Main

networks Archives

May 16, 2007

What, me Published?

Well, not really. But some ideas that I helped develop and data I helped gather/generate were recently written up in this paper from the Harvard Business Review on Big Seed Marketing by my pals Jonah Peretti and Duncan Watts. The basic idea, embodied in the open source ForwardTrack project I helped create, is, as Kottke says:
Instead of relying purely on viral marketing or mass media marketing alone, big-seed marketing combines the two approaches so that a large initial audience spreads the marketing message to a secondary audience, yielding more overall interest than either approach would have by itself, even if the message isn't that contagious. "Because big-seed marketing harnesses the power of large numbers of ordinary people, its success does not depend on influentials or on any other special individuals; thus, managers can dispense with the probably fruitless exercise of predicting how, or through whom, contagious ideas will spread."

While I was not listed as an author in the HBR paper, I am listed on the as yet unpublished version on Dr. Watts' web site (abstract).

August 20, 2007

Cars vs Transit is like Packets vs ... Packets

A genius I know once wrote an article that some say brought the distinction between packet-switched and circuit-switched networks into the popular consciousness. In the decade or so since, we have seen packet-switched networks take over the world, and unfortunately some people find themselves tempted to abuse this bit of history in arguing related points. The gist of it will always be something to the effect of: "the thing I support is like Packets, the other thing is like Circuits, and since we all know that Packets beat Circuits, I must be right." (This is not unlike how many in my extended family will take anything they think is sufficiently bad and compare it to Hitler.)

For example, a recent piece (of what I don't know) by Stephen Fleming of the Georgia Public Policy Foundation, entitled In Transportation and in Technology, Packets Beat Circuits, starts off "Why are so many mass transit policies doomed to failure? Because packets beat circuits. Let's explore an analogy."

You can imagine where it goes from there ("cars are like packets, mass transit is like circuits, so cars are better"). The guy claims to have worked in digital communications for 10 years; I wonder if he's just bitter because he was on the wrong side of the packets vs circuits debate.

For the sake of all 6 people likely to read this, I hereby debunk this terrible analogy:

The fundamental aspect of a circuit-switched network, as stated in the first sentence of the Wikipedia article on Circuit Switching is that it "establishes a dedicated circuit (or channel) between nodes and terminals." That is, the bandwidth for the flow is reserved end-to-end for the life of the circuit. Traditionally, when I make a phone call, part of the 'space' on a bunch of copper wires connecting where I'm calling from to where I'm calling to is reserved even if no one is saying anything over the line.

Clearly a road network is not circuit-switched -- when you start out from your house you don't have a dedicated lane all the way to your destination (if you did, I might just own a car). In a circuit-switched transit network, not only would I have a seat on one R train from Union Street to Union Square, I would have a seat on every R train over the same route for the duration of my trip. Realizing this, the analogy breaks down completely.

Fleming's main argument as to why Transit is like Circuits is that the bandwidth hierarchy, in traveling by train, then bus, then feet, is like the digital transmission hierarchy of telephone (i.e. circuit-switched) networks. Perhaps he lives and works directly on top of an interstate and has never driven by highway, then arterial, then local street in his car. Or never noticed that the bandwidth on his home broadband connection is orders of magnitude smaller than the trans-Atlantic fiber lines that connected me in London to his web server in Georgia.

In a circuit-switched network, I only use as much bandwidth as I need at that moment, and only on the single link I'm currently traversing. When I get to the end of that link, I am put in a queue until the next link on my path is ready to receive me. It's true that it's a bit more obvious to see a road/auto network as analogous to a packet-switched network, but only because of the apparent simplicity of the rules of the system. Transit networks are of the same nature, it's just that the way packets (i.e. people!) are queued and switched where links connect is more complicated and constrained than on roads.

Speaking in data-network terms that we are all familiar with, transit mops up auto when it comes to bandwidth (total bits, or people, per unit time). The problem with transit is, in some circumstances, higher end-to-end latency (the time it takes for the first bit, or person, to get where they're going). But once you get the flow started, we all know that one track of even light rail service can carry the same number of passengers per hour (or was it bits per second) as 7 lanes of freeway or 17 lanes of street (see here).

Unfortunately, people actually read and believe this kind of proof-by-bad-analogy thinking. In a recent newsletter from the Reason Foundation, Robert Poole, Reason's Director of Transportation Studies, claims to have had his "Aha!" moment when reading Fleming's piece. Too bad for him the "Aha!" wasn't a realization to be much more careful with his analogies.

November 6, 2008

You Know What I Did Last Summer?

I spent 10 weeks last Summer as an intern on the strategy team of Transport for London's (TfL) London Rail division. This part of TfL is responsible for the London Overground, the Docklands Light Railway, and Tramlink, is the presumptive operator of Crossrail (if and when...), and serves as TfL's interface with the National Rail network. My general task was to help London Rail start to make use of the oceans of data spewing out of the Oyster smartcard ticketing system, but I spent the bulk of my time working on a project that came to be titled Oyster-Based Performance Metrics for the London Overground. I've posted my final report and slides and outline for the presentation I gave to TfL executive management.

Rather than try to explain the work, I've just cut and pasted the executive summary from the report and included some of my favorite figures (with no explanation). It's not a terrible paraphrasing, but if there is a lot of really good meat in the document if you are bored and hungry. Snooze on...


The London Overground is a pre-existing rail service in London whose operating responsibility and revenue risk were recently granted to Transport for London (TfL). Here we discuss the prospect of using data from the Oyster smartcard ticketing system to evaluate the performance of the London Overground explicitly from a passenger’s perspective.

The core idea behind our approach is to directly measure end-to-end individual journey times by taking the difference between entry and exit transactions stored by the Oyster system. The focus of this study is Excess Journey Time (EJT), calculated on a trip-by-trip basis as the difference between the observed journey time and some standard. In this case, the standard is determined for each trip with reference to published timetables, indicating how long the trip should have taken under right-time operations. A positive EJT indicates that the journey took longer than was expected.

Excess Journey Time is interpreted as the delay experienced by passengers as a result of services not running precisely to schedule. The distribution of EJT indicates reliability. We validate these interpretations using a detailed graphical analysis, and then aggregate them to the line and network level over a variety of time periods. Our analysis is conducted on large samples of Oyster data covering several months and millions of Overground trips in 2007.

At the aggregate level, relative values of Excess Journey Time are largely in line with expectations. The North London Line has the highest average Excess Journey Time of all lines on the London Overground, around 3 minutes, and the widest distributions (i.e. least passenger reliability). On all lines, there is significant day-to-day variability of Excess Journey Time. For the whole London Overground, and for the North London Line in particular, Excess Journey Time is worst in the AM and PM Peak timebands.

The current performance regime for the London Overground is the Public Performance Measure (PPM), which measures the fraction of scheduled vehicle trips arriving at their destinations fewer than five minutes late. Over time, EJT shows a strong correlation to PPM. There is clear additional variation in EJT, indicating that it captures certain information about passenger experiences that PPM does not. This variation tends to increase as PPM decreases, particularly in the AM and PM peak timebands, which suggests that the effectiveness of PPM as a measure of the passenger experience decreases as service deteriorates.

Another quantity of interest derivable from Oyster data is the time between passenger arrival at the station and the scheduled departure of the following train. The spread of this distribution of this quantity indicates the degree to which passengers arrive randomly (i.e. "turn up and go") rather than time their arrivals according to schedules. We have found that on the North London Line, especially during the AM, interpeak, and PM peak periods, passengers tend to arrive randomly. This is apparently in contrast to conventional wisdom for National Rail services, and has distinct implications for crowding levels and timetabling practice. In an appendix to this report we look at this in detail, and recommend that even headways be prioritized in timetabling the North London Line.

The Overground is, by design, part of a larger integrated multimodal network. Oyster data, by nature, is somewhat ambiguous in representing passenger trips on such a network that involve transfers or multiple routing options. This poses certain problems to our methodology, but also presents the opportunity to quantify and understand the experience of passengers across the entire network. We discuss these problems, potential solutions, and opportunities at length, as well as other applications for this methodology, and future research directions.

We have concluded that Oyster-based metrics are effective for monitoring and identifying problems as experienced by passengers on the London Overground. They may be even more effective for use across the whole of London's public transport network, particularly as Oyster is in the process of being rolled out to all National Rail services in the Greater London Area.

About networks

This page contains an archive of all entries posted to Frumination in the networks category. They are listed from oldest to newest.

money is the previous category.

nyc is the next category.

Many more can be found on the main index page or by looking through the archives.

Creative Commons License
This weblog is licensed under a Creative Commons License.