Ambient Agora Homepage

The Ambient Agora project has pretty much died off. But, they have a few more papers on the work at their project website: http://www.ambient-agoras.org/

There is a movie of the system in action at:

http://www.ipsi.fraunhofer.de/ambiente/paper/2004/AmbientAgoras_MPEG2_prante.avi 

Interactive Public Ambient Display

There’s a guy at Toronto who built a pretty awesome public ambient display, and also tried to emphasize a few zones of interaction.

http://www.nonsequitoria.com/v.php?s=research&f=interactive_public_ambient_displays&i=phases.gif

I don’t know if he’s still working on it, as his CV hasn’t been updated in a while.

Fun quote on BT

Back in 2005, when a journalist asked Steve Jobs if iPod would support BT headphones, a part of his reply was that Bluetooth is a technology in search of a problem. The comment struck in my mind as “BT is a solution in search of a problem”. IMHO BT has failed to live up to its hype. May be it was hyped too much, may be it was the faulty implementation.

I told my friends that a day will come when you can walk into your house and calls to your land-line will be routed to your mobile via BT. Has that day arrived? Was I sleeping?

Traits of a PD

Over the last couple of weeks I have been brainstorming different ideas for public display apps. I’ll put down some of the generic traits that were recurrent in my ideas. These ideas are also gleaned from the readings. None of the ideas are mutually exclusive, so many of the apps I imaged had a few of these ideas.

• Ambient information: Public display can serve as excellent tools for ambient information. My definition of ambient information is information that is valuable to user at the given moment but not critical. I would love to know that my friend is looking for someone to fill the empty slot in basketball game tonight, but I can afford to miss the information. However when the user becomes habituated with the interface, we can reliably expect the user to look at the display.
• Conscious interaction with the user: Many design ideas required conscious interaction between the user and the display. My idea of allowing people in a room to play scrabble using their cell phone to input text is an example. Applications having such interations will have to deal with coordination of inputs from users more diligently than other apps.
• Context based interaction: I commute between north and central campus regularly. If the public displays at North and Central campus is going to show me when the next bus (to the other campus) is leaving when I’m near the display - it is an example of context based information. The displays have to be aware of their physical location, and attributes of its surroundings. This will also limit the mobility of the display. I can’t make my laptop screen as a public display, while I’m not at my office. Or can I?
• One on one interaction: Although they are public displays, in some contexts we can assume that there will be only one user interacting with the display. Like the Unicase displays at Accenture labs. The display can handle interaction with multiple users, but it is primary designed for 1-on-1 interactions.
• Aggregating interaction: Opposed to the previous idea is apps that need multiple people interacting with it to showcase its skill. The ScreamMarket application is an example where interactions with many users is encouraged. Personally I’m not favorable to these applications, they need to sustain their novelty to keep the users interested.
• Serving as completely different tools: - Public displays equipped with camera can be used to store and broadcast messages. Instead of spending time creating fliers, students can storing and broadcast messages across campus using the public displays. They also be used to communicating across buildings connecting users separated by two displays. They can also be used by the dean to eavesdrop on the rumors spreading in the corridors.

Project Oxygen

It would be worthwhile to take a look at Project Oxygen to get some context on work that’s been done in the past.

Finally Found a good Survey Article on Ubicomp Systems

So I have been looking for a good survey article on Ubicomp Systems for a while, and I finally found one:

Endres, C., Butz, A., and MacWilliams, A. 2005. A survey of software infrastructures and frameworks for ubiquitous computing. Mob. Inf. Syst. 1, 1 (Jan. 2005), 41-80.

This is a great review article, simple, to the point, and pretty complete when it comes to providing information about a lot of Ubicomp systems.  I really wish I had read it about a year ago, but it seems somewhat hard to find.  In any case, it doesn’t provide very much depth on any of the systems, but provides a good place to start when hunting for a framework to use for a Ubicomp project.

Publish-Subscribe for Sensor Networks

I’ve been doing a good amount of research trying to figure out if the publish-subscribe model I have been working on for pervasive computing systems is novel in any way. So far it is at least novel in I can’t find anyone that has released source-code — so far. [correction I was wrong]

Papers focused on Systems: 

PLIM Context Aware System - Jabber based approach, Jabber works well for high level data, but it creates a lot of centralized overhead if it’s used as the primary transport.

