After the success of the Pyramind mixer event in San Francisco in October, the Manhattan Producers’ Alliance, which is based in New York City, decided that this might be a great thing to showcase over there.  Since I was in town anyway for the Spectrum release events, and would have my BCMI equipment with me (I was due to perform the demonstration in the East Village at Spectrum over the weekend), it seemed like a great opportunity.

The format of this demo/presentation was always likely to be different to the Pyramind event, due to the fact that we had arranged for it to take place at Dolby Labs in midtown Manhattan.  Instead of a studio complex, this was a screening room, and as such was more geared towards a lecture-style event than an informal, hands-on walkthrough.

With this in mind, I decided to preface the demo with a broader discussion of ‘where are we going with this’?  It seemed appropriate to try and address the broader context of BCI/BCMI, especially in light of who the audience was.  Since this was taking place as part of the Audio Engineering Society‘s quartetly meeting, there were a lot of engineers, producers and musicians, and slightly less in the way of Bay Area tech-nerd types.  I decided to alter my presentation accordingly, to discuss how brain computer interface technology might be applied in the future not just to games (which is, to pardon the pun, a no-brainer) but potentially to film and other linear media, as well as in a therapeutic context.

And so, after rambling on for some time about this, I handed over to my mad-scientist friend and colleague, Dr John Long, researcher at NYU and fanatical robot-builder.  While he extolled the virtues of Brain Machine Interfaces and even delved into a little philosophical conjecture on the subject, I strapped myself into the Emotiv headset, ready to do some serious sound-shifting in this wonderful theater.

(Cut to post-performance) Although the demonstration went ‘well’, and certainly generated no end of questions from the audience (so much so that I think we kind of overstayed our welcome!) – there were a few, very simple things that I feel I could have, and should have, done differently.  Speaking to an audience member at length afterwards over cocktails confirmed this.  “You should be the performer – the star” – he said, and he’s right.  There I was, controlling this cool brain music system, and I was tucked away in the corner with no light on me, mumbling explanations of what I was doing that no-one could hear.

Instead, I should have thrown off my ‘presenter’ jacket, marched up with confidence to the ‘sweet spot’ seat in the surround theater, and hooked myself up to a microphone.  Sitting amongst the audience would have created a better connection, and the microphone would mean I could mumble instructions (too much projection or articulation would have made for noisy EEG signals) that could be heard by everyone.

I’m certainly planning on developing this 2-part presentation/demo, and I think the ‘lecture’ part really got people thinking about the possibilities.  That said,  I fully recognize that they also need to get turned on by the idea, and being a confident performer who knows how to put on a good show for the audience will likely get me 99% of the way there.

Highly risky theremin demonstration of 'proof of context'

Last week, I gave an Ignite Talk.  And not just any Ignite talk – this particular event was being held at Google I/O – one of the largest developer conferences of the year, in the Moscone Center, San Francisco.

Walking into the room, which could easily have housed a major team sporting event, I noticed that it was about a third full – that is to say, about 400 people.  Then the Ignite audience came in, and the room swelled to over 1000.  I steeled myself, flashing through my slides in my head as I rehearsed my spiel one final time.  For anyone not familiar with the Ignite format – it’s five minutes.  You get to talk about whatever you’re passionate about, but once your five minutes are up – that’s it.  Next please.   And here’s the kicker – your slides auto-advance every 15 seconds.  20 of them, in quick succession, with NO speaker notes (that little bit of information I found out the night before the talk).

It’s a pretty terrifying format for the uninitiated, and I was giving a talk about a subject that has significantly more depth than ‘the best way to tie your shoelaces (this was an actual Ignite Talk).  Namely, I was talking about experimental music, and how it can sound bad, and how maybe we as artists can work to achieve a balance between proof of concept and what I call ‘proof of context’.  Throw a live theremin/sampler demo (all of 30 seconds long) into the mix, and there is significant potential for disaster.

That said, my talk went pretty flawlessly (you can view it here – I come in at 38 minutes).  There were a few minor stumblages, and the ending was pretty hasty (since I was out of slides), but overall I was happy with how I delivered it.

Then comes the question of the content.  I’m pretty convinced I will have pissed off at least some small contingent of experimental composers with what I said.  Basically it was along the lines of “let’s do experimental music, but PLEASE, can we try and make it sound pretty?”.  A gross oversimplification, to be sure, but my point was essentially that many experimental composers, at least historically, have been obsessed with algorithm to the detriment of aesthetics, and it doesn’t have to be that way.  I use as an example of this my work with BCMIs (Brain Computer Music Interfaces).  A large scale project that we conducted involved all sorts of cool technology and science, but the output was not, shall we say, particularly musical.  If we had just tweaked a few things on the output end, we could have created a more musical context, but we were primarily concerned with making the damn thing work.

Of course, in building systems such as in Music for Online Performer, certain aspects must take precedence, and time is often of the essence, but if there is anything I learned from the project, it is that without a musical context, such an endeavor becomes less a composition than a science experiment.

Some composers who consider themselves on the leading edge of their art will probably take umbrage at my suggestion that we consider the audience’s reception of our work, but to these I say ‘let’s have a discussion about that’.  I am certainly no stranger to the avant-garde aesthetic, but let’s just say I’ve bought the T-shirt on that one, and am quite willing and able to discuss its merits and shortcomings.  If my approach of incorporating context into our work seems to fly in the face of ‘traditional’ modes of experimental composition, is that not in itself an approach that challenges the status quo, thereby qualifying it as ‘avant-garde’?

So, if the Ignite Talk taught me anything, it was that five minutes is a long time on a stage, and that even complex ideas can be distilled very effectively if faced with these constraints.  For me, the talk forced me to take a position on my art, and elucidate it in front of a broad spectrum audience that didn’t necessarily have a context for my talk’s content.  It was an amazing experience, and one I would highly recommend to anyone who has something they want to formalize, clarify and expound about their passion.  As the Ignite slogan goes ‘Enlighten us, but make it quick’.

I am currently a little over half way through building a training model for our BCMI (Brain Computer Music Interface) project, and it’s no picnic.  The goal is to select 50 2-minute excerpts of piano music in order to build an experimental model for classification by emotional signature.  This will eventually be used as the basis for EEG ‘training’ by the other performer in this piece, the illustrious Tim Mullen (Swarz Center for Neuroscience, UCSD).

The real trick is, given that I’m using genre rather than chronology as my basis, what do I pick as my musical material?  I’m currently at 7 categories, including the usual suspects (Baroque, Romantic, Jazz, etc…) but when we get to the late 20th Century things seem to become less obvious.  Where do we separate the serialists from the maximalists?  Many of them were doing both, and in fact at the same time as they were writing their parametric music they were writing beautiful melodies.  As the 20th Century drew to a close, there was a pervading perception that the purist attitudes of the avant-garde were not sufficient, and the new eclecticism was (is) being hailed as the savior of modern music.  More thinkings to come as the experiment progresses…