— ITPG-GT 2621 001 CLOUDCOMMUTING

Archive
Author Archive

rumors

 

Rumors Game – Jiwon, Michelle, John and Haylee

To potentially improve communication between ITP students (especially between two classes) by better understanding social divisions/cliques and flow of information in the community by testing viral spread of information in ITP network via a Rumor Game.

Misinformation/rumors are a form of information that have a natural tendency to spread virally in social networks and could be used to create an interesting game to learn about the topology of the ITP community/network. Through a rumor game, could we learn about the social cliques of the ITP community, and how interconnected people are? And how this affects viral spread of information in the ITP network and conversely, other networks?

Last year, 2nd year students dominated the ITP list serv and first year students did not use as frequently but could often be found posting on Facebook. It is also harder to meet students in the other classes due to how the curriculum is setup. By understanding the current flow of information within the community, are there opportunities to improve the community and communication between various members of the community? Especially given the diverse makeup of the ITP community with many people from various countries and cultures. One of the strengths of the ITP community and program is the network of connections students make while attending the program.

 

Read More

I visualized the citibike membership data (April ~ June. 2014) with D3 js.
You can find the visualization here

+ I explained about my process here

 

 

Read More

o   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.).

I took a surveillance camera class last semester and For the assignment, I had to take pictures of surveillance cameras around me. It turned out even from Union Square to school I found tons of surveillance cameras that I could not count the whole number of those.

Mostly Police department or Government own them. The goals are to take care of safety /to check traffics/to check the speed of each car. How it works: If police find the car running too fast, they try to find the owner of the car, which is really easy for them. Then, the owner will get a ticket shortly. I don’t really know how stakeholders receive the feedbacks of those cameras, I guess if someone broke a camera, there might be a sensor to detect the status. They are tons of cameras in NYC, which is a bit scary but people try to ignore that fact and they are basically sensors.

The fun fact is now it is really hard to tell if this machine doesn’t connect to a network or not. For example,  as this book mentioned, even air-conditioning systems have sensors for detecting temperature and humidity and they are easily connected to various networks. Now we can say, vacuum cleaners have eyes.

o   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?

I am thinking about car sharing.

  1. Carpool. I heard that many of people who live in New Jersey (near george washington bridge) do carpool to split the expensive toll to go to Manhattan. The way how it works is people who drive cars are waiting in front of the bridge and people who want to take their cars just ask those drivers if they can ride their cars together. I heard that there are some internet websites to find the carpools. I don’t know why crossing the bridge costs a lot but I am pretty sure if more places require expensive tolls, more carpools can exist. The control problem can be it doesn’t have any system so we never know about supply and demand.
  2. Car rental (like Hertz) People living in NYC rarely own their cars because this city is too crowded and no space for parking. How it works: You have to go to rental place and have to return at there. The way how you can get there is not considered. You can reserve the car and it  helps to control distribution.

For me, if there is any way to use Mod car like a Citibike. It would be really great. I sometimes need a car from IKEA to my home or when I have to move, I need a car and it should be one way trip. I don’t want to go back to rental car shop to return it.

o   Read MIT’s report on driverless automobiles (ignore all technical stuff and focus on results). Suppose that in few years we have driverless automobiles that can be used in shared schemes without the need for redistribution. Is this a plausible solution/future for you?

Yes, I think so. According to this book, the author also mentioned that the control of distribution can be a challenge of automobiles but there could be some possibilities to manage supply/demand issue based on real-time data from vehicles themselves. Now Citibike is using truck to do a redistribution but if these driverless automobiles connect to a network (mesh network might be better solution because of cost, as author mentioned). Even these automobiles can detect traffic flows and these information can change the price. And I think it can be a really good solution to keep the spatial and temporal distribution of automobiles supply.

o   Read both articles of DeMaio that illustrate the history, future and challenges of bike sharing. Do you think in the future we will need better centralized control, more self-governance (incentives for users), or a combination of the two?

I think we will need better centralized control because in terms of redistribution and placing the right spot for all users need a sort of strategy based on real time data. In terms of convenience, self-control/customized one would be better but it has a complexity including cost. I think self-governace bike sharing costs more than centralized control sharing because it requires much more demands.

Read More