ICS203B

surprisingly relevant to many of our recent discussions, it seems many folks in the UK have managed to drive INTO A RIVER because their GPS nav system directed them that way. despite the actual body of water directly in front of their car, they CONTINUED TO DRIVE INTO THE RIVER. yes, clearly there is more going on here than simply an issue with prioritizing digital information (such as, at what point into the river do these people think, "hey, maybe my car doesn't belong in a freakin river"), but it's still worth looking at.

from engadget:
"We've heard reports before about the dangers of driving while under the influence of GPS, but it looks like drivers in the UK have taken trust of their navigation units to the extreme. Twice in the space of the last two weeks, we've seen reports of British drivers taking serious risks because they trust the info displayed on the small screen more than what they see through their windshield. In the most recent case, drivers passing through the village of Luckington have found themselves landing in the River Avon, by following a GPS-recommended route that pointed to a bridge that has been closed for a week. Despite warning signs on both sides of the road, and nothing but water straight ahead, local villagers have found themselves pulling an average of two cars a day out of the river for the past week. "When you ask what happened, they say, ‘My sat-nav told me it was this way,'" one resident told The Times. Meanwhile, the village of Crackpot (yes, that's really its name) has had to deal with drivers whose navigation systems have directed them to the edge of a cliff with a hundred-foot drop. So far, there have been no serious injuries, but drivers have found themselves stranded on a rocky path. "It's only a matter of time before something happens," said one resident. Listen, we like GPS as much as the next sense-of-direction-impaired driver. But we also do our best to use our eyes, as well. And if you don't, well, there's a village in England named for you, and it ain't Luckington.

Read - Luckington

Read - Crackpot

Posted by Teh Greg at 06:29 AM | Permalink | Comments (0) | TrackBacks (0)

A general comment:
It may be problematic to try and create technologies that people have to adapt to rather than creating technologies that solve or ease a particular problem/task (ie have a technolgy adapt to the task).

The Active Campus project discussed in the Dourish, Barkhuus paper, specifically the Active Class project was designed to "broaden participation in class by lowering barriers of interaction" however according to table 2, even though students may have brought a laptop, PDA, etc to class, they did not necessarily use it to contribute to Active Class. I personally believe that asking students to contribute verbally to class discussions but at the same time an online discussion, etc does not help facilitate discussion. Our class for example we are expected to contribute to the class on the blog and during the actual the class (not necessarily to both at the same time). This seems like a better solution to what Active class aims to do.

One issue I have with mobile technologies is that users are given too many applications with which they use them for (to enhance social interaction), thus forcing people to continuously focus on the device, rather than people. This may hurt social interaction rather than help it. Though I hold strong opinions towards this I am curious to what others who own mobile devices feel about my concerns.

I have a problem with the fact that the Living for the Global City article focussed on people who graduated from "elite" universities. This as noted by the others but I still question why they solely focussed on three people from this class. Couldn't they have also focussed on three people from other walks of like? Also, I think a city such as Stockholm may also have been an ideal location to study.

Posted by ftsonis at 02:07 PM | Permalink | Comments (0) | TrackBacks (0)

Thinking about the conversation on Tuesday about the Pickpocketing game and the less than stellar responsiveness of its PDA's user interface, I wondered if the game would be as attractive if the interface was more responsive.

I came to a conclusion (based on the responses of Johanna and others and my own experiences in similar situations) that it probably wouldn't. It seems the faults of the game made it more interesting and deepened the experience for the user because they had to devote more of thier attention towards learning how to exploit its emergent properties. Which leads me to believe that the failure of perfect seamlessness between virtual and tangible interfaces can be potentially beneficial from an experiential point of view. Maybe having the seams is better if the target of ubiquitous computing is not purely enhanced productivity but a better productivity to pleasurable experience ratio.

Also, I think the Urban Probe paradigm seems to tread on thin ice as far as actively experimenting on people (in the "Lost Letter Technique") without their consent.

Posted by Rubikzube* at 01:50 PM | Permalink | Comments (0) | TrackBacks (0)

just a quick thought about some of this urban computing ubicomp work: who is the target audience? who are the intended users? from the Duruz and the Brown and Chalmers papers, it seems like the target audience is people who have lots of free time and the liesure of mobility; in these cases, older (presumably retired) women and tourists. This also seems the case with the Mainwaring et al. paper; the demographic they chose is more likely to be voluntarily mobile. in this case, there's a lot more to the paper, but I think it's an important trend to notice. Paulos and Jennings don't target this audience as specifically, as pretty much anyone in an urban area can use a trash can. however, trash can "stalking" #3, from which most of their data is derived, took place in the early afternoon, when people who are not voluntarily mobile will likely be at their jobs. even in the Barkhuus and Dourish paper, where they say that their population's locations and mobility are constrained by a schedule of classes, they are more or less incredibly mobile in between classes, choosing whether they will study in the library lounge, meet their roommate for lunch, or take a nap somewhere. going beyond these readings, a lot of ubicomp literature in the past several years has focused on gaming with mobile devices (e.g., the Barkhuus et al. from Tuesday). who has the time (and equipment) to play these games? young, urban "hipsters" with enough disposable income to buy the gadgets and enough disposable free time to play.

this isn't necessarily a problem; lots of technologies are developed with a very specific user demographic in mind. rather, it's something that should be explicitly considered in the description and design of these technologies.

