"Porthole streams music from your favorite music player to all your AirPlay speakers, in sync." Slightly cheaper alternative to Airfoil.
15 October 2012
I have totally failed to finish updating my vanity domain to be more than just a holding page, so my weeknotes for freelance work will have to start here, for the time being. And as for indexing them from zero – well, why not. And as for doing them at the beginning of Week 1 rather than the end of Week 0: well, I was out, and Late Weeknotes seem to be the trend.
A gentle week to kick off with, mainly focused on some business development (ie: having lunch or coffee with people), administration (setting up accounting software, picking project codename schemes, and beginning to maintain a pipeline) and working on the project I’m doing for the Royal Shakespeare Company.
That’s developing really nicely. I’ve still not written much about it at length because it’s evolving a little as I work on it. Still, rather than being cryptic, now’s a good as time as any to start talking about it.
The project is called Spirits Melted Into Air, and it’s a piece of work about logging actors’ positions on stage during performance and turning that data into secondary artworks. It arose from the idea that many people’s many interactions with Shakespeare are with the text, when, in fact, the RSC’s work is about performance: performance which is shaped not by a text (given how few stage direction we have in the printed versions of Shakespeare) but by an actor, a director, a motion coach, perhaps a fight choreographer, and (crucially) the audience on a given night.
I’m producing this data through a piece of software I’ve written – in Processing – which allows me to trace motion (by hand) from video. It’s a little crude, but is producing valuable results. Then, I’m writing more software to output that into useful formats, and turning that into art.
I am developing it in the open, albeit somewhat cryptically, over at Github, where you can find several Processing experiments and some diary notes. But really, it probably won’t make much sense until I write it up properly.
The big leaps forward this week were acquiring some source material from Stratford, whittling it down to size with some ffmpeg voodoo, and making my simple 2D demo work in skewed 3D space. Oh, and beginning the graphic design of the output.
Lined up for
next week (it’s this week really): two days design work, sketching and exploring an interaction space with the Good Night Lamp team, two days thinking about the RSC, a little business development, and Playful on Friday.
Good week 0, really.
MattB writes down his tips for language processing/machine learning; useful that somebody's done this.
"Riders pass through openings in a waterfall created by precisely monitoring their path via axel-housed encoders, creating the thrill of narrowly escaping obstacles." Brilliant.
This is one of the better sets of instructions for this – contains some useful commentary.
A Redis-based queueing system for ruby; much more efficient than Resque, from the looks of things, and really nice deployment configuration (ie: someone's bothered to write cap recipes and similar). Definitely going into the toolbelt, I think.