Generative Music the 2nd

In my search for Generative Music artists I came across two very interesting blogs. The first is lawriecape.co.uk. This is ideal for what I'm researching into, he's a Flash developer who has played around with Generative music and Generative art in general. He even wrote a midi flash plug in which I'm currently investigating.

He has also done work with other artists, for example this is the result of a collaboration with Joshua Sadler:

Musicgen2 by Catfurnace

It is this quality of art I want to achieve in this project.

Generative Music

Although I should be slow to tie myself down to a specifics at this stage I had a look at some Generative music pieces. Brian Eno s one of the pioneers of this style of music. He has made a big impact with his music apps. His apps are very interactive, touch based affairs, take for example Bloom:



This is an interesting approach to generative music but not the one I'm looking at at the moment. I also came across another iphone app called ANTracks on the ACM database. '... describes a generative music composition application on a mobile phone. Music generation is a well-suited application domain for multi-touch and reality based interaction. Thus, we choose to implement our concept on Apple's iPhone/iPod touch. Based on selected harmonic scales the user creates a musical expression by controlling a number of virtual ants. The movement along pre-defined patterns allows users with little artistic talent to create harmonic expressions. The mobile device can act as a multi-touch input device that controls an electronic synthesizer program on a desktop computer or as a stand-alone application.'

This again relies on a touch interface but the ants can wander by themselves (according to an algorithm) and only trigger certain notes. It is these two aspects I am interested in. Originally my idea was to reflect the emotional feeling of twitter but that would be rather restrictive. After research into digital art the best pieces are interactive. So I'm now leaning towards the same idea (art from twitter feeds) but instead of restricting the field to predefined words I'm considering letting the word that's being searched for to the user, so instead of searching for 'sad' they could search for 'john', either way, art will be created.

Art

The main purpose to my proposed project is art. In order to avoid the danger of getting to caught up in the technical issues (ie seeing the trees but not the forrest) I decided to look into digital art. I came across numerous interesting artists and exhibits. A number stand out though, David Rokeby was definitely a trail blazer. San Marco Flow is my favourite, it offers almost an alternative view of reality and how we perceive space. However his installations are huge in scale and possibly beyond what I could pull off by myself.

Another interesting artist was Golan Levin. His site is called FLONG and also has other collaborators on it. This has numerous exhibits, eyecode is one of the most interesting.

It might be possible to incorporate an interactive aspect to my project. This would probably add to the appeal of the installation and merits further investigation...

Moving Music

SInce I set out one of my aims as reflecting the positive or major tone of a tweet via music I did some investigation into music. Why does music move us? According to most sources, the main reason is social conditioning. Minor scales are used widely in eastern music and don't sound depressing. In western society however we associate sad and minor and major and happy. For example how does the following make you feel?


Whereas a major key makes you feel good or happy:



There are a number of other things to take note of though, tempo for example, even a sad song sped up doesn't sound as moving. Take for example the stirring Moonlight Sonata.. sped up four times.



now compare that (especially the intro) to normal speed:



The difference is staggering. There is a very interesting piece on this on exploaratorium. So not only does the scale or key matter but the tempo comes into account as well. This is definitely worth bearing in mind with this project.

Twitter-to-Flash Proof Of Concept

After some research to see how I would go about getting this information and a couple of tutorials, I created this Flash file. I called it TwitSearch.

For simplicity sake I'm looking for Birthday on twitter as this is a term mostly used with 'happy', however sometimes people prefer tweeting 'enjoy your birthday' or the like. This code displays HAPPY if happy is used and SAD if it is not. Obviously the system used for the project would be far more advanced (and presented better) but this is just part of research. However it works a charm and can be viewable here: Tonys Simple Twitsearch Flash File.

Twitter to Flash

Rather than be twiddling my thumbs waiting for the arduino to arrive in the post I've been researching other ways to get data from twitter. Twitter restricts requests to 150 per hour. This leaves you with one request per 24 seconds (3600/150). Obviously one note per second would sound rather dull. Therefore each update would have to play a series of notes, this changes the project slightly.

With regards to getting data from twitter, one way to do it would be Flash.

Luckily flash is very helpful in regards to this, or twitter is to be more exact. The

SWX Twitter API 

is explained in good detail, the site defines it thus: 'SWX stands for SWF Data Exchange Format. It’s a new, native, data format for Flash that uses simple SWF files to store and exchange data.' It also has a tutorial I am trying out currently.

That's all for now,

                             Tony

Data From Twitter

Doing some research into how to grab data from Twitter for use with Arduino I found this project on displaying twitter info on an LED.



This article is particularly useful as the author had very little experience with python before the project. This is somewhat of a relief and it also includes the code used in a handy Zip File...

Hear The Web

My main idea for a project is an art piece /  museum installation that reflects data from the internet via audio. To do this simply I am breaking the audio up into to sections major (positive) and minor (negative). Inspiration for this comes primarily from This project. In it there is a table that changes colour depending on the tweets using the hashtag #ledtable. Obviously for this project a different hashtag would have to be used, for example #news or #today. Instead of changing lights I want this to play a different audio file (let's say for example from the key of A) if it is positive it plays a major note or negative it plays a minor note. In this way someone would be able to hear the overall mood of the web (I'm fearing a lot of minor notes in these times for news!).

However after some searches on Twitter.com for #today a more equal balance seems to be available. The way this project would work is that it reads a twitter feed and if there is more words from a positive array than a negative it would play a random major note from the key of A. This might not sound very pretty but imagine for example if some horrible disaster was to occur and most the posts would use 'dead', 'sorry' and 'pray for' then the music would take on a dark tone and music can be very emotive so the user would get an instant emotive response even before knowing why they are feeling better or worse. I think this is an interesting dynamic that has some real artistic and perhaps even social merit. 

Tony