a public text box is a basic information sharing utility for a shared space. in this case, it’s a twitter account associated with a specific machine in a shared space rather than a person(who could be anywhere).

we tried this on macosx and linux. windows not tested but should work fine afaik.

1) download software and libraries

Processing: http://processing.org, Control P5 Library: http://www.sojamo.de/libraries/controlP5, Fullscreen library: http://www.superduper.org/processing/fullscreen_api/, Twitter Java library, twitter4j-core-v#.jar, http://twitter4j.org

2) the code https://gist.github.com/1267563

3) where to put things, this will vary slightly with your OS, but should look about the same. if you open processing and save the same sketch and then search for that it will probably be something like /Documents/Processing/sketch_023d

So you want to create a folder processing/publictextbox and publictextbox.pde will go in there. Then create /processing/publictextbox/code and controlP5.jar, fullscreen.jar, and twitter4j-core-version#.jar files will go in there.

file locations for processing scketch and libraries, will be a .txt file with keys in it too

4) register and get twitter keys, put keys into .pde code

5) test it out yo! do yer best debugginn.

6) make some keyboard modifications. maybe if you’re going to put this in a public place you should make a keyboard that has no ESC or CTL key so people can’t get out of the app once it’s running?

7) build a box. make it look nice. think ATM, public telephone, arcade game. get crafty and creative. i believe in you!

8) maybe you will have an auto shutdown and restart that will allow the installation to run long term without crashing as much?? see video for mac version of this. on linux it will be crontab. on windows the answer is ??

subscribe to the feed and watch the strange messages role in….
http://twitter.com/#!/publictextbox

could you make the same for something like a reddit/facebook/?? account?

some random notes about using old mac 10.4 machine…

http://www.webrelatedstuff.com/blog/ffmpeg%20and%20ffmpeg-php/ffmpeg-installation-on-mac-osx

https://developer.apple.com/downloads/index.action

Things you can learn to make if you want:
A videobooth, a public timelapse server, command line video remixing, computer timelapse self surveillance, youtube upload scripts, a public text box, and lots of good low level stuff about shell scripting and installing linux, systems thinking, etc.

What you should bring:
Your current computer or an old computer. Any old pc(win2k+) or mac (osx+) can probably be turned into a public information appliance. So bring one of those if you have it.
If you have an old webcam, bring that. If you have an old point and shoot camera from this list, please bring that:
http://www.gphoto.org/doc/remote/.

Where: Industry Lab coworking space. 288 Norfolk St, 4th Floor. Near Inman Sq, Cambrij.
When: 4 Wednesdays in November, 7-9pm. Nov 9,16,23,30

RSVP to dan:
broadcasterproject@plebiandesign.com

If you are interested but can’t make the time, you can still email and maybe some additional one-off sessions will be held at other times.

send us an email broadcasterproject@plebiandesign.com and we’ll make it happen in a jiffy.

thanks!

“face to face is the highest bandwidth”
for all our computer and technology advancements, nothing compares to the information bandwidth of face to face communication. this is what builds trust and long term productive relationships. so if that is true, why do all our devices make us spend more time communicating through a screen and typed characters?

“parallel vs serial devices”
a blackboard is a parallel media device. people can stand next to each other, both have write/read access to the surface and face to face bandwidth with each other. how would you design digital devices that have interfaces more like this? it’s so strange to see a classroom full of kids staring at individual screens or a train car full of people staring at their phones. phones and laptops are inherently serial devices. serial devices that tunnel long distances lead people to be disconnected from their local physical environments. like i am right now!

“copying is the advantage of digital”
plain and simple, the distinct advantage of digital is it’s ability to be easily copied. existing old school players want protection but all the young kids don’t care about that if you get to them first. a media company service based company based on public domain media flow, more like a utility. metered up, metered down. do whatever you want with bits once you have them.

“publishing and record keeping as a service to the marketplace”
craigslist(goods, services and property), reddit(conversation), ebay(goods), github(software), and facebook(conversation) are all good examples. also, courtrooms, stock exchanges and news wires as old school versions. the ability for individuals to publish and broadcast is powerful. how does this become linked to a physical place rather than a personal device? the digital nature still allows mobility but doesn’t require ownership or maintenance of part of individual.

“media and group witness”
reading a newspaper vs going to church or a sports game.

“freedom of speech and write access”

youtube channel=http://youtube.com/allasiatwvee

The scripts are similar to those used in the Timelapse memory server. It also has an audio component which is similar to the methods described in this post.

The machine is an old hp pc(~$140 from craigslist) with ubuntu 10.10 (free if you don’t count internet to download and cd to burn it on) on it. Webcam is a playstation 2 eyecam(~$40). I spent some money on lumber to build a box around it but mostly used scraps.

