Location Based Disservices

A rather curious way to use my coordinates

Using one of my most prized iPhone apps, MLB At-Bat the other day, I was struck by the above. This was a curious bit of crappiness. I had heard something about this app using location info, but I wasn’t expecting it to use my coordinates against me so literally. Clearly there are *legitimate* contract rules for blackouts and all that hubbub, and I guess having this requirement, if it means live games, is worth it.

Perhaps in my naïvety I’ve been sitting here thinking about using earth-coordinates to enhance some kind of mobile experience, when clearly there is this whole other realm of Location Based Disservices. How many other strange and anachronistic restrictions are out there that could be LEVERAGED in such a way? Certainly there are countless others in sports.

A few obvious ones spring to mind:

• Our iPhone NFL app can force us to watch the Raiders lose every week.
• If we are in Utah, we can be prohibited from downloading episodes of Big Love from the iTunes Mobile store! (second that for the iBeer app).
• Warning: all calls placed from California can not be routed to South Carolina

Abstract Reality

[image by flickr user Drewje]

In preparation for Where 2.0 next week, where I’ll be on a panel called “Mobile Reality,” I wanted to jot down some recent thoughts on the matter, both as a way of helping me better understand what I might say next week, but also a kind of beginnings of a more deliberate effort to think through some of the projects I’ve been involved with recently, both at work, and on my own stuff.

(I should mention that all of what comes next is entirely in the context of my personal work, and in no way is endorsed by or has anything to do with my current employer.)

I guess Mobile Reality is an attempt to define some different kind of Augmented Reality (AR) that is about focusing more on real spaces and location awareness, and stuff that runs on mobile devices. (Although AR in general I think is based around the same ideas idea of “enhanced” mobility). So perhaps this new variation is slightly redundant, except maybe to contextualize this stuff as being available on “today’s” mobile devices instead of on some crazy setup with a 100 pound backpack full of gear. In any case, I’m going to just refer to this stuff as Augmented Reality, and completely ignore the whole hologram side of this, where you show your computer’s camera a AR Code, and some 3D model is magically displayed on top of the code. That’s just called magic. [note: see this]

So most instances of Augmented Reality don’t interest me a whole lot. Julian’s been writing about a bunch of overlapping ideas, so I’m not going to spend much time discussing why I’m not 100% on board for this particular vision of the future. As He writes,

All [this] kind of stuff that would turn spatial experiences into some kind of database inquiry seem very much different from what I enjoy about the world when it is mixed with humans — curious interpretations of objects and moments that are not salted with uniform resource locators, pop-ups, soft synthesized voices telling me that I’ve got mail or to turn right at the next intersection.

What he’s getting at here is important, at least to me. The entire idea of many types of AR is to ostensibly make our lives easier, to help us learn about our environment. See Wikitude for a pretty insane example of how this kind of future might look. But in this entire process of becoming more at one with the world through this technology, the world becomes entirely mediated by a computer. It literally becomes a simulacra of itself, as though we weren’t close enough already. In some cases, AR requires use of special glasses that beam the real world into your vision through a camera mounted on the glasses, effectively taking the human part of you completely out of the physical world that the system is supposedly augmenting.

There’s a kind of Uncanny Valley effect that goes on in these experiences. The more that reality is augmented, the less real it becomes. The similarities to the Uncanny Valley here is not surprising considering that AR is a descendant of virtual reality, which itself was an offshoot of computer graphics, where the term comes from.

Clearly in this genealogy, there is a tendency to try and make virtual things real. AR I think is a little different in that it aims to (by definition) make the Real World itself more real, or at least more defined and accessible. In all cases, the inclination is always, “wouldn’t it be cool if…”

Ok, so that’s enough of that. But what I, and I think probably lots of other folks are interested in, is the idea of applying Augmented Reality to virtual spaces. This is the idea of lots of pervasive games and stuff, layering virtual worlds on top of the real world. This concept video from HP is a pretty archetypal version of what this kind of entertainment looks like. A kid, using his HP device, sees a virtual world layered on top of the real world. He’s running around, dodging virtual enemies in the streets, navigating the normal, real world which acts a kind of game field. So this is kind of one of the promises of AR as well, a way to create mobile and pervasive game spaces that blend (or mix) realities.

I think this stuff is pretty cool, in general. I think everyone is on board with the idea that your real life can kind of turn into a game at any moment, your cubicle or whatever turning into some kind of evil lair. If you haven’t seen The Game, or heard about Majestic, maybe you should check those out if you are interested in this kind of thing.

So, most of these type of things are based on this idea of spatial correspondence. If I’m in a park in the real world, there is some virtual space I’m walking around at the same time, that shares the spatial characteristics of the real world. The virtual world is the real world, reskinned, with enemies and obstacles that conform to the same constraints as you would in the real world. The HP video linked above is full of examples of this. I was also part of an early attempt to do similar things (albeit much weirder). Using spatially correspondent worlds makes sense I guess if you really want to turn the real world into a game board. Paradoxically, this kind of Augmented Reality really does more to connect you with the real world, forcing a more complete connection with physical space as you navigate the virtual layer.