Posted by Eric Baumer at 01:25 PM | Permalink | Comments (0) | TrackBacks (0)

first, i want to support sara's post. there are definitely some unattractivly naive statements in the paper. the paulos and jenkins paper had this bizarre thread running through it about emerging technological urban spaces; however, i think perhaps they intended this as a catalyst to make a statement that technology in urban spaces shouldn't be studied using traditional research approaches that favor the abstract over the concrete. i dont think they primarily inteded to say that technological urban spaces are new or not studied, but instead that the way they are studied leaves something to be desired.

(click ze link below)

i did happen to like several strands that echoed phil agre's words from "computation and human experience." ideas about a more situated ethnographic style that shys away from ideas of objective scientists whose observations come from somewhere outside of social culture. i like their motivation to nont only accept their influence as observers on the system, but also to exploit it. they mention how they philosophy bypasses classical design approaches, but i wish their motivation section had been longer (though probably inappropriate for chi). i also enjoyed the ideas that values such as efficiency should be not assumed but questioned.

on the flip side, a few statements about deconstructing the essence of urban life felt dangerously reductionist to me without being explicit about it. and that study of the trashcan was hopelessly lackluster in my eyes. the augmented trashcan was the most interesting part, althoguh im still not sure what i think about that whole setup. im just going to leave it at that.

i only have a few comments on the mainwaring, et al. paper. the idea of cacooning interested me. the approach that one was either cacooning or not seemed far to black and white to me. often my ipod is not a means to "escape" from the world but to reduce my boredom, or satisfy an urge to hear some music after a long day of quiet, controlled thinking. during a bike ride, why am i trying to escape from the city or not? surely there are times when one is doing this, but if the scale moves from "escaping the city" to "being immersed in they city," then the gradations along that scale are not discrete. not all "cacooning" actions are meant to remove the user from the world, and to create that type of binary distinction between these two extremes seems far too digital for my tastes.

also, the assumption that cacooning is anti-social (yes, they said could be seen as anti-social, but many have harped on this point) simply reveals that one has categorized it as "either youre part of the world, or you arent." i can't think of one time in my life when i completely shut off from the world. even when im asleep, someone calling my name loudly will awaken me. what about sharing a space THROUGH the technology used for "cacooning."

barkhuus and dourish wrote about the interesting concept of nomadic students and issues of reliability. its kind of interesting to think about reliability in the physical world. i mean, who would expect a trash can to suddenly not be there anymore? maybe if it's too full it isn't useful (though many people simple construct trash houses on top of trashcans despite the lack of space). barkhuus and dourish spoke about reliability in terms of nomadic students who can't rely on resources being present or not, and so shape their interactions and actions in light of that lack of reliability. what about switching it around? maybe people act less nomadically in order to preserve the reliability of resources. when i was a student i didnt go to the lab i worked in for social interaction, i went there because there was always reliable power and free water. clearly, people go to certain cafes or areas in order to preserve internet connectivity. i think this issue of reliability, not only power but also social and cultural reliability, has a significant impact on how we manage and move through the world. while this is probably obvious to everyone else but me, i hadn't really thought of it before. so there.

with regard to active class, i think the point about looking at the relationship between tech and local cultural practices is a nice one. active class took a standpoint of looking at social interaction through the lens of a professor, not the lens of a group of students. a classroom atmosphere is as much about education and asking questions as it is about facilitating inter-student interactions, groups and friendships.

Posted by Teh Greg at 01:08 PM | Permalink | Comments (0) | TrackBacks (0)

The Paulos and Jennings paper lists as part of its motivation, "This research investigation must begin before urban inhabitants acquire strong mental models and expectations from the current emerging suite of standardized urban applications."

The statement gives the impression that inhabitants are entering their environment naively, treating it all the time as a brave new world waiting to surprise them. In fact, people already have a host of mental models and expectations, both for the urban environment and for personal technology. I would argue that Paulos and Jennings's crucial moment has already passed, if it ever existed, and while it is certainly true that a city's ecology is continuously being remade - more rapidly than ever - pre-existing expectations persist through change. Technological expectations are merely overlaid onto this existing structure.

I'm intrigued by the idea of reworking these models to create a better fit, but the task should be approached knowing it's a reforming effort, not a creative one (in an ontological sense).