Files:
AVcapture.sh https://gist.github.com/834255
audiocombine.sh https://gist.github.com/834248
videoloop.sh

The machine also uses the youtube uploader stuff described in this post.

A twvee(twee-vee) set is sort of like a tv set except that it’s 2 directional. Maybe it should be a 2V set?

UPDATE:
using motion now to capture frames. other incomplete notes here

The basic idea is this. You’ve got an old computer lying around. You’ve got a webcam or something else that can take pictures. So you can make yourself an information appliance. A simple one function shared device.

Can a timelapse server for a shared space serve a purpose like a logfile for a shared software development project?

Supplies:
1. Old pc/mac that is sitting a closet somewhere.
2. Capture card + external security cam OR usb webcam
3. optional web connection

Our current system is an old Dell Dimension running Fedora 11 Linux. Should work fine with most any linux distro.
Set up the machine:
1. install streamer(xawtv), ffmpeg, imagemagick, scrot(optional)
2. install scripts

The script for making timelapse movies is here:http://gist.github.com/475719
File includes many old comments but it’s a rather simple script.
Basic functionality:

initialize to a daily directory.
while [ 1 ]
do
if (new day)
then {make movie with ffmpeg, reinitialize to new daily directory}
fi
write timestamped jpg to daily directory
wait 20 seconds
done


The script for playing back the movies(videoloop.sh):

#! /bin/bash
while [ 1 ]
do
#mplayer -loop 0 -fs -shuffle /home/timelapse/Videos/copies/cam*.mp4
# vlc works too
mplayer -fs -shuffle /home/timelapse/Videos/copies/cam*.mp4
sleep 2 # this line is to make it easier to break out of.
# esc to leave playing. ctl-c to break out of script which will restart after 2 seconds.
done


don’t forget to make the scripts executable.

chmod 755 *.sh

If you’d like the movies to upload to a youtube channel automatically every night, you can do that as well. Instructions here.

A sample movie from the system:

Another earlier overview that shows some more detail about the actual connections:

dell d630 running ubuntu 10.04 and cheese 2.30.1 [more info on specs]

Actual booth is not installed yet. Should be installed later this week.

Come down to the library and leave a video message of support as they are facing a $20 million budget cut. This broadcaster is focused around the idea of allowing visitors of the library a chance to voice their support and easily send messages to elected officials about the importance of the library. Lowering the barrier to entry for letter writing.

Video Overview: (which you can definitely skip)

INSTALL SOFTWARE:
FFmpeg (OS X howto), video utility, format conversions, etc.
Imagemagick:all kinds of image processing goodness. Like a command line version of GIMP or photoshop.
Sox: Audio processing.

We’ve used all of these programs on both Linux and Mac. Linux installs are pretty easy with package managers (apt-get, yum, etc). Mac is little more work for ffmpeg and imagemagick but not too tough.

[Optional] VideoDownloadHelper: Firefox plugin that allows you to download easily from lots of video sharing sites in mp4 and flv format at a variety of resolutions. There are Linux command line downloaders as well but we haven’t tried that yet…

GET SOME MOVIES Maybe you take them with your own camera, maybe you download from online. You’ve probably got a bunch floating around your hard drive already. Start with short ones. Feel free to grab some from http://youtube.com/brooklynmobile, or any other broadcaster, all are public domain.

EXPORT MOVIES TO FRAMES
mkdir frames
ffmpeg -i yourvideoA.avi frames/A%4d.jpg
ffmpeg -i yourvideoB.avi frames/B%4d.jpg
This will export individual frames of the video to A0001.jpg, A0002.jpg, etc
For longer videos you might need to use A%5d.jpg or A%6d.jpg but start with small videos first.

DO SOMETHING TO THE FRAMES and RECOMPILE MOVIE
This is where ImageMagick comes in handy. ImageMagick can do all kinds of image manipulations. We will give you a few examples.

EXAMPLES:
Blending two movies on top of one another. Copy the following code into a file called blurmachine.sh and make it executable[type chmod 755 blurmachine.sh at the command line].

#!/bin/bash
# boranj!
counter=10000 # initialize counter
for f in A*.jpg; do # step through frames
let "counter+=1" # increment counter
# apply some Imagemagick effects to the 2 images and write a new image
composite -blend 50 A${counter:1}.jpg B${counter:1}.jpg -matte ./mixedFrame${counter:1}.jpg
done
#Use FFMPEG to make frames into a movie again
ffmpeg -r 30 -sameq -i mixedFrame%4d.jpg ../outputmovie$DATETIME.mp4


Now you will have a movie with no sound that will look something like this:

ADD AUDIO BACK TO THE MOVIES
After using the above script file, you will have a movie with no sound. You can add the soundtrack of your choice, or add the mixed audio from the original movies.
To extract audio from the original movies and sum it back together:

