E, E.

Parameter 6 establishes the tempo of the loop: 123bpm. Parameter 7 sets the note length: 2 being a minim, 4 a crotchet, 8 a quaver and 16 a semiquaver. The loop length was established as 3 bars by Lee Connolly. The sample preset from Symptohm is 72 (literally counting through the sample files in the Symptohm library!) and the remaining four parameters establish the filters applied.

Loop 1 – Symptohm

Symptohm (Half-BIT #Netaudio) by Half-BIT

More to come! Stage 2 will be the mash up and sequencing in Audio Damage’s BigSeq2.

Phase 2

Phase 2 of our Netaudio project applied a VST plugin named BigSeq2 from Audio Damage to the source file above. Once again, the values were requested without conveying to the contributors which parameters they governed. The table below illustrates how the values were applied:

Loop 2 – Symptohm plus BigSeq2

Symptohm BigSeq2 by Half-BIT

Thank you to all the contributors. That’s my bit done. Now it’s over to you, F.