In trying to define this kind of stuff, I’m realizing that the kind of stuff that I most interested in is similar to the virtual worlds stuff, but much less concerned with spatial correspondence.

I’m really compelled by the idea that mobile games or other experiences can be used to further connect people to physical environments. At the same time, I’m realizing that my work, and the work of others that I find the most interesting in regards to this, is really experimenting more with the idea of Abstract, or Abstracted Reality. To me, this is more like using aspects of the real world as game pieces rather than using it as the game board itself. There are any number of components that define the real world, from concrete objects like trees or buildings, but also less tangible things about how we interact with the real world like distance, speed, if this house is my house or this bar is where I meet my friends on the weekend.

This is one of the nice things about foursquare, for instance. On the face of it, it’s a type of mobile social network where you can see where your friends are. The thing that’s interesting to me about it though, is that it begins to use some of these more abstract layers of space (call it place, if you want) in a game context. For example, if I check into my bagel shop 3 times in one week, I unlock the “local” badge, or if I am at my local pub more than anyone else, I’m deemed “the mayor” by the game. In this way, dennis and naveen are using a more abstract concept of the physical world, one that connects me to my world a lot more than seeing some pop up overlay of the pub menu, or a little sound blip that tells me where the nearest place to get beer x is. (clearly, it’s getting closer to the beer part of the day…)

It’s probably also useful to look at how traditional games go about getting players to figure out the game environment. Games are often all about figuring out an environment, whether it’s a puzzle, or figuring out how you can get up to the coin on that high platform. But that’s probably enough of this for now.

Where 2.0!

Scheduling note: I’m speaking on a panel at this year’s Where 2.0 Conference, which should be a lot of fun. The panel is based around the idea of Mobile Realities, and I think a lot of it will be talk about how people are pragmatically creating alternative realities using the mobile platforms that are with us today instead of the laptop+backpack+differential GPS setup of days past, present (and unfortunately) future. Instead of dreaming about circa 1987 views of the future like 100% spatial corresponding “mirror worlds,” I hope we can focus more on what people are actually building now, and and how they are using various constraints with the current systems create cool new interactions and experiences.

If you want to come (and you should), use the code: WHR09FSP for a 25% discount!

Constraints and Innovation: LBS Edition

For just about as long as I can remember, the top item on most mobile/location nerds’ wish lists has been access to real time, always on GPS information. For many years, this dream was stuck in *5-years out* purgatory, mostly taken up by researchers wearing robot suits or less-than-casual citizens geared up with specialized PDA hardware.

At some point, the N95 came out, which allowed developers to access GPS information to some extent. But the N95 was still super expensive (relegating it to PDA status), and developing stuff for it was still kind of a pain in the ass. Then some version of iPhone software came out which used the Skyhook Wireless SDK to allow your phone to have a flavor of GPS based on WIFI triangulation. Even later, the v2. iPhone came out, which had full GPS integration. But we were still not satisfied. We need GPS apps to run in the background! To make matters worse, the Skyhook data wasn’t accurate enough to really track us in real time. Jeez!

But it’s been interesting to see some of the stuff that’s worked and not worked with this constraint. One thing that immediately becomes clear is that LBS apps tend to be more simplistic and less grand in their ambitions. Not being able to track every GPS coordinate you stumble upon, at every instance, has forced people to think a little harder about exactly why they want to use location information in the first place. Another thing that the constraint highlights is that apps that are designed around “where am I now” persistence don’t function very well. The Brightkite app is an example of this, as I would imagine other “here now” social location apps. Instead of embracing the constraint of an articulated location, chosen at the discretion of a user, these apps ignore it, hoping that in the future the always on gps will fix the problem. This may well work for them, but it’s a failure in terms of getting people engaged in the service in the short term.

What’s more interesting is an application like Foursquare, which focuses not so much on what latitude and longitude you are at (the space) but on the venue, event, etc. you are engaged in (the place). This works much better in the context of an articulated location mechanism. If I want my friends to know I’m at this bar, I can open the app, find nearby venues, and then tell the system which one I’m at. This may seem like a process that, using the powers of technology, could be easily automated. But when done right, a system like Foursquare I would argue functions better when this level of interaction is required.

Clearly there’s more examples of where this works and doesn’t, but frankly I’m running out of steam (way over 140 characters at this point).