# AUDIO EXTRACTION
ffmpeg -i movA.mp4 movAsound.wav
ffmpeg -i movB.mp4 movBsound.wav
# AUDIO MIX
sox -m movAsound.wav movBsound.wav mixed.mp3


Add audio back to your movie:
ffmpeg -i movNoSound.mp4 -sameq -i mixed.mp3 -ab 192k movSound.mp4


OTHER IMAGE MANIPULATIONS

Wavy overlay: Script file.

and similar example with more slices.
For this example, you need to create a mix image first. That is done with this line.
convert -size 640x480 gradient: -evaluate sin 20 wave_gradient.png

Place it in an appropriate location and make sure it’s correctly reference in your script file.
This type of combination has all kinds of options. You can take this a lot of other places if you wish to dig deeper. See this forum post and this usage examples page.

Four movie grid montage: Script File.

Six panel version of same effect.

Lots of other possibilities if you poke around Fred’s Imagemagick scripts as well.

Let us know what you come up with.

A few things you will need:
Gmail account and Youtube account. These will be linked. You can use one gmail account as the developer for multiple youtube accounts but each youtube account needs it’s own independent email address, and for multiple gmail accounts you might need to get some dummy address at yahoo.com or mail.com as an alternative email. Then get yourself a developer key while logged into your developer gmail account.

Python 2.6 is probably already installed. But if not, install that too.

We’ve gotten this working for MacOSX and Linux (ubuntu 9.10 and fedora 11 anyway).
It’s probably not too hard to get it working with Windows but we’ve never tried.

Files you need:
uploader2.py at http://gist.github.com/319734
edit login information for youtube and gmail accounts.
edit developer key. edit the keywords and description to fit your project.

videochecker.sh at https://gist.github.com/e2569bca9313b1056c0e
check the path.
Thanks to Sean Levin for writing these.

Gdata and Atom directories. They are zipped and downloadable here: http://www.divshare.com/download/11521457-3da Unzip and place into same directory as uploader2.py.

Copy these files into whatever directory you will be uploading videos from.
You will now need to edit both of these files as noted above. Do you have a favorite text editor for simple code? We like smultron on mac and gedit or emacs on linux but any will do. Whatever music you like.

make sure both files are executable.

chmod 755 uploader2.py
chmod 755 videochecker.sh
then try it out manually. from the video directory prompt try running the videochecker.
yourvideodirectory$ ./videochecker.sh
Got Upload directory
You have a save directory
Not running
just before loop start
inside for loop
Uploading OMPD-20100301-20100302.mp4
uploader-log: running doit with the_time=2010-03-02T00:00:53 and the_filename=OMPD-20100301-20100302.mp4
uploader-log: created media_group
uploader-log: created youtube entry
uploader-log: created YouTubeService
uploader-log: attempting to find previous login token in token file
uploader-log: token file found
uploader-log: set login token to DQAAAHUAAA ...
uploader-log: trying login with captcha_token=None and captcha_response=None
uploader-log: login succeeded continuing on to upload
uploader-log: saving token (DQAAAHYAAABXiMigRrgi5 ... ) to token file (youtube_login_token.txt)
uploader-log: starting to upload video entry
uploader-log: completed upload of video entry
uploader-log: OMPD-20100301-20100302.mp4 uploaded to service successfully. Returning success to calling script (0)
yourvideodirectory$


if you’re getting an error with videochecker, try changing the first line to
#! /bin/bash -ex and it will output lots of extra info as it runs.

Ok. So you made it that far and now you want to set it up so it runs automatically each day or week or hour. Time for crontab. What is a crontab you ask? It’s sort of like asking your alarm clock to run a program instead of make a loud sound. It’s just a simple text file that you add some instructions to. The only problem is that you usually edit it in vi (a text editor). vi is nice and all but it’s rather confusing if you haven’t used it before. With vi, there are 2 modes, insert-text mode and command mode. 3 most useful key sequences are:
i – enter insert mode,
esc – exit insert mode
ZZ – save file and exit program.
Want to learn more: http://heather.cs.ucdavis.edu/~matloff/UnixAndC/Editors/ViIntro.html

In the terminal window (here your current path matters not), enter
crontab -e
type i to put the editor in INSERT mode. (see text in bottom left of editor)

- then make your crontab look like:
# m h dom mon dow command
*/10 * * * * /home/yourlinuxboxname/Videos/Webcam/videochecker.sh

(which runs that script every 10 minutes.)
or

# m h dom mon dow command
* 3 * * * /home/yourmac/Movies/AUTOUPLOAD/videochecker.sh

(which runs that script at 3am each evening. But only when the computer is awake.)

- hit the ‘esc’ key to get out of insert mode.

- then type ZZ to save and exit.

Let us know how this works for you and we will try to keep an updated version of these instructions.