Thirunavukkarasu Sivaharan, Gordon S. Blair, Geoff Coulson: GREEN: A Configurable and Re-configurable Publish-Subscribe Middleware for Pervasive Computing. OTM Conferences (1) 2005: 732-749 (I could not find a research group page for this system)

MundoCore -  is a pretty complete low level core for sending messages around in a UbiComp environment.  It was built as part of a larger Mundo Project at Technische Universitat Darmstadt in Germany.  This group look like they have a lot of recent releases and lot of systems style research for Ubicomp.

Aitenbichler, E., Kangasharju, J., and Mühlhäuser, M. 2007. MundoCore: A light-weight infrastructure for pervasive computing. Pervasive Mob. Comput. 3, 4 (Aug. 2007), 332-361. DOI= http://dx.doi.org/10.1016/j.pmcj.2007.04.002

Aitenbichler, E., Lyardet, F., Austaller, G., Kangasharju, J., and Mühlhäuser, M. 2007. Engineering intuitive and self-explanatory smart products. In Proceedings of the 2007 ACM Symposium on Applied Computing (Seoul, Korea, March 11 - 15, 2007). SAC ‘07. ACM, New York, NY, 1632-1637. DOI= http://doi.acm.org.proxy.lib.umich.edu/10.1145/1244002.1244348

[ This claims to be a user oriented paper, but in my opinion lacks evaluation of some of their important claims -- i.e. the complex interactions they claim their system is useful for are not really talked about ]

Papers more focused on attributes of systems:

Taherian, Salman; Bacon, Jean, “State-Filters for Enhanced Filtering in Sensor-Based Publish/Subscribe Systems,” Mobile Data Management, 2007 International Conference on , vol., no., pp.346-350, 1-1 May 2007
URL: http://ieeexplore.ieee.org.proxy.lib.umich.edu/iel5/4417106/4417107/04417180.pdf?isnumber=4417107∏=STD&arnumber=4417180&arnumber=4417180&arSt=346&ared=350&arAuthor=Taherian%2C+Salman%3B+Bacon%2C+Jean

[This paper emphasizes that state-filters are pretty important, we use state-filters in our system, although, probably abstract the statefullness in a let elegant way then these researchers would recommend.  The authors also talk about the difference between Source-Side and Scoped filtering -- Both important concepts for building scalable pervasive systems. (Worth citing down the road) ]

Cilia, M., Fiege, L., Haul, C., Zeidler, A., and Buchmann, A. P. 2003. Looking into the past: enhancing mobile publish/subscribe middleware. In Proceedings of the 2nd international Workshop on Distributed Event-Based Systems (San Diego, California, June 08 - 08, 2003). DEBS ‘03. ACM, New York, NY, 1-8. DOI= http://doi.acm.org.proxy.lib.umich.edu/10.1145/966618.966631

[ This paper has a nice look at the problems people run into in publish-subscribe architectures with mobile devices.  The authors propose the buffer + replay of old events model, which seems to be a good approach in many situations]

Migrations And You!

This week’s post from me to you is about migrations. “What are migrations,” you ask? They’re ways to update the schema of a database without actually touching icky SQL syntax. I actually like touching SQL, but if you don’t then migrations are the way to go! They’re also useful for keeping the changes of the database saved in some kind of format so you remember what you changed.

So here’s some posts about migrations! The first is from rails wiki. This is a pretty sweet source. I used to refer to this a lot when I was learning RoR. It’s the first link on Google, so it must be good!

http://wiki.rubyonrails.org/rails/pages/UnderstandingMigrations

Up up down down left right left right b a start
Just because we use cheats doesn’t mean we’re not smart
So here’s a migrations cheatsheet

http://garrettsnider.backpackit.com/pub/367902

Here’s a link from International Business Machines about migrations.

http://www.ibm.com/developerworks/java/library/j-cb08156.html

Cell phone detector

Cell phone detector

Cell Busters - Cell phone detection and prevention. detectors, blockers and jammers

Cell Busters - Cell phone detection and prevention. detectors, blockers and jammers

I wonder if any of the old S60 phones can be reconfigured to do detection/location…