Oberlin workshop and MTS algorithms

Darren Kelly (kelly@x4u2.desy.de)
Fri, 17 Oct 1997 15:06:58 +0200 (MST)


Dear Richard Povall and dance-tech,

> I told this group some time ago that I wanted to organize something here at
> Oberlin in 98. We're looking at the possibilities for early fall 98, and
> probably aiming more towards a workshop with lots of active participation
> for a small group, rather than a large conference. The large conference,
> of course, will be happening in February 99 at Arizone State, and I hope
> that outcomes from my smaller workshop might feed into that event.

Thanks for telling me about your planned workshop. At least I chose
the right address for my suggestion. I can see that - as a newcomer to
your community - it'll take me a while to find out exactly what has been
done and planned; I'll do my best to find out as much as possible about
work already done and planned by others, and dance-tech is certainly
helping there, but I'm sure to ask some silly questions and to propose
some old-hat ideas.

Which brings me to the point of my planned list of possible MTS
algorithms: performance idioms achievable from input signals measuring
motion. I'd be grateful to learn whether you know of such an existing
list, especially published. It's difficult to find time to do literature

> Now, that we would want to concentrate on the heavy duty issues surrounding
> hardware sensors (although as some of you know, my own interest lies much
> more in video-based sensing) seems unlikely. I do see the workshop being a
> mix of sessions of technique, technologies, aesthetics, and hands-on work,
> though.

Although my own intrument employs transducers, I'm interested in video
and other methods too; the sensor implementation was just the easiest
to realise with limited time and funds. Indeed I looked very closely at
video methods from the outset, and other technologies. These didn't seem
suitable for achieving one goal of my application (drumming by dancing),
but have advantages for other performance idioms. Certainly my concept of
MTS algorithms is not restricted to any particular "front-end technology".

The only requirement for an MTS algorithm is that the input signals are
somehow related to motion. The whole idea is not to concentrate on the
rightly denoted "heavy duty" issues of sensor hardware, but rather the
possible mappings between the measured quantities (irrespective of
the hardware) and the output, i.e. signal processing algorithms.

I'm certainly interested in attending your planned workshop, time and money
permitting. If you see fit to have me contribute there in some capacity
(based on the CMJ article(s) in preparation) I'd welcome the invitation.
To judge by the pictures on the Oberlin web page the campus is beautiful
in the fall.

As for my own suggestion of a workshop with an emphasis on the general
mathematical description of these methods, I'll wait to see what the
response is to my planned CMJ paper(s) (see below). I've approached
Applied Mathematics at Sydney Uni about my developments, and hopefully
I'll get a part-time position researching and teaching there. (The rest
of the time I'd like to dedicate to my own work and music). I'll also
approach the Music Dept. there, but to my knowledge these techniques are
not well known there. If I do get such a position, maybe I'll have the
opportunity to host a workshop there (always difficult, Oz being so far
away from everyone). David Rodger performed his work on his MIDI Corset at
La Trobe University in Melbourne. Maybe he'd be interested in coorganising
such a workshop.

CMJ papers and MTS algorithms

I've decided to split my CMJ paper into two parts, one concentrating on my
experience with accelerometers - and the algorithms to which the property
of acceleration lends itself - the other paper concentrating on MTS
algorithms INDEPENDENT of front-end technology (sensors/cameras/tracking).
It is to this second paper I invite contributions, news of existing lists
etc. Please do mention, however, the input and output devices if you wish
to contribute.

I also understand that these algorithms form the "tricks of the trade"
for future performances, and that people might hesitate to reveal all
before a performance, so feel free to restrict your contributions to
methods already demonstrated publicly. The aim is to review all the
possible mappings used to achieve performance idioms in a unified



Darren Kelly | | |
| \ 0 / |
DESY -MPY- (Deutsches Elektronen-Synchrotron) | * o |
Notkestr.85 | - D E S Y - |
22603 Hamburg | o * |
Germany | / O \ |
| | |
phone: +49-40-8998-4569 | |
fax: +49-40-8998-4305 <--Note change | |
e-mail: kelly@x4u2.desy.de | |
Amandastr.40a, | HOME |
20357 Hamburg | zu Hause |
phone: +49-40-4322602 | The PLAYhouse |