Heuristic Nolli Map [Site]
Harvard GSD

Bus [Site]
MIT Design Lab

Elens [Site]
MIT Mobile Experience Lab

iRate: Random Credit Scores
poverty-action.org
.NET CF, OpenNETCF RAPI, J2SE, Swing, XML, Soundex

Augmented Reality Game
MIT/Microsoft iCampus

Dictoot
Django, MySql

FlickrFinder
J2SE, Swing, Threads
Get the .jar file.

Implementation of CoverFlow
Processing

StudioShare.org

• A site for photographers to find gigs and put idle resources to work (eg. Sublet Idle Studio Space, Rent Equipment).
• 4 types of Profiles ( Studio Owners, Pro-Collaborators, Collaborators and Equipment )
• Each resource has its own Calendar component attached that determines a resource's availability. Searching results for 'Canon 1Ds Mark 3' within 75 miles of Zip Code 02139 display all relevant entries in the database; Only those free during the specified dates can be booked.
• A Threaded Messaging System similar to Facebook's
• Payment System using Paypal

Python, PIL, Django, MySql, YUI, JQuery, Google Maps

Prototype of a Tag Cloud
J2EE, XML, Threads, Flickr, YouTube

Flash Widget
AS3, XML, PHP

Prototype of a 'My Places' FB-Application
PHP, MySql, FBML, MySql, Yelp

    I strongly believe the next big thing will come from LBS / AR / Mobile Social Networks. This belief has been shared by many for years, yet for several reasons it still hasn't materialized; I believe that the following issues need to be addressed for this dream to be accomplished:

- Merging the real world with the online world

Having venue pages on Brightkite and Yelp is nice, but the 'exploration' feeling one gets when browsing the web (think StumbleUpon, Digg or just following links on blogs) is missing. How are users supposed to find out about venues and thus make the whole LBS-concept work? Through a styled textfield and a search button? Through an index of Venues in zip code 02139? I believe there needs to be an underlying infrastructure that links both real and virtual world for LBS to reach its full potential. Such infrastructure would serve not only as a conceptual container for social capital linked to a particular physical location, but more importantly, it would serve as means of connecting real life locations with each other, pretty much the same way links connect HTML documents today. Without a 3d panoramic recreation of space, LBS boils down to just another styled version of a hyperlocal news aggregator.

I know of a couple companies that claim to be recreating the real world online. For several years now actually. More important than how much venture capital they've raised, what really seems to matter when it comes to determining success is:

• How low-cost their image capturing process is
• How good the resulting images are
• How efficient they are at capturing the indoors ( Shooting by hand VS Using some sort of Innertial Guidance System to tag panos ? )
• How viable the business model is. This I think is particularly important. One can't cover the entire country on a round of series A funding.

As much as I enjoy these sites, I must admit I get bored after a while. As a time waster and a way of discovering new places they're nice, but there's no reason for me to come back every morning to check out my neighborhood online; It will be exactly the same. Nothing will have changed. Also, Indoors are much more interesting than outdoors, but the business plans behind them don't allow for a reasonable coverage, so after a while of poking around the site I get bored, and leave.

1. The We'll-sell-ads model: I'm not a fan of this model. It has not been proven to work for social networks thus far. To make matters worse, look at the cost of hosting/serving those panoramas alone: Assume the average panorama is 0.1MB and that one gets loaded every 15 seconds (ie 4 per minute). With the average user spending 4.5 hours per month on FB, that's 0.1MB*4*60*4.5 = 108Mb/month/user. Let's dream about taking those 250 million active users from FB and that's 250,000,000*108/1,024/1,024= 25,749.21 Terabytes per month. Good luck paying that bill Waldo.
2. Charging for Membership: Enough has been said on the subject, but this may be not that bad considering there's really nothing out there in the market that allows users to walk anywhere they'd like.
3. Charging Business Owners to shoot their locations: How many business people does one need to convince a business to pay for a panorama? How much time is invested? And the payoff? Forget it. This is not scalable. The value of a product gets determined by the user and the user only. Not by cluelessly asking for money, especially in a recession. Which brings me to the next model
4. Free: Free is good when it comes to scaling coverage and number of users; The question becomes how does one cover the linear costs associated with covering a location ( one-time fee ) and the exponential costs associated with bandwidth/hosting. Interestingly enough it seems like the latter is in fact the answer to the question. There's a vast array of financial transactions associated with physical locations: from tourism ( book airline tickets to get there, book a car to get there, rent a night at that location ) to tickets (concert tickets) to even virtual goods. And these happen multiple times over the course of a year ;-) Will this make up for the cost of hosting/serving/shooting immersive media ?

- Mobile Platform Fragmentation

This has been getting better mostly thanks to Apple (iPhone) and Google (Android). It's not as bad as it used to, but it did drive some of the earlier mobile social networks to the grave.