Posted by Sara at 12:10 AM | Permalink | Comments (0) | TrackBacks (0)

one question implicitly raised by the Barkhuus and Dourish paper is that of who does the analysis of a deployed system. on the one hand, you can have the the system architect/builder/designer/implementer perform the analysis. on the other hand, as is the case in this paper, you can bring in someone separate from the design team to perform the analysis. is either one necessarily better? what are the situations in which one is preferrable to the other?

I suspect that neither is necessarily always better than the other. furthermore, I suspect that there are not any general situations where one is preferable to the other. to be sure, each approach has its own strengths and weaknesses. choosing the approach is a matter of deciding why you are doing your analysis and what you want to get out of it.

when I was at the AAAI conference last summer, there was a group from CMU doing studies of a robot demonstration they had there. rather than doing the evaluation themselves, however, the group had brought on an anthropology student to study the way that people interacted with the robot. as I understand it, there's actually a large push for this study of HRI (human-robot interaction) with an anthro-/socio- emphasis at CMU. bringing others like this can be really beneficial, as outsiders won't have the same experiences with the technology as those who built it, having to explain the system to others can help the designer concretize their thinking, and those not involved in the creation of the system will bring in different biases than the creators.

however, this approach does have its problems. such non-CS types are not always available to do the analysis. furthermore, when they are, it takes time and effort to bring them up to speed on the system, describe to them what the intended locus of study is, etc. it also takes time for them to translate their analysis back into lingo that can be understood by the computer scientists. there's been some work done on the role of ethnography in system design (Hughes and Anderson come to mind), arguing variously that ethnography should not just be in the service of design. that's not really what I want to talk about it, but it's certainly related.

so, (someone clever apparently said), if getting ethnographers to do analysis for our systems can be so difficult, why don't we just train our computer scientists to do ethnography? indeed, here at UCI, there's a qualitative methods class, generally taught by the professor of this class, that attempts to give CS types an introduction to some anthropological and ethnographic methods. in general, this is great; bringing different perspectives to your work usually opens your eyes to seeing things in a new way and opens your mind to new insights. however, I think that a lot of CS people also realize that ethnography is really hard, in-depth, time-consuming work. it's already hard enough to design, implement, debug, and deploy the system, says the computer scientist, and now you want me to do all this ethnography, too? in trying to avoid all the tedium of ethnographic analysis, we end up with systems implementers doing "lightweight" ethnography, such as (no disparaging intended) Paulos and Jennings' urban probes, or people just doing qualitative observations and calling it ethnography. in certain situations, these methods, can be interesting, provocative, and occasionally lead to important topics for study and discussion. however, I believe that the majority of the time, they are used in place of more time-consuming, rigorous methods for qualitative, anthropological, or ethnographic study.

there's also a note to be made here about corporate anthropology (e.g., Intel's peoples and practices research group). however, (a) I don't know enough about this area, and (b) this post is already long enough.

so why does there seem to be this big push to turn computer scientists into sociologists, anthropologists, and ethnographers? I understand trying to bring new perspectives, and I definitely understand trying to give an appreciation for other disciplines' methods. however, why not, as is done here, bring in others to do the analysis of your system rather than trying to do it yourself?

as a side note, there's also the question of authorship. if I build a system and someone else does the analysis, who gets to be author on the paper? if it's a paper about the system with an evaluation section, do I credit the analyst I brought in? if it's a paper about the analysis of the system, does the analyst credit me? here, it seems that authorship goes to the analysts, and the system builders get a bunch of citations and acknowledgements, but I wonder if it could have gone down differently.

Posted by Eric Baumer at 11:12 AM | Permalink | Comments (0) | TrackBacks (0)

I liked the game a lot.
But I think that the idea of 'walking on mines' does not match exactly with losing coins.If it were a video game then the players may have been blown to pieces.
God forbid :)
I have another idea for using the mines. Every player should be entitled to inititiate another operation : blowing mines. They will blow the mines to destroy the coins that are going to be picked up by their opponent player who is nearer to the coins.This may intensify the competiton and enhance interaction between players.

Posted by Ishani Chakraborty at 10:51 AM | Permalink | Comments (0) | TrackBacks (0)

I have certain questions regarding Ticket2Talk and AudiSpeakID . I feel it though its a good idea to display the name of the speaker in AudioSpeakerID , but sometimes displaying on board may distract audience from the main question .
I am also not clear about how we can change our profile information within the conference if we are more interested in knowing some other topics in Ticket2Talk .Also, how we can make sure Ticket2Talk is going on handle multiple users when they are near the display screen. If a group if person arrives at same time, how can we distinguish which information is for a particular user .

Posted by Hironmay Basu at 03:28 PM | Permalink | Comments (0) | TrackBacks (0)

I've got a couple comments on the discussion points posted here, and then some more comments that I'll save for class discussion tomorrow. here are the questions I'm addressing:

T2T:

6. the best part of this system is that it links names to photos. hmmm, nametags anyone? as frank said, "these guys wouldn't have published a paper on nametags, so they made an extremely complicated nametag."