And, of course not meant to dismiss the idea that persistent tracking can lead to some really amazing applications, stuff that lots of us have been scheming on for years. Locale, for Android (which supports background processes), is one early idea that has a lot of potential. Instead, it seems warranted to revisit the idea of articulated location, and look at it not simply as a bridge towards full, persistent tracking, but as a valid interaction scheme in it’s own right.

area-code business

Collected some area code data & put it into json format. This is for personal purposes, but it might be useful if you are trying to some some super simple area code lookups.

ex:

{"zip":"203","country":"United States","state":"Connecticut","cities":["Waterbury","New Haven","Bridgeport","Stamford"]}

{"state":"Utah","zips":["385","435","801"]}

• areacodes.txt
• zipsbystate.txt
• zipsbycountry.txt

Elephant Paths in Animal Crossing

Not really sure the extent to which this happens in other virtual environments, but one of the really great new aspects of the new Animal Crossing game is it’s support of Elephant Paths. The basic idea of the Elephant path is a visible path, formed by “unofficial” movements through space. In this sense, unofficial is defined as outside of established path infrastructure like paved pathways or roads. In many virtual worlds, Animal Crossing included, there aren’t many official pathways to begin with. The larger the world, the more frustrating it can be to navigate. While the world of Animal Crossing really isn’t that huge to begin with, the game requires that you take many of the same routes to accomplish certain tasks. The designers, in this revision of the game, keep track of the movements of your character, and if you trod along the same ground for a few days in a row, a visually distinguishable path will form. These paths enable much quicker and more decisive movement through the space.

The larger point to all this is perhaps thinking about how elephant paths can be created and leveraged in other environments. Certainly implementing such paths in larger online worlds seems obvious. But what about other stuff? What about web browsing trajectories? What about more semantic paths? Most elephant paths are about the quickest way to get from point A to point B. With a more layered approach, we could try and account for running paths, shopping paths, art paths, or other paths that had more encoded information.

Pink Ponies: None of this is possible

I love j-pony’s writing style, since it reminds me of talking directly to him when he’s got this kind of flabbergasted look in his eye. His take here on a Garmin GPS unit is perfect.

Quickly: RJDJ

Pretty fair amount of coverage of RJDJ lately, but worth mentioning quickly, that, despite my general feeling that none of the “tracks” are that compelling, it is a mobile application that attempts in some way to facilitate the kind of “where I am now” value. Sitting at a cafe this morning working, and listening to the busy street and the 2 different sets of architects around me through this weird PD filter… is pretty nice. The “Single” echolon is my favorite, in that it basically just takes what you would be hearing normally and amplifies it as an echo chamber. Simple, sure, but more effective I think as a mobile experience than the other tracks, which all seem to sound a bit like squarepusher or something, and I can make very few 1:1 matches between my environment and the music.

The place you are

One of my routines is to take my dog out for a walk around 5 or 5:30 every day, which takes about 45 minutes to an hour. Recently I’ve noticed that the perceived time this activity takes varies drastically from day to day. On days that I’m in the mood to be outside, and don’t have some pressing thing to do, the walk seems to take no time at all. On the other hand, when I’m antsy and want to *get it over with* that half hour seems at least twice as long.

This really shouldn’t be surprising, but still important I think from the perspective of mobile design. How we perceive time in mobile situations, from being bored to being passively or actively engaged by some mobile experience should be of real importance of how we design any mobile experience.

Most mobile games are of the *time suck* variety, and tend to be casual (for example my fellow juror swinging his arm wildly around in the back of the courtroom the other day, engaged in a game of iGolf). Even just the process of messing around with my phone the other day during the walk, having it crash, restarting it, hard-restarting it, became a huge time-suck and by the time the phone was functioning again, I was back at home, almost completely unaware that I’d been walking.

I think one of the things that comes almost standard with mobile design is immersion, at least in a sense. Many of the prescribed applications for this medium are for “killing time.” If time is killed successfully, you become so immersed in whatever you are doing, that you have that moment where you snap back into reality and all the sudden you are next in line at the DMV. So immersion, in a sense, is free if you design something that is relatively engaging. The question is then (as I’m stumbling towards some kind of point), how do you create mobile experiences that don’t do this; how do you make stuff that asks us to engage *more* in a 1:1 ratio of real time vs. perceived time. With mobile design, it seems pretty easy to design for time-suck, but how do you design for the place where you actually are instead of the place you’d rather be?

DMV line image by flickr user charliereece

Viewfinder

Viewfinder is a collaboration with Michael Naimark and Erik Loyer, as well as with the Institute for Creative Technology. It was funded by a Google research grant. The basic premise is:

Viewfinder is a novel method for users to spatially situate, or “find the pose,” of their photographs, and then to view these photographs, along with others, as perfectly aligned overlays in a 3D world model such as Google Earth. Our objective is to provide a straightforward procedure for geo-locating photos of any kind, and our approach is to engage a community of users for a certain amount of human help. We specify that a 10-year-old should be able to find the pose of a photo in less than a minute, and we are convinced that this goal is achievable. While we are not entirely there yet, we are getting closer. This is our progress report.

My main role on the project was building out web and flash UIs that interacted with a google earth “server,” which we built to fundamentally fake the idea of a google earth image api. Using the UIs we build, people could fine tune where their photos were taken in google earth. Think clicking on a map, then seeing a street level view from that spot within google earth.

Links

• New York Times writeup
• Youtube videos