Eclectic Tech Carnival Weaving Networks

Launch Lounge Playground
Venue: foyer STWST
days: thursday, 12th - saturday, 14th
time: 10:30 -14:00
facilitators: b#, kriztina, magde

LAUNCH LOUNGE

the playground knows no onedimensional knowledge but includes approaches,
knowledge and the creativity of its participants only.

the playground provides things, technics and materials for free use. the
difference to a workshop is, that we want to encourage to simply play with
things without the fear of not knowing or doing something 'wrong' - in the
playground the categories of right or wrong don't exist anymore.

every day we set a topic, a goal that we want to reach by evening. the process
of doing is more important than the outcome.
at the end of each day people present what they have been doing during the day
and how they were trying to reach the set goal.

located in the foyer of stwst you can also just drop by and check out what's
being worked on currently or participate from the beginning on and use the
equipement creatively

we take the playground literally and just play. the only thing you have to
follow is a common goal, set in the morning.

equipement that is there for use:
computers, microphones, mixing panel, picture camera (foto and video)

participants: maximum 6

TOPICS:

- working with text - how to generate and transform text
transform texts into something else: other texts, movies, comics, pictures....

- how to stream text?
a lot of things are streamed nowadays. but how does this work? and have you
ever seen streamed text? how can we do that ourselfes?

- voice, speech and language
speech without language and language without speech - how the fuck can we do
that? stop over and find out!

LAUNCH LOUNGE - REPORT

Launch Lounge was a concept for playing rather than working and learning through playing. It was on purpose going away from the workshop concept of a teacher in front and an audience who is taught but was rather driven on invention and creativeness rather than knowledge.
The three days that we were doing (Magde, Kristina, b#) turned out to be fun and we all learned a lot of things about streaming servers, unknown software and developed ideas about our topics that we could not work out all. This is a short overview about the things we did.

Day 1: Transforming Text

Day 2: Streaming Text
When we talked to a few people they asked if chatting actually would not resemble a kind of streaming text. But for the brave three of us this would have been way to easy! So head over heels we tumbled around in the internet and discovered really interesting tools, that could guide us closer to our aim.
First we tried out a tool called TBT – the time based text. It turned to be out to be a tool mainly for programmers to record certain commands and record the keystrokes at the same time. Hmm – interesting but how could we use that?
We came up with the idea afterwards to try to make every second a screenshot of our desktops to stream those while we are writing. It ended in a bash script taking care of the screenshot (thx didi!) but the transform into video was quite tricky though and did not want to work due to well-known linux compilation problems (damn the mjpegtools!). If it would have worked out how we imagined the jpeg snapshots would have piped through to the jpeg2yuv which would have made an mpeg stream out of it and then further piped through to the oggfwd and ffmpeg2theora to an icecast streaming server. so much for the theory - the praxis didn't work due to problems within the jpeg2yuv.
Anyway – we did not want to give up. So we actually fell back on a tool from the day before – the freej tool and voila! It's not really a real streaming text – but it was well enough. The first file below - streaming-text.ogg - is the part of the recorded file from the stream.

This is the bashscript (picstream.sh) taking care of the screenshots of my desktop:

#!/bin/bash
for i in $(seq 0 100); do
import -window root ${i}root.jpg
done

Commandline theory of piping:
./picstream.sh | jpeg2yuv -It -L0 -j abc_%04d.jpeg | mpeg2enc -f3 -o stream-text.m2v | oggfwd --> ffmpeg2theora [icecast server]
the oggfwd and ffmpeg2theora we could not try anymore! so check back for right syntax....

These are the linx to the programmes we checked out:
http://tbt.dyne.org/ - time based text tool
http://freej.org/ - the v-jing programme with which we actually streamed the text (you can easily connect with that to a icecast server!)

Day 3: Language without speech, speech without language
On the third day we were trying to think about this topic. We didn't make it though to discover language without speech although some things came to our mind like sign language of course or whistling languages. Gloria told us about the japanese Ponsori singing style – it's without words and rather tragic. There is not much documentation on the web though unfortunately.
So we started with languag without speech – which can be retrieved over various ways. What we tried were two attempts:
cutting up an existing audio and taking out just consonants/vowels. And the speech synthesis mbrola which works with various language databases. You can listen to the outcome in the files below:
lws.ogg - pure consonants taken out of a speech file
in-test.ogg – transformed file thru mbrola. The .pho file was created with german database and it was sent thru the hindu language database.