ASID:

1. Why are we reinforcing associations that a person is involved in that may jade our interpretation of their words? why not present their most recent projects or paper titles?

Crafting Participation:

1. interesting: getting to know the work as a satellite observer, watching others explore and discover it. getting to know the work through social interaction.

read on for my thoughts...

T2T:

6. the best part of this system is that it links names to photos. hmmm, nametags anyone? as frank said, "these guys wouldn't have published a paper on nametags, so they made an extremely complicated nametag."

in relation to a remark you had about the ASID system, yes, computer scientists have this tendency to use computation as the panacea (or sledgehammer) with which they attend to address (or smash) every problem across which they come. here, I agree with your other questions that they are trying to overcome their lack of social skills with some computational system. however, I think this is more than just nametags. in my experience, you can recognize a person's face at a much farther distance than that at which you can read his/her nametag. plus, it's uber obvious when you glance down at someone's nametag. while, as mentioned in the paper, there are certain norms at conferences that allow this, it is nevertheless quite awkward. I'm not saying T2T is wonderful, but I think it's more than just "an extremely complicated nametag."

ASID:

1. Why are we reinforcing associations that a person is involved in that may jade our interpretation of their words? why not present their most recent projects or paper titles?

I think this is very interesting. you're concerned about reinforcing your own biases, but what about the biases of the person who's asking the question? in order to understand a particular individual's approach, it may be quite useful to know their affiliation, especially since people don't often make their biases evident. we can't avoid having biases, but being aware of our and others' biases can help us (I think) make sense of our various perspectives and comments/questions on a given topic.

Crafting Participation:

1. interesting: getting to know the work as a satellite observer, watching others explore and discover it. getting to know the work through social interaction.

actually, I don't feel like this is really new. it's definitely interesting, and it may be that studying it in the context of a museum exhibit like this is new, or was in 2002. there's a paper by Amy Bruckman from 1998 called Community Support for Constructionist Learning, which talks about social behaviors in a MUD-type game designed to teach kids creative writing skills and simple programming concepts. one of the behaviors she found was that certain children would take on the role of mentor towards others; the "master" would do something, the "apprentice" would see this action and ask how it was done, and the "master" would proceed to teach his/her pupil. granted, this example is much more about explicit teaching.

another example that's not so much about explicit teaching is some of the behavior my research group saw when we deployed the EcoRaft project last summer at SIGGRAPH (you can check out http://orchid.calit2.uci.edu/~wmt/movies/EcoRaftUCI.mov for a video about it). occasionally, people would explicitly tell one another how the installation worked, generally if they knew each other before hand. more often than not, however, people would watch each other (strangers) interact with the system and learn how to use it from those observations. granted, that's a little different than what Heath et al. are describing, because their exhibit is about discovering some non-obvious functionality in the piece, whereas the EcoRaft is explicitly about teaching participants about restoration ecology. however, the things that participants were learning from each other were similar: the interactional properties of the system.

Posted by Eric Baumer at 12:42 AM | Permalink | Comments (2) | TrackBacks (0)

Do the systems in the "Augmenting the Social Space of an Academic Conference" promote interaction or do they actual inhibit it? By providing certain information on individuals, it may pigeonhole that person, causing others to focus their interaction on whatever information is given by the system or even avoid interaction because of bias obtained from that information. A valuable part of human interaction is discovery; as we interact more with one another we learn more about each other. Althought the AutoSpeakerID and Ticket2Talk may serve as “icebreakers”, I can see people becoming too reliant on the information presented instead of exploring other areas of shared interest.

Posted by Doug Kim at 10:08 PM | Permalink | Comments (3) | TrackBacks (0)

frank and i have worked out the following topics for tomorrow's discussion. please read em over before tomorrow and make some posts, damn it. =)

in the spirit of saving space, the body of the post is contained after the jump (translation: click the link below).

Picking Pockets on the Lawn:

1. what was lost and what was gained by using the PDA as a focal point as opposed to using it as a reference tool?


2. the PDA was a tool that primarly functioned as a focal point, and was seldom moved to the periphery. the focus of the game was on the digital not the physical.


3. Why did they prioritize digital actions over physical (e.g. pick pocketing and running away was not an effective strategy, but pickpocketing and shielding was)?


4. Why force the user to reconcile the world with the digital world (i.e. 'spy looking' to check to see if the information on the PDA was an accurate representation of the world)?


5. over time people were less collaborative. oooooooook. what might have caused this? perhaps because the physicality of the game was a novelty, not a priority.


6. the only interactions that were truly meaningful were digital. not only were digital actions more effective, but simply moving outside the digital space made all subsequent physical interactions trivial.

