— ITPG-GT 2621 001 CLOUDCOMMUTING

Archive
Author Archive

Saki//David//Dan

For this week, we worked on the final implementation of the networked side of our game. Because our game consists of a array of networked station that players pick up and drop off stock, we had to create a server and database that constantly updated each stations stock and bonuses as well as each players point totals.

When a player arrives at a station, they swipe their RFID card and received an alert on their phone. If the player has stock, they are alerted that they can drop off their stock, or if the station is full , they are denied. If the station has only one available slot, the player is alerted that if they drop off they will put a system wide penalty into effect. If the player has no stock, they are alerted that can pick up from the station, or if it is empty they are denied. If the the station has only one stock, they alerted that they will put a system wide penalty into effect.
Because of these dynamics, each player needs a constant update of the status of the entire system. This includes every players scores as well each stations current stock and any bonues that are in effect. We were able to accomplish this using a node server connected to a mongo database. Each player accesses a client-side webpage that is continuously updated via websockets from the server.

The front-end of the client has a simple data viz showing each player points and station inventory. In the next couple days we will refine the front-end to more clearly let each player know whether they are holding stock or not. We will also experiment with further game dynamics to better create an exogenous network.

Read More

By Dan//Saki//David
StrategyOfTheCommons-02

For our final project, we are looking at the stocks and flows of a peer to peer mobility system as an example of the tragedy of the commons.

We are asking two questions:
If people are aware of the negative effect they have on the overall system, will they take measures to prevent it?

Can individual incentives be used to counterbalance the tragedy of the commons that currently exists in systems like CitiBike?
GAMEGUIDE-01
To help answer these questions, we are creating a game played in the physical world that deals with taking and leaving stocks at stations.
Our system will incentivize leaving stock at empty stations and taking stock from full stations.
Our system will penalize leaving stock at almost full station and taking stock from almost empty stations. If one does choose to do one of these actions, a system wide penalty will go in to effect, for the duration the station is completely full or completely empty.
graphic_interface-01
Using RFID card readers connected via Arduino Yun to a central server, players will take and leave stock and be served a personalized webpage on their phone informing them of the overall scores of all the players as well as the incentive or penalty for leaving or taking stock at the station they are at.

We will play test this game to figure out the best scale and point proportions to create a dynamic game and balanced system. The scale could be the ITP floor or even the entire Tisch building.

 

Read More

 

 

by David Tracy, Dan Melancon, Saki HayashiquantifiedFoos
We created a simple application that can log foosball matches using RFID cards.  Arduino Yun sends the RFID number to a Node.js server connected toMongoDB database.  Using simple GET requests from the Yun, we can record and store Foosball players, teams and game results.   A simple LCD screen displays basic feedback to let the player know their info was recorded correctly.

Get Requests:

New Players: localhost:8000/newPlayer?playerID=1111111

New Game: localhost:8000/newGame

New Red Team: localhost:8000/redTeam?playerOneID=1111111&playerTwoID=2222222

New Blue Team: localhost:8000/redTeam?playerOneID=3333333&playerTwoID=4444444 Game Start: localhost:8000/gameStart

 

FullSizeRender

Read More

Screen Shot 2014-09-23 at 8.02.36 PM

Working with Saki and David, I made a NetLogo user agent model that simulates the behavior of users with different demographic types, in this case blue and red users. Each station has quota of blue and red user types that it wants to fill, as well as an overall user limit. Each user tries to find a station in which their type can fit in.
The purpose of this simulation is to determine if user demographic has an affect on the overall usability and capacity of the system. For our midterm, we will try to make multiple demographics by which to evaluate it as well plugging in some real values for the stations quotas, based on real citiBike trip data.

Read More

Read my blog post about this visualization here.

Read More

Compare different models of sharing that exist (or you can think of) in MoD systems (e.g. vehicle sharing, parking sharing, ride sharing, etc.) What operation/control problems do they have? What would the ideal form of sharing be for you and how would its resources be controlled?

When I think of an ideal sharing model, I tend to think of a different type of sharing. Whereas citiBike and ZipCar are a kind of private sharing systems, in that the vehicles are discretely being used on ones’ own defined timeframe, I think of a system with algorithmically-created shared shuttle routes. Something more akin to the routing systems in large distribution warehouse used at places like Amazon.
In these warehouses, individual human workers still have to parse the bins of products and find the correct products to fulfill online orders. However, they not merely given a single customers order and told to search the enormous warehouse for these different products. That would be hugely inefficient. Instead, a centralized computer reads all the incoming orders and dictates to each individual worker a ideal paths to fulfill some products that contribute to concurrently filling many people orders. The end goal of fulfilling each customers order is still accomplished. However, by splitting it up and then reassembling it them algorithmically an ideal path is taken.
In my imagined system, the user would input their destination on a their mobile device and a centralized computer would pair the user up with other users whose origins, routes, destinations are determined to be compatible. Then a shuttle or bus would take this ideal route. Adjustments and stops and route changes could be made on the fly to account for more users, traffic, etc. Obviously, there would be a whole new set of problems with this type of system, but I feel that truly shared vehicles/transit as opposed to ‘shared private’ vehicles such as citiBike or ZipCar could be a more efficient solution to congested transit.

In ME++ Mitchell talks about “electronic nervous systems” of intelligent urban environments. Discuss an example of an intelligent urban system you are familiar with and discuss the elements of the feedback loop, how its form of governance works, and who are its stakeholders (goals, decision makers, evaluators, etc.).

An example of an intelligent urban system is the modern Postal System. The various postal routes are the links that can deliver correspondence or packages between everyone with an address. There are a few forms of feedback for the consumers, most importantly the timely receiving of expected packages and letters, as well as close to realtime online tracking numbers. Every user knows that some mail comes every day and some mail only comes periodically. Much of this are unwanted and immediately discarded, but not because of an inefficiency of the system. Some packages are ordered from stores by the users and some packages come unexpectedly. The system is governed by the federal government and paid for by the consumers by the purchasing of stamps. As well, I would guess each mail carrier governs some specifics in the way they deliver the mail each day. The goal is provide transportation of correspondence and other shipped goods. At some point it was decided this was something that should fall under federal jurisdiction, but it seems like its not a profitable undertaking. But because almost everyone relies on the Postal Service for important bills, letters, etc., we are almost all stakeholders. There is probably some internal evaluation that can be done on the specifics of the system, but from a consumer perspective, the mail system works when your mail is received and delivered when expected.

Read More