Links for Wednesday

Samsung unveils a 7" flexible LCD screen

Somone at UIUC pimps' out their bed

Discussion about Astor Place on Curbed

And the folks from Dodgeball came and gave an inspiring talk at ITP last night.

Got links you want to share? Send them to me.

Today, Jenny, Karl and I researched and finalized our Physical Computing final. We settled on the idea of using the segment displays to create a form of digital graffiti. The idea is that people will be able to change the letters on the displays and bring new contextual meaning to the 'tags'. This means that we're looking at using 16-segment displays instead of the 7-segment display's we've been working with. We've also set up the beginnings of a documentation blog to capture our research and findings.

Interactive stick figures on Engadget

I think the Castiglioni emulation kit is my fave.
[via Core77]

NY Times article on Zaha Hadid (NY Time login required).

And the Sound dome project site has been active over the weekend.

If you've got links you'd like to share, send them to me.

Today Jenny and I worked on our Physical Computing final. We've been working with seven segment LED displays. First, we got the Allegro shift register (a.k.a. the bit shifter) to write serial data to numbers on four separate seven segment displays. Then we got our MM5450 LED display driver up and running. It isn't multiplexing yet which means it's not doing what we need in order to drive the dual displays but we're moving in the right direction. If anyone has experience multiplexing with this chip I'd love to hear your experience.

Have a great Thanksgiving.

Fasticon released iSipme Icons, a freeware set for personal use.
[via macnn]

Sound related: Babble attempts to drowned out neighboring cubicle noise.
[via archinect]

Now you too can fight aids@home with your spare CPU cycles. This is a great cause put together by some amazing people. If you aren't already donating idol CPU its definitely something worth looking into.
[via engadget]

Hulger released a few very nice Bluetooth enabled retro telephones.
[via engadget]

And apparently Long Island City is the place to be (NY Times login required).
[via curbed]

For the Spatial Design final we're proposing a sound installation in the hallway at ITP. The general concept it that sound, and what people hear can make people more aware of the space that are in. The sound domes (custom speaker systems) will be mounted from the ceiling in secession through the length of the hallway. The sound will travel from speaker to speaker down the length of the hall. Min and I recently completed the first prototype. The group is still debating audio content. You can learn more about our process and documentation one the project website.

Some older and newer keyboards:

optimum
Bluetooth Virtual and the ergodex

And, an interview with Jonathan Ive. I think crediting him for 'inventing' the iPod goes a bit too far.

Over the weekend I built a simple counting device using a switch, a PIC chip, and a seven segment display. On the PIC, I used edge detection to trap the button press event. It's been interesting learning about physical interaction, especially coming at it with my background in software interaction design. Its funny because I tend to think in terms of button press events and software constructs. The ideas are perfectly valid in terms of an approach to learning physical interaction. However, there is a bit of translation. Or, it's possible that software based interaction is abstract in the sense that there is no physical button so, we use terms like 'button press event' to describe the action of pressing a button.

Learning how to use edge detection is important because often times you don't want to know if sensor is on or off, rather you want to know if the sensor was triggered or reached a certain value. The documentation and PIC code are up in the Physical computing section. I also posted the tricolor LED and DC motor labs as well.

The Processing code I used for tracking video in my PCOM and Comp Media midterms is now online. The code requires a video camera. I've included stills too. For the PCOMP midterm the video feed was analyzed and used to calculate the percentage of change in each cell of an invisable grid. If the threshold for change was reached in that cell an LED was illuminated. I also recently updated the Spatial Design section with current work too.

Yesterday, Adam Asarnow, Rocio Barcia, Josh Knowles and Charles Pratt presented in Applications of Interactivity (Red Burns participatory lecture series). If you're at ITP you already know that each student presentation is a reaction to the guest speaker from the previous week. So yesterday, the group presented a reaction to Clay Shirky's presentation from last week.

Clay's presentation was very interesting, focusing mainly on information structures. Specifically, how old models of categorization (like the Dewy decimal system) breakdown when attempting to categorize large volumes of digital information. He also pointed out that these systems rarely withstand the test of time. This is why we now have categories such as 'Former Soviet Republics' in the Library of Congress. Old models are ridged and do not accommodate change. I do agree with this point. It would be redicilios to look up United States history under 'Former Colonies of the British
Empire' His point is that users shouldn't have to guess what somebody else thought they might be looking for.

The solution Clay described is: allow users to 'tag' their own content. By empowering users to define their own categories, there are no pre-disposed structures which the information must fall into. Examples of systems where this has been done effectively are Flickr, delic.io.us and Google. The use of 'tagging' content items has grown in popularity over the past two years mainly because the system is flexible, relatively simple, give everybody a change to contribute and provides amazing new ways to slice and dice the content.

In Applications class, what the group presented was a web site they created which allowed users to tag other class mates. I could go in and add words which I felt described myself or other people in the class. This brought up some very interesting points. The first has to do with the validity of dynamic categorization.

Because the tagging was done to student profiles, the tagging only represents a snap shot in time of when that person existed in the way when were described (for example, if I were described as nice and next year I became cruel and bitter, the information would be invalid, and, this is by no means the fault of the system, rather a mere fact that people change over time). But this does raise the question: Can tagging work for time based content?

My immediate reaction is yes, but, there needs to be some kind of indication of when the content was tagged this way. Possibly, current tags could hold more weight, or have tags expire. Would this even work? I'm not sure, possibly.

In the end the site was a very large success. The site expanded exponentially with user tags, validated Clay's point about user driven data structures and made a very interesting presentational. However, last night Josh took the site down. People were telling lies.

I could see this one coming, I just didn't think it would happen this fast. Because the systems did not require users to login there was no social accountability for who posted the tags. Therefor people could poke fun at, and in some cases make things up about other people. Social accountability goes a long way. Look at eBay for example. If you're bad, you get bad ratings, and no one will buy stuff from you. You're held accountably for what you do by the community. However, in this case there was no form of social accountability, and therefore people could do and say what they pleased with out being help accountable for their actions.