Situated Displays: T2T

1. while it tries to promote social interaction, T2T probably does just as much to hurt interaciton by allowing people to be passive voyeurs.


2. how weird is it to have digital image of the people standing right in front of you splayed on a large display?


3. isn't it actually detaching a person from their identity (despite the claims to the contrary)?


4. so, we make this huge display that compensates for our social inadequecies. screw learning how to introduce yourself to your colleagues you haven't met, just look at the sign! actually, it's quite dehumanizing.


5. how important is the dynamic element of this system? why not just have a static rotation of attendees that changes pictures every 15 seconds? oh right, because if you're standing near the display and you see someone on it, they might be standing there too. or they were there 60 seconds ago.


6. the best part of this system is that it links names to photos. hmmm, nametags anyone? as frank said, "these guys wouldn't have published a paper on nametags, so they made an extremely complicated nametag."


7. i proposed a more socially relevant visualization (though still clearly problematic, it at least doesn't attempt to make up for a lack of social skills): what if the displays just showed the groups of people that were standing right next to each other? we could see who was talking to who, in what groups, which might invite one to join or avoid groups depending on those involved in it. at least its not a big, expensive nametag.


8. on a positive note, the tool is useful in the sense that one can use it as a focal point and gather information without further action, or the user could work through the tool using it as a point of entry for conversation (assuming they already possess adequete the social skills). can move attention easily between ready at hand and present at hand.
   

Situated Displays: ASID

1. Why are we reinforcing associations that a person is involved in that may jade our interpretation of their words? why not present their most recent projects or paper titles?


2. the system takes a very computer science approach to solving a 'problem' of people not hearing a person's name or affiliations. hey look, we had a loss of information! let's put up a big display that reduces error, taking a lossy activity to a non-lossy activity.


3. tends to distract from the person, again placing attention on computers instead of humans.

Crafting Participation

1. interesting: getting to know the work as a satellite observer, watching others explore and discover it. getting to know the work through social interaction.


2. a system design that, through discovery of its use, leads to an understanding of how it works.


3. by moving focus between the art and the other people, one can learn about both people and the art.


4. no clear distinction between people and the piece. how does this contrast with the pda from picking pockets?


5. with picking pockets, collaboration was supported through use of the PDA, whereas in this piece collaboration was enabled through the social interaction with the support of the artwork.


6. tim & mary: the properties of the interactive piece were transferred and tested on the non-interactive piece they were looking at.


7. how does the artwork move between focus and periphery? is it a quantizable measure? who could measure it?


8. on that note: how does the idea that the work move between focus and periphery affect the interactions of everyone involved? is it the individuals or groups that move its focus, or is it both? all of the above?


9. what might the authors mean by the difference between in and through interaction with the art? in and through = ?

still consolidating topics from the excluded spaces paper. this should be enough to get some people posting.

..hopefully.

Posted by Teh Greg at 07:02 PM | Permalink | Comments (0) | TrackBacks (0)

Going through the discussions last week and chapters in the book , my mind was intrigued with certain questions I would like to post on the blog. We divide ubiquitous computing strategies in inch ,foot and foot depending on the size of the device . I was wondering if the distinction is clear and we can represent all devices and application models as to me , devices like digital desk technology can be somewhere between foot and yard as we use differnent application models to represent it. With Reactive rooms clearly moving out Computers from box on desk to environment and getting involved in relationships , It becomes pertinent to me to explore designs with more generic applications rather than being confined to one application itself and how designs could be extended to other settings, adapting to changes in the relationship between environment and activities in it. Are we being able to interactionally equate physical and digital media? One important question in social computing is how the different incompatible approaches with different sets of assumptions and commitments have helped us in understanding social action . Is there some underlying equalities in all of them. We all being embodied, all our social and physical actions being embodied,how the embodied interactions is meaningful. I hope to search through the answers of these questions in the coming week

Posted by Hironmay Basu at 03:01 PM | Permalink | Comments (0) | TrackBacks (0)

I have a bunch of random thoughts on the readings and the class discussion from Tuesday so I thought I'd simply vomit the contents of my brain on the blog.

First of all, in response the the debunking of Cartesian dualism in class, (and bear with me because I'm totally obsessed with "low" technology and "ancient" schools of thought and medicine), there is a tendency in Buddhist thought to be aware of the incredibly strong link between physical sensation and mental state, and also the fact that through training it is possible to divorce one's mental state from the constant stream of sensation that flows into it. So while I would agree that Cartesian dualism is not a natural state of existence, with sufficient attention and practice it may be possibe to achieve something that may resemble it. Why one would want to do that, I have no idea... just a thought.

Also, I was wondering why the natural human drive to create causal relationships between events wasn't really addressed in the readings, although it was probably just assumed. In any case, all of the approaches to tangible interaction in the readings seemed to rely on this fact and quite a few of them seemed to exploit the power of this drive novelly in creating relationships across the virtual/physical divide. I just think that it's interesting that people seem to want to create these relationships inherently.

