This talk begged for a backchannel discussion on sciviz but I am not sure that the larger group of students is up for the task or interested in it. If we were a hardcore research school, these folks would hear an earful about their project. Perception in sciviz is a big topic right now because we can finally make an attempt to look at gobs of data but we don't know how to judge what we are looking at. Earthquake data has been visualized but we don't know what we are looking at. The example of Labanotation by Peggy is a little far fetched because Labanotation is more of a documentation attempt for dance than choreography therefore by default subject to interpretation. On the other hand, for years we fantasized that visualizing scientific data will help us solve problems and it simply isn't true.

Just look at typography and graphic design. And let us not forget our good friend Tufte who collects good and bad visualization examples. Sure they are not real-time animations of data but the same problems exist. We have to find ways to weed through what is important and filter the data. And still we don't know if what we are looking at is right. After all, data collection methods are not absolute: they are also subject to filtering methods, calculation and storage. And how many times will designers hear from scientists: "But my data can't be wrong!".

I remember when visualizing curves on scientific calculators was marketed to us by a math teacher as the next best thing that would make everyone learn math easy. I wish I could meet the guy again, slap him and tell him that he told the same crappy story with numbers on the whiteboard and the same crappy story with math on the calculator. He was simply a bad teacher...

Good and bad data, good and bad visualizations, good and bad interpretations...they can happen to anyone. Don't let that be you...

I think my biggest concern with the "perceptualization" of immense amounts of real-time data is the UI. For example, the musification of port activity is a fascinating idea, and I can understand how it might be easier to notice certain activities or changes by listening rather than looking. However, I know if I were listening to that music for more than 10 minutes I would tune it out as background noise. Same with watching the space dome. Maybe I'm just not the right kind of person to have that job, or maybe I'm just untrained. Still, how do you make significant/meaningful/suspicious activity catch the user's attention without blowing it out of proportion? In real time?

Marientina and Jess, you guys make good nuanced points about perception and data. We have so much we're tracking these days, there's ample opportunity to study or manipulate so much seeming certainty.

I must admit, I was pretty much completely and pleasantly floored by this presentation. Particularly the rendering of data as music. I thought that was transcendent - the ideas of windchimes of the internet, taking data and making something audible of it. I think it was Bolas who raised the issue of rendering a few times - if you make the data look like flowers, people are prone to like it or respond to it in unnatural ways.

These men didn't seem opposed to that - in fact they seemed delighted in their audio/visual data fiesta. I was left wondering, what kind of personal data could I mine, and what kind of a show could I make of it? As you said Jess, most data streams you're likely to tune out, no matter how pretty they are.

I had a friend Don Pearman, a contractor in the Bay Area, who carried a fire department scanner along with him at all times. He was a fire nut - if he heard the right severity of call, he would drive over there with his cameras in his trunk and take pictures of the flames and firefighters.

Most of the time, the radio was silent, or a low chatter in the background. But he could filter it out. So what data would I want to hear constantly, that I could learn to respond to from slight audio cues? I can't think of what personal data, but I think that's my own mental limitation.

There are increasing numbers of sensors produced and only a matter of time before I'm able to measure, say, my blood-sugar level with sound and hear when I might be starting to crash from lack of nourishment. My Aunt Lori wears a pedometer so she can make sure to take at least something like 70,000 steps a week for exercise and health. Sensoring herself! Attach that to a GPS, attach that to a map, attach that to Patholog, attach that to the tornado-data maker and run that through a projector and your walls will seem like some kind of carnival you made with your footsteps. Huzzah!

Crappy story....hum. Maybe it's because of that the pure scientific fun cannot be easily appreciated by ordinary people--like me--who have not the necessary related knowledges. I don't know. Maybe there're other reasons. The midline blows away only after enough charges by both sides. But before that, misunderstanding, prejudice, set rules, suspicion historically separate two parts. Who knows--Maybe someday musification becomes more than the complements to the verbal system e.g. sound identification of public signal system, aural identification of CI.... if humans are to be trained and ruled from their childhood on. Isn't that we do have successful data visualization?

I admire scientists' modest appreciation on the "beauty of the purity of mathematics & simplicity of the physics" etc and understand most of their immediate ideas of medium utilization come along that fashion. However, what's new from the artists' experience? Any difference or new message?

(It seems engineers know more potentials of the emerging technologies--actually, they do. If the right sphere of their brains are likewise developed as do artists, it becomes easier to make choice in terms of the "artistic" kind of MEASURE & PROPORTION of medium utilizations for them--much much easier than others.)

Justin, the blood-sugar musification idea is creepy but promissing. GPS Attachment... I hope you'd brought out during the Mobile class. Hum, yes, forecast musification may also work agreeable than a talking head--at least, I'm long tired of years of focusing on that talking head in that small light box.

I got to hear the speakers in my morning class as well, where they went into the methodoloy and testing behind such visualizations. A large and very hard part of this work is proving that such visualizations actually help cognition...and they found some interesting points in this matter. For instance, they found having the displays in a 3D environment is easier on the brain than, say, switching between two windows that are on top of one another. Still, finding such elements, and proving they work beyond anecdotal evidence involves long and costly experimentation.

Tracking a lot of the space exploration going on right now (Cassini, the rovers, the space telescopes, etc.) quite a bit of wha we see is also "visualized"...some of the most famous photos are under under infrared or ultraviolet light, for example...but such images show a great deal more than visible light. There's still an issue of confusion, and taking a visualization too literally.

That said, I've been finding a lot of visualizations works of art in and of themselves. That's also a fine line to walk...the best examples cannot compromise either the art or (of?) the science.

What Jess bring up is interesting. I think a lot of the time, the form that the visualization takes is something that deserves heavy consideration. Its beautiful to turn the activity of a network into sonorous melodies, but why is that how the network activity sounds? I guess the intent there is with the artist.

The system they devised, Flatworld(?), was a great platform for information visualization development. I especially liked the structure that acted as a sort of heartbeat for the AI. However, I think that using a platform like that in the end might be too formally imposing for my taste. I think a lot of information visualization would be more successful with a lighter economy of information. Not everything needs to be in a 3D immersive environment.

One information visualization designer who I particularly like is Martin Wattenberg. His work is playful and beautiful. I particularly reccomend his Shape of Song project, but there is pretty much nothing that isn't good at his website, Bewitched.

Oh... I guess a can't use html tags in the comments. Well for your information:

The Shape of Song: http://www.turbulence.org/Works/song/

Martin Wattenberg's Bewitched: http://www.bewitched.com/

wish I could have stayed and seen this talk in its entirety, but I had a large project to finish. That said, what I did hear intrigued me a lot. I'm personally interested in metaphorical representations, allegories and visualizations.

I'm not sure, but it seems like there's confusion that comes into play when people consider scientific methods as too singular a path to empirical discovery, or that every visualization has to meet a standard of usefullness. Some correllations are of course wrong or unprovable, but the amount of insight and possible routes for investigation that result from visualization outweigh the problems. You have to root out the useful ones, which I think are pretty few and far between. Transdisciplinary understanding is important, particularly in a time when so much postmodern outlook divides and marginalizes individuals.

I think visualizing data is for practical usage to understand complicated information more intuitively and faster. When I saw the works presented by the speakers at first time, it seemed to be difficult to understand what kind of information I can get from them. They seemed to be rather aesthetic than informative. But after watching it more, I got to know that how many things I could directly recognize through what I see. Sometimes letters of our language seem to be too abstract to express certain things but what we see is directly comes to our brain and the resolution of our eyes' visibility have big bandwidth of deriving informations from even really detailed change of what we see. That reminds me of the article recently I read about a savant who's trying to explain his mysterious ability. He said that when he gets an answer of two 3digit-numbers multiplication in a second he actually said he can visualize numbers being calculated inside his brain and it's so natural for him. I think he implies the power of visual language and its potential. Visual language is borderless and doesn't require any translation. Personally, I think I became more appreciating how efficient it is studying here as an international student.

The music was great. It really made me groove.

God help the company that uses such a system and employs me to overlook network security. I'd be busy grooving to the resultant synth jazz solo while some guy is busy hacking the system. Okay, maybe not that bad, but...

Yeah, you definitely need a composer to reinterpret the possibilites of music. Working within that sort of minimalist motif, I definitely can hear a sort of soundstage of the picture set. Maybe it was specifically working within the realm of genres, so that each type was represented by...classical? Romantic? Jazz? Or maybe having totally different tonal centers. Maybe it would be easier to differentiate between Dorian mode and Japanese pentatonic...

I don't have the CS background as some of my peers and unfortunately did not uderstand most of what two of the presenters were talking about. So, my apologies on a seriously weak "report" this week.

That said, the music part of the presentation was much more tangible. I was intrigued by how the randomness of it all didn't sound like musical throw-up.

Thanks for signing in, . Now you can comment. (sign out)

(If you haven't left a comment here before, you may need to be approved by the site owner before your comment will appear. Until then, it won't appear on the entry. Thanks for waiting.)

URL:
Remember me? YesNo

Comments: