Hans-Christoph Steiner, about Solitude.
Hans-Christoph Steiner has accepted to answer some questions about Solitude.
Q: Damien Henry
A: Hans-Christoph Steiner
Where and when was Solitude played for the first time?
It started out with my spending lots of time at home using Csound to process samples of piano. I was listening to a lot of Duke Ellington at the time, so I was playing with samples of Duke Ellington. I was getting some really nice sounds, but I had lots of difficulty composing using Csound, so the idea was shelved until I learned Pd better. You can also download my sketch made in Csound, called “Solitude in Csound”. The completion of Solitude was actually a class project for a composition class I was taking. It seems that I need someone else to impose deadlines on me in order to complete a work like this. So the first public playing was at the end of the semester concert for the NYU Music Tech department. There hasn’t been other public performances that I know of. I suppose since I composed that piece sitting at a computer, listening with headphones, it seems somehow natural to me that people would also listen to it sitting in front of a computer with headphones on.
Why did you choose the Free Art License ?
I am a strong believer in the ideas laid out in the GNU GPL. What motivated the GPL is a bit different that want motivates me. Stallman and the Free Software Foundation focus on the ideas of freedom, which I do think is important. For me, the ideas of sharing are more important. If we do not share ideas, we are all the poorer for it. If we did not share ideas with our children, then would remain uneducated. If our children do not build upon these ideas, then progress would halt. This goes for art as well. I have seen so many interesting pieces of art locked up behind copyright and licenses, waiting for the artist to win the lottery of a record contract or things like that. I want to see those ideas out there and shared. When I was learning computer music, I learned a lot from the compositions that many people freely shared, mostly in the form of Csound source code. Back then in the early nineties, internet bandwidth was precious, and sound files seemed so large back then.
The other thing is that now lawyers like Lawrence Lessig/Creative Commons are trying to use copyright law for things that it was never intended to be used for. Copyright law, as Lessig describes so well, was intended to regulate commercial copying, and certainly not things like attribution, like the Creative Commons licenses do. There are long standing traditions of enforcement of attribution in the art world, we don’t need copyright to do this. On top of that, the law is really expensive and not accessible, so why would we want to bring the law into a system that doesn’t need it? I think the Free Art License best reflects these ideas. If you are interested in knowing more, I have written an article on this topic for the FLOSS+Art book, organized by GOTO10.
Are the Pd patches available?
Yes, you can download, run, and modify the complete composition. The patches are available under the GNU GPL, so people are free to play with it, remix it, etc. I would love to see some remixes or even completely new works based on this. (I suppose I should make the code a bit cleaner). In fact, this patch started out as a remix of Miller Puckette’s “sequencer” data structures example that is included with Pd.
Why did you create a graphical score (instead of using a text file to save all the parameters)? Is it only because it was the best or easiest solution, or are they some other reason?
Mostly, I wanted a composition environment that allowed fluid exploration of the musical ideas that I heard in my head. While no composition environment comes close to imagination, the environment I created based on Pd’s graphical data structures ended up being a pretty good environment for me to work in. That is, once I forced myself to stop programming and debugging, and start using the tool.
Is all the entire score visible or are there some hidden parameters ?
The whole score is visible, there is nothing hidden. And it is quite different from most scores because it is read directly and literally by the Pd program to generate the sound. That means that every time you run the program to interpret the score, it will generate the exact same sound.
What kind of musical parameters are linked to graphical ones ?
Yes, the overall color represents which sample is being controlled. Each data structure has two elements: a top, lighter color one which controls the playback of the original sample, and a bottom darker color one which controls the amplitude and panning of the playback. Time runs from left to right for the whole piece. Each sample and pan structure has it owns little “containing area”, you can see the thin, straight lines defining those areas. Those containing areas define the upper and lower limits of the parameters, like panning or length of the chunk of sample to playback.
For the sample playback, the bottom of that area represents the beginning of the sample, and the top of that area represents the end of that sample. Then the structure itself cues up chunks of samples to place at a given point in time. So if the sample structure looked like a very narrow line rising linearly from left to right, that would then just directly playback the sample. Changing the slope of that line changes the speed of the playback. More importantly, the height of the sample structure at a given point represents which chunk of the sample to start playing at that point in time. So if the sample structure is thick at a given point, spanning low to high, then a large chunk of the sample will be played starting at that point. If the line remains thick, then there will be overlapping chunks of sample playing. That’s where the sound smears come from.
For the bottom, darker structure controlling the pan and amplitude, it is easier. The thickness of the line is the amplitude of the sample playback. The position of the line in the area controls the panning. That’s it.
Is the look of the graphical score important to you or is it only a tool to create the recording?
When I was creating it, I wasn’t thinking in terms of graphic design. I was thinking in terms of a flexible way of representing the things I wanted to control. I didn’t really think much about the colors, beyond making sure they were easy to distinguish from each other. I think the colors look quite garish. The way it looks is important since it represents what the program is doing, but I think I could have made it look a lot nicer.
Do you think that the look of the graphical score influenced you during the composition?
The look of it definitely influenced me, since it was one of the made ways that I was understanding how things were working. My experience composing this piece was a constant feedback loop of changing the shape, then listening to the sounds, so after a while, I could really easily begin to read the shapes and hear the sounds in my head. There seemed to be a natural connection for me. It could be improved a lot, but that would require much more complicated programming.
Will you create some other graphical scores in the future?
I don’t have concrete plans yet for another project like this, but I definitely have something brewing in my mind. I would like to spend a good chunk of time first making sure that the composition environment feels quite fluid and easy to work with. Then I would like to focus on making the tool an aesthetic experience as well. Then of course, it needs to be bug-free. Once those bits are in order, then I would like to focus on nothing else but the composition I have in mind.
I have been thinking a lot recently about ideas of layering in music, and I have been gathering a lot of examples of pieces that use layering in interesting ways. I have a lot of sound snippets in my brain for how this next composition should sound, but I haven’t figured out what the source material should be. I might explore some synthesis to use, but also mix it up with samples as well.