And finally, I was really intrigued by the notion of sources and sinks in Ishii's paper, and how it may be possible to build on that and treat computation as a "source" that one "sinks" into an physical artifact to create a computational device. In essence, instead of embedding the computation into a single device, make it embeddable into a slew of devices, each of which manifests the computation differently.

Posted by Rubikzube* at 01:48 PM | Permalink | Comments (0) | TrackBacks (0)

"Tangible Bits: Towards Seamless Interfaces between People, Bits and Atoms" - Ishi, Ullmer

I agree that manipulating physical rather than virtual objects is more natural and intuitive. It is not surprising however, that Ishi's first tangible works consisted of ideas being extrapolated from virtual paradigms back onto physical ones. Could Ishi have jumped to an entirely new physical paradigm without this intermediate stage?

"Getting a Grip on Tangible Interaction: A Framework on Physical Space and Social Interaction" - Hornecker, Buur

"too many tangible interfaces aim for direct one-to-one
mappings, remaining literal and missing out opportunities
for employing magical metaphors or for providing the user
with computational re-representations of information [26]
and transformations of input (highlighted by the theory of
distributed cognition [16, 23, 25])" (Hornecker and Buur, 4)

I wonder if this approach would actually work with task-oreiented scenarios. It seems logical for an arts interaction, which is trying to portray abstract meaning which can be interpreted differently by various individuals, but abstracting specific tasks can be problematic. As with the Clavier, users are using their entire bodies but essentially are just triggering "buttons" - there is no real complex movement being interpreted by the system.

"From Interaction to Participation: Configuring Space
through Embodied Interaction" - Williams, Kabisch, and Dourish

"We also noted a woman who put the compass up to her ear, as if expecting the sound to emanate directly from it. These were the most noticeable illustrations of the general tendency to focus on the physical objects as the source of the sounds and regard the digital system as transparent. Universally, when a participant’s attention was attracted by a sound associated with a certain object, they turned not towards the physical source of the sound – the speakers – but to the causal source of the sound, the object." (14)

I wonder as we try and apply computation onto physical objects, as we are accustomed to actual objects emitting sounds, rather than audio feedback coming from somewhere else, how we can design systems to make participants understand the goal of a system, without explicitly stating it. Obviously with this work, exploration was intended. As noted by the authors, this correlation of sound to action is confusing if auditory feedback is coming from somewhere else, but their intentions were to create a system to use objects to collaborate with a system as a whole. I guess i dont really have a point.

Posted by ftsonis at 12:55 PM | Permalink | Comments (0) | TrackBacks (0)

i just want to set down a few brief reactions and comments.

first, i have to disagree with you a little bit eric, not completely tho, in your statement that "the ways in which the physical properties of a tangible interface can take on social meaning aside from their primary function" is something that is vastly overlooked. i would say in many related disciplines it is something that's pretty strongly analyzed [as i am reading sharon traweek's beamtimes and lifetimes just now i see another example of that: the way various dectectors take on meaning for physicists, and also they ways they display meaning, such as the style of the group who built them]. in any case though i see two interesting aspects here: one is how do certain artifacts affect fine grained interactions, and the other is how do they fit into a broader social whole. as an example, what does the way in which i'm using my cell phone right now say to you as another person in the same space, and what does the general trend point to overall. these questions i think have been looked at in various settings, but, here is where i agree with you eric, often they are a bit overlooked by HCI proper, especially in the design of new technologies. often these two categories of social meaning, which are not the direct purpose of the device, are not addressed when technologies are designed or evaluated. as greg was saying, ishii et al. seem bent on packing more "ambient information" in to a space without regard for the "side-effects" [which i would argue are just as important as the intended effects] of their designs. without consideration of the broader picture, current cultural metaphors, etc. the designs seem a bit detached from real interaction.

now - all that said, i want to briefly touch on what i originally wanted to post on. eric sort of opened the door for me in asking about the alternate types of meaning which artifacts can embody and engender. what cut across most of these papers for me was space/spatiality. but often when we tackle these topics from an HCI perspective, we sometimes ask how can we design technologies to fit in a certain place... but more along the lines of what Williams et al. and to some extent Hornecker and Buur touch on, i think it is worthwhile to ask what objects residing in a place can tell us about it. Williams et al. focus more on the varying levels in which the objects can be used to explore a space; here though, i mean more to ask how a place, already populated with all sort of objects, technological or not, can tell us something about a place. though this venture would not be an explicitly HCI one, i think for the design of new objects for a place, it would be a worthwhile venture to try to examine what it's current character is via the things that are already present. taking a look around from this very low level could potentially open our eyes to considerations often missed by the broader view. just maybe ;]

Posted by roxy at 01:04 AM | Permalink | Comments (0) | TrackBacks (0)

this post is mainly focused on issues i had with the ishii/ullmer paper, though the williams/kabisch/dourish paper begins to open up and explore questions related to those i pose in the comments below. apologies to the other papers. perhaps ill comment on them later.

in the tangible bits paper, while i like the idea of exploiting user expectations in an interface, i'm concerned about the attempt to satisfy the idea of generality using a tangible interface. the personal computer has become a general purpose machine, such that its impoverished interface controls are meant to manipulate the vast space of applications the computer can run. what's nice about non-general-purpose, physical objects is they only do certain things. a toaster makes toast. all of its controls are specific to making toast. you can toast many things, but in the end you end up toasting something. the metaDESK seems to attempt to be a general-purpose machine with a specific-purpose interface. they state that "Tangible Geospace" is “an application that runs on the metaDESK platform.” it isn't clear what this implies. is the metaDESK platform meant to run multiple applications, as the personal computer does, or is it something else?

(please note the link below if you wish to continue reading. i didn't want to take up the whole page with my commentary.)

if the metaDESK IS a general-purpose machine, attempting to design a general-purpose tangible interface seems like it would re-implement the bottlenecks of the mouse/keyboard/screen interface, albeit in a somewhat more 'natural' way. attempting to design an interface that meets the needs of hundreds of thousands of applications is already difficult using the standard gui + mouse + keyboard. expanding this to a tangible interface significantly increases the complexity of this problem. what physical artifacts will be general enough for me to manipulate all types of applications, but specific enough that i can use my real world knowledge to operate them? maybe have a bucket of dials, knobs, levers, switches, flashlights such that i'm supposed to figure out when and when not to use each? why limit the potential and substantial power of physical interface controls by an underlying commitment to generality?

if the metaDESK is NOT to be a general-purpose machine, the interface that was designed specifically for interacting with a map is deliciously arbitrary . i don't think i've ever found a physical object and placed it on a map hoping the map would reorient itself around the object; however, i have looked at a map and tried to trace a route along it with my finger, or folded the map so i could look at two distant location side by side. so, how exactly is the "tangible geospace" satisfying my expectations of how it should be used? it's true i often use my hands to pick up objects when using a map, but not in the ways their application permits. I realize the goal was to build something that drew away from the desktop metaphor, but they also comment on how satisfying a user’s expectations is important. i feel it simply does not accomplish this.

it seems a better tangible interface could be designed if it were specific to a certain application. while the metaDESK has left behind a traditional mouse/keyboard/screen, it has brought along with it this crippling notion of a general purpose machine. williams/kabisch/dourish state that they "are inter-
ested in the ways in which the migration of computation into the everyday environ-
ment might reconfigure the relationship between people, objects and space." it is this interest that i feel should have been an important part of the ishii/ullmer paper, but somehow is not.

//possibly unwarranted and utterly bizarre rant about ambient displays
the ambient display ishii and ullmer talk about (water ripples representing website traffic) seems to not only hide the digital nature of computation from the world, but also support the notion that information is separate from body and can be redisplayed using any medium. it does nothing to elucidate the constraints and limitations of digital computation, but actively seeks to hide away those limitations by displaying digital information in a completely unrelated analog form. showing information in an abstract form of water ripples seems dangerously separated from the computation that underlies it. the metaphor of the relationship between the physical world and the digital is subtle and arbitrary here. suppose, instead of an empty water tank we had silicone fish such that each fish swam and turned for each hit the website received. they argue the information was presented in a way that allows for peripheral absorption, which I agree is a good idea; however, from my slight modification to their ambient display, all of a sudden the digital computer world seems mysteriously life-like in its behavior. Of course this is not the intent of the ambient display, but often intent is quite easily disregarded once technology is absorbed into culture. as opposed to pushing toward a more public understanding of the inherent constraints and limits of digital computers, this type of display would reconfigure the digital and present it as something non-digital, perhaps even life-like.

where the metaDESK made explicit the connection between the physical and digital through interaction, this ambient display seems to subscribe to the notion of software/hardware duality. that is, the information from one system can be excised and placed into the body of another, preserving the information content in the new medium. The medium is NOT critical, and is, in a sense, an afterthought. priority is ascribed to information, not the body. while clearly ambient displays are an attempt to present digital data to analog humans in a manner that is more 'friendly,' i believe it's important to think about the metaphors this technique embraces. the computer has used the tank of water to fill in gaps in computation (its lack of an ability to 'poetically' display information in the periphery). This brings to mind the way humans were harnessed to fill in the gaps in computation for the SAGE system in the 1950's. i just don't like the idea of harnessing the world to supplement the inadequacies of some computer, regardless of how philanthropic the intent.
//end rant

//ok, maybe more rant…
later, the group begins applying this duality to the metaDESK in the discussion section. they suggest that buildings could cast shadows that are not a function physical parameters, and provide the example that a shadow would represent "the research publications or sponsorship inflows of various buildings." it is this type of physical/digital separation that begins to creep into the work covertly. on one level, the relationship between a shadow and the flow of research publications into that building is not grounded by any of the user's "expectations" the authors so fervently wish to accommodate. on another level, this idea that information can be moved about using arbitrary relations to physical objects harkens back to the ambient displays rant. a simple change would transform this problem into a virtue. for instance, in stead of basing shadow generation on an arbitrary connection to digital data, perhaps we might use the light as one often uses light: to illuminate the details of an object. in this sense, if we shine the light on the building, its surface becomes bright and facts and figures about the building appear on its structure. in this way the light acts to reveal data in a slightly more expected way (more light usually allows us to see more detail). the light simply shows us more detail about the building. i've never used a flashlight to look at the shadow of an object in order to find out more about the object. if the game is about exploiting user expectations and building them into a physical interface, this section seems completely out of place.

Posted by Teh Greg at 07:54 PM | Permalink | Comments (2) | TrackBacks (0)

I'll go ahead and comment on what seems to be an interesting theme running through parts of these readings.

chronologically, it comes up first in the Ishii and Ullmer paper, in the very last paragraph. "When his mother kept household accounts, he was aware of her activities by the sound of her abacus, knowing he could not ask for her to play with him while her abacus made its music." this is interesting, as it takes a non-central aspect (the sound of the abacus) and gives it a social meaning (for Ishii, that his mother is occupied with household finances). the ways in which the physical properties of a tangible interface can take on social meaning aside from their primary function is, as far as I know, not a topic pursued further by Ishii's research group.

the social aspects of tangible interfaces are raised also by Hornecker and Buur, particularly in their discussion of what they term Spatial Interaction. "full-body interaction," they say, "acquires communicatie and performative function," that when interacting with a system, one is also interacting after a fashion with all observers present. they relate this to a case study of CLAVIER, an inherently performative interactive device. additionally, I think this performative aspect may also apply to interactions that are not inherently, explicitly, or intentionally performative. bringing in ideas from Ishii's abacus example, the act of typing on a keyboard emits a very distinctive noise, which lets everyone within earshot know that typing is going on. however, there is also a performative aspect in this task, as well. as I type this comment, my lab mate sitting across the room from me hears my keystrokes and knows, in the periphery of his attention, that I am hard at work. thus, there is a performative and communicative aspect to this interaction with my keyboard, even though the interaction itself is not intended to be a performance.

Williams et al. comment on such socially tangible (does this term make sense?) activities of the participants in their system. not only did participants "learn" how to use the system by observing others (moving through iconic, instrinsic, and instrumental interactions based on their observations of others), but they gained peripheral awareness of each other's activities through the sound scape and monitor each other with looking at each other.

I've not yet gotten to the Benford et al. paper, but I wanted to get this posted now. I'd suspect some of this stuff comes up there, too.

it's curious to me that this is such a widely known open question to the research community, and yet it has not really been addressed. Hornecker and Burr comment to this effect, but they don't offer any suggestion as to why. I'll venture a guess and say that the social aspects of tangible interfaces are caught up with the ways in which social groups repurpose and/or appropriate (depending on your preference of terminology) artifacts, a process of which we don't really have a great understanding. I'm not sure that exploring one will necessarily elucidate the other, but I suspect that a solid understanding of both will be necessary to really adress this issue.

Posted by Eric Baumer at 04:38 PM | Permalink | Comments (0) | TrackBacks (0)

I can imagine a couple of ways that things might go from the tangibility stuff we're talking about this week. One topic to proceed to would be to explore the collaboration angles a little more; another would be to look at a different scale and talk about some of the recent work on urban computing as a ubicomp domain.

Vote via comments...

Posted by jpd at 03:24 PM | Permalink | Comments (2) | TrackBacks (0)

The image of "Sparky" to which I repeated alluded in class yesterday is
here.

Posted by jpd at 09:29 AM | Permalink | Comments (0) | TrackBacks (0)

I've set up this blog in order to create a space for discussion outside of class meetings for ICS 203B this quarter.

The primary reason to have it is to share seed our discussion sessions by getting some questions, reactions, and responses to the readings circulating in advance of the class meeting. The discussion leaders will manage the in-class discussion, but the in-class discussion should, ideally, both draw upon and lead to a broader conversation that we can carry on here. So, you should check here regularly, and post, too; it's all part of the same consideration of class participation.

The second reason to have it is as a place to share anything else that might be of interest -- URLs, news items, and other material that might be relevant to ubicomp and interaction issues. One thing you might want to look at is the LUCI (Laboratory for Ubiquitous Computing and Interaction) web page, at http://luci.ics.uci.edu, which collects some interesting materials.

Posted by jpd at 10:34 PM | Permalink | Comments (1) | TrackBacks (0)