Midterm – build a network

Posted on Friday 2 March 2007

Network Effects – Midterm – Design and Implement a Network


The general idea of my midterm is to implement a network of parking referrals for residents to swap parking spots on the street with each other and perhaps develop relationships and get to know each other a bit more.

The nodes in my network are everyone living in my building with some non-residents included in the interaction, such as building staff and perhaps visitors of residents. The primary or most important nodes are going to be the doormen, as trusted connectors to the otherwise unconnected residents these staff see everyone coming and going. Located at the doors on street level they also see a lot of what is going on in the neighborhood including people coming and going from their parking spots. The job of a doorman is not to just to open the door but to help the residents by receiving packages, helping load and uload cars, help children get safely to school buses etc. The secondary nodes are all of the residents. We have 21 floors of people that come and go constantly, many with cars that need to be parked on the street. The residents are usually disconnected and seem to like it that way. The makeup of the building of a mixture of owners(or shareholders in the coop) and renters, the ratio is approximately 60/40. I don’t know for sure, but I suspect from casual conversation that a higher number of people who own their apartments also own a car. Of those that park their cars on the street it seems that a higher percentage of the people I talk to are renters. The challenge with the nodes will be getting people in a coop to interact that normally keep to themselves.

The protocols for arranging connections in this experiment grow naturally out of the way we all think of each other in the building. Addressing is how you might think of each other and how the doormen keep track of everyone. Residents are identified by first name and then apartment number. (i.e. john/4e) personnel by first name and title/function in building. Transmission of content or the information as to when and where a parking spot is located is handled by the doormen. To give you an ideal of the flow and to explain the drawing below, here is how the system would work. I have given the doormen a stack of bright green index cards with the following copy:

“I will be: , I might be: (circle one) Leaving my parking spot at _____am ______pm on
m t w th f Sat Sun (circle one) Name__________________ “

The cards are kept on the table that the doorman uses to keep the sign-in sheet and other notices and administrative documents. The bright green is meant to be a visual reminder of availability of a parking spot, so the doorman has instructions to put out a green card when a spot becomes available. If “John” knows that he will (or might be) leaving at 12pm on Wednesday, then he fills out a card and it sits on the desk. Residents would learn of it’s existence through the doorman and hopefully through each other as it becomes successful.
If a resident sees a card or speaks to the doorman and learns of the spot, then the doorman marks a check on the card with the interested parties name. Information that the spot has a “taker” is passed back to “John” who then knows to pay attention when he gets ready to leave so that he might look for “Susan”. The doorman helps identify whether Susan is moving her car or if she is coming in from somewhere, the doorman also has information from the card about the make and color of the car. So transmission of the actual parking spot is through the primary nodes, with information about the networks existence through the secondary nodes.

Predictions: The principal activities of people who park on the street are one of two activities. The first is simply coming from or going to somewhere and needing a spot close to home that will be good for at least one day until 9am the following day, or preferably several days. This leads to another behavior of on the street parking and that is the alternate side switching ritual where to avoid the street sweeper and the more importantly tickets from parking enforcement people keep track of the days they need to move and move their cars to the other side. I anticipate that this activity of moving will account for as much of the networks traffic as anything else.

My success goals would be within one week to have 8 referrals and at least 2 successful parking spots. I feel that the culture of the building might be sort of reluctant to associate with each other in this sort of arrangement. I do feel there will be many questions and I hope that the doormen and super will be helpful in getting the word out.

Post mortem:

OK, now that the network has run I can describe what I observed and what kind of success I had in implementing what I wanted to do with this. In the beginning I picked to people that I considered sort of super-nodes, the connected of the connected in our building. All of the residents, my self included depend heavily on these personnel who keep the building running. The doorman and the super were involved in my project and I depended heavily on them to help run it. These super nodes were also critical points of failure if they didn’t perform the tasks that were required.

In the first few days that the project ran I observed a fair amount of interest in the cards. Quite a few people noticed them and asked what they were. The doorman explaned the idea to them and asked if they were interested in participating, many said they were and that it was a good idea. A few people asked me about it and I also explained how it was to work and what I was trying to do. After the first day there were not cards filled out, however the doorman received several possible notifications. To explain further, he said many people said they might be leaving but that they didn’t want to fill out the card until they were sure. On the second day of the project I ran into Louise, a woman on the second floor who I observed parking along with me during the alternate side ritual. We had never spoken, but this time she brought up the cards in passing and said that she would fill one out, she later went back and did this. I arranged to switch my car to the other side of the street due to that exchange. By the second night another card was filled out to exchange a spot with one of the doormen leaving his shift at 11pm. The interesting thing in this network was that the doormen were connected in this sort of arrangement already, most of them travel by car to queens or the Bronx and swap places among themselves. The doorman was most helpful in running the experiment was the day man Joel, he had already been formally tipping me off on where to park and when people might be going. The cards in general had tepid response at best, I had an expectation that more might be filled out at least, then I thought the success rate of connection would be lower due to people not connecting at the right time.

By the third day the referrals stood at the following:

Actual connections:
I got a spot from Louise in 2c
Louise (2c) later got a spot from John /building personnel
John in 20h got a spot from me.

Cards filled out

I think that my idea of the doormen as super-nodes was flawed not because of the fact that they weren’t in touch with everyone, but that they were inhibited in getting the word out. When I talked to Joel I asked him if he had the cards out he said yes, but that he felt reluctant to actually talk about their existence. When pressed further he seemed uncomfortable so I didn’t pursue it. I think that Jane Jacobs point on the sidewalk connectors was that they performed a duty based on implicit trust and disression. The idea of privacy comes into discression and that is were it all broke down I think. I am guessing, but the existence of the cards became a piece of information on how people were coming and going. Also, parking on the street may not be something that certain people are comfortable broadcasting. There may be a financial class system where some people might not want to admit that they couldn’t afford to garage their car. If I redesigned the system, I might work on something that was more of a back channel that didn’t involve the doormen. Also, the doormen already had their own network and that might have involved a certain conflict of interest.

Sorry, the comment form is closed at this time.