Musichackday Stockholm comes to the end. People start submitting their hacks to the musichackday wiki. There are some awesome projects to explore.
Simon Hohberg and Robert Böhnke build this radio that plays songs from soundcloud by city and tag.
The swedish company Winston Design build holodeckapp.com a webapp that allows bands to create their own profile with content from soundclound, songkick, tumblr and last.fm – it is like flavors.me for bands.
Also at an experimental stage is AMIE by Jan and Wolfgang who worked with me at Musichackday Stockholm in Hamburg. The most impressive parts are in the code. Jan created a JavaScript based Synthie which he uses for his other project as well. Even though there is still some flash involved most of the sound creation happens in JavaScript which I think has a great future. Check out their video about the hack.
My most favourite project so far is HacKey. The mashup between last.fm and Echonest answers the simple question “Are you a major or a minor person?” by analysing your most popular tracks. If I got it right, the whole project is done in JavaScript, including API calls, analysis, datavis and playback of the songs. Great hack, so far! I haven’t used last.fm in last time but in January 2006 I was 80% major!
What are your favourites? I hope to discover more via live stream when the presentations start.
Yesterday RyanAir changed my plans. Today I changed my plans, too. I did not feel fit enough to start building my SoundCloud / Echonest comment based remix machine and hacked around, but did nothing real this morning. After reading about the new Metreo Charts in last.fm’s API I finally build “My City vs. Your City“.
“My City vs. Your City is a JavaScript based app that compares to what artists people listen to in different cities.
Multiply color (to get green) for intersection of circles
Compare different times (Hamburg is 60 % of New York 5 weeks ago)
Do you have any other ideas?
Update: I use data by last.fm to create the charts. At the time of building last.fm only offers data for the 100 cities listed on the website. I’m sorry, but cannot add more cities.
I find myself in a bus on the way back to Hamburg and not where I was expected to be: in Stockholm. My flight from Lübeck was canceled due to bad weather and Ryanair was not able to provide any alternative. No other way to get there.
So sad.
But on my way I met some nice guys who where on their way to musichayday, too. We spontaneously decided to start a remote hackday tomorrow in Hamburg.
Musicians, coders, hardware hackers: Come to mindmatters office (Schlachthof / U Feldstraße). We’ll for sure watch the live stream from Stockholm and do some work on our planned projects as well. We start with breakfast at 11:00 am 30th January. Join us!
The remote hackday seems to be a good plan to turn our energy towards musichackday into something valuable.
I hope to soon have the chance to really join a hackday to continue and start interesting discussions and wear one of those sweet Echonest tracksuits!
I’m really looking forward to meet some guys from RjDj. As I still love their concept of composing interactive music I never made it into PureData. I hope someone can push me in the right direction to manage the first steps and get an idea why PD is right. I’m also interested in what they think about providing and working with analysed data. I think it would be good for composers to work with values like ‘intensity of movement’ or ‘equality of movement’ and not only have direct ‘physical’ values from the accelerometers. What do they think about compositions for specific locations or movements? Is there a chance to realize some ideas I had while working on my Bum Tschack Wippe project with RjDj? I’ll try to find that out.
My initial project idea is to work with the timeline comments of Soundcloud tracks. I wanna dig into the data to reflect how timeline comments increase the value of creating and listing to music. The ideal result is a remix machine that uses word analysis of the timeline comments and music analyses provided by Echonest to create new songs form Soundcloud tracks. Therefore I’m excited what Echonest new APIs will look like. On my road to that remix machine I’ll try to create some data visualizations and a Soundcloud player that displays the timeline comments (maybe I’ll reuse some parts of the lyrics visualization I did last week).
Another project I find quite interesting is the Web Synthesizer, a javascript synthesizer running in the browser. Jan from Hamburg shared the idea on his blog. I hope he will push some ideas from the guys at bocoup forward. And I’m sure, that there will be a lot more interesting projects!
I’m really looking forward to be in Stockholm tomorrow.
Peter just published a post that sums up all hackday projects around the world. My chance to follow these super-interesting links, that i missed on hackday:
Thank you all for chatting, giving tipps and sharing such great tools as trackmate and reacTIVision are. The livestream worked pretty well an it kind of felt like beeing and working with creative and productive people in one place.
I reached my goal to create a ‘fritzcrate to audiosoftware via midi learn interface’. Well – it is a prototype yet, based on dirty code but it works for now. Hopefully I’ll find some time next week to add a special fritzcrate sound set to make some noise then.
lusidLearn is a software that should help people connect lusidOSC messages to audio-software without programming.
I’m not a midi expert. Therefore I’m not sure if lusidLearn will ever work but I’ll try to build a prototype in processing today.
The software should work like this:
1. Setup Mode.
lusidLearn waits for an new uniqueID [Object on the crate-surface] and saves this object as ‘Learn Key’ Object.
2. Learn Mode.
The user enters ‘Midi Learn’-Mode in audio software and clicks a controller. Then the user adds the ‘Learn Key’ Object on the surface. The next Object that is moved by the user (for lusidLearn it is the next lusidOSC message with changing values) is mapped by lusidLearn to the controller in the audio software. (lusidLearn will send out a new midi signal, which is mapped to the controller in the audio software).
3. Play Mode.
For every lusidOSC Event that was mapped before lusidLearn sends out a midi-signal.