Each week we will be focusing on blogs from around the world dealing specifically with urban environments. We’ll be on the lookout for websites outside the country that approach themes related to urban experiences and issues.
• Streetsblog questions the conventional planning wisdom that successful transit depends on density. Using the counterintuitive example of the comparatively dense Las Vegas with Vancouver, Jarrett Walker discusses the importance of how we measure density and the role for transit-supportive design.
• Pop Up City advances the idea that it’s opportunities for romantic partnership (rather than work) that attract us to the city. Urban ecologist Remco Daalder argues, “Sex in the City is not just a slogan, it is an important fundament for Amsterdam’s economic success.”
• DNA Info has pictures of New York’s latest pedestrianization project: Union Square. The $500,000 pedestrian mall, on Broadway between 17th and 18th streets, has carved out new public space and improved intersection safety.
• We all know dog poop could be put to better use than smeared on the bottom of a rubber soled shoe. As WorldChanging reports, Matthew Mazzotta acted on this knowledge: creating a dog-poop powered street light in a Cambridge, MA park.
Photo from WorldChanging
Do you have a World Wide Wednesday worthy article you’d like to share? Send the link to www (at) spacing . ca
12 comments
The geography of Las Vegas and Vancouver couldn’t be more different. Downtown Vancouver is basically located on a peninsula, meaning there isn’t room to build sufficient road capacity to serve it without building lot of expensive bridges, so people have to turn to transit.
The Vegas-Vancouver thing is a blog post commenting on a blog post commenting on a book and a little hard to follow, but in reading through I see that the author of the book had an earlier publication where he went into great detail on one of my favourite topics, Melbourne vs Toronto. If any Toronto bookstores have a copy of this lying around it would be worth a look:
“A Very Public Solution: Transport in the Dispersed City”, Paul Mees, 2000
The StreetsBlog post was commenting on a post humantransit.org which is commenting on Paul Mees book Transport for Suburbia.
Sheesh..
Closer to home (UofT) Profs Miller and Daboe state ” employment density was the most important variable ”
If Spacing is really interested in the Issue, you should interview Prof. MIller.
Also of interest is this very good paper from Amal Ibrahim “Investigation of the Relationship between UrbanSpatial Structure and Travel Demand in the GTA” available here…….
http://www.collectionscanada.gc.ca/obj/s4/f2/dsk2/ftp01/MQ28832.pdf
or the short version here…..
http://southofsteeles.blogspot.com/2010/07/job-city-before-transit-city.html
The Mees book recommended by Iskyscraper is available in the Toronto Reference Library
Glen,
I took a quick peek at the shorter version, and would like to comment on what I read there:
“The Graph speaks for itself. Traditionally transit ridership is correlated with employment,”
True, pretty clear from the graph.
“not population density.”
Bullshit. In the case of Toronto, the population density is almost held as a constant. How can you derive any conclusion about the correlation between ridership and population density when one of them is a constant?
Yu, when one of the three variables changed (employment) ridership followed in kind. When a second variable changed (price of oil) ridership went up predictably (not just in Toronto). Population was nearly constant yet ridership fluctuates greatly during the period. If there was a strong correlation between ridership and population, as you seem to suggest, then ridership would have remained constant. Which it did not.
Glen,
Sorry but it is hard to discuss this with you what you need a statistics 101 lesson.
Yu, my graph was to show a strong correlation to both employment levels and fuel prices, regardless of population density. The