(Michael Chinen)

I did a performance a few weeks ago at the Linux Audio Conference in Maynooth, Ireland. I had no idea that the cars run their like in Japan – on the left side. I guess this is the urban equivalent of galapagos island evolutionary theory.

The performance was done as a 4 hour long installation in front of the main concert hall. This was nice because it made the whole thing really informal and people would come up and talk. It’s probably the only performance where I got multiple requests for specific sorting sonifications (e.g. ‘Please do mergesort on a size 30 array with 5 discrete values next’). It had the techy-conference problems of having the entire audience being male except for a few wives.

One of the nice things about LAC was that it was very low on expenses. This is something I have been thinking a lot more since my gradual departure from the academic ivory tower while still doing some experimental music and living like a cheap bastard in Berlin. LAC was completely free to register, and my total transportation (inc. airfare), food, and hostel costs were less than 200 EUR. For comparison, I’ve been accepted to ICMC the last two years but decided it wasn’t worth going to as they generally charge around $300 or more just to register. I don’t quite get this as they should be using university venues and funding to host. LAC is much smaller and they still manage to pull it off. Sometimes I wonder if ICMC does this specifically to keep non-academics who won’t have institutions to fund them out. ‘Pay to play’ kind of is a stupid model at any rate.

I was hoping that FFmpeg folks would be there, but they weren’t. I mean to work more on a seeking API for them over the next few months.
Instead, I met up with a friend from my Tokyo Denki University lab, Rennick Bell, who did a presentation on using haskell as an interface for livecoding with supercollider. Rennick also came to Berlin after which gave me a much needed reason to visit the SuperCollider meeting at NK.

I also met some interesting new folks working on new and interesting things.

Flávio Luiz Schiavoni and Giuliano Obici were fun Brazilians (apologies if this is a stereotype) who did a pretty nice installation called Concerto para Lanhouse:

Flavio also has created a system for networking music called medusa (pdf) which I am kind of interested in to use with someone for livestreaming performances.

I also met Daniel Mack there, who is working on writing a OS X version of pulseaudio. This sounds great because then I finally pipe audio between my linux and mac boxes (I have four). Apparently JACK works on mac, but I have never gotten it to. On the bus ride home from the airport I also met him again, meaning that we both live in Berlin so maybe there will be some work we can do together or I can help test.

Tim Blechmann of SuperCollider also presented a new system called ‘Supernova’ that uses parallelization in SC for better performance. I’m interested in that from a code standpoint. Apparently the work from Supernova is being or going to be used in boost libraries. Once we get the stable release of Audacity out I’m going to look at our on-demand engine, so it was nice to be able to talk to Tim about these ideas.

As far as my own musical work I’ve been modifying lstn somewhat – now there is some 8-bit visualization that actually looks kind of cool. But for the most part I have been focusing on Audacity and contract work. The stable release seems as always 2-3 months away despite my crunching about 8 bugs last month. God damn those P2 bugs. Since I’m not doing summer of code this year I hope I can pound it out.

To keep contributing to the music community though, I think I am going to do some writeups of the environment (both musical and japanese) in Berlin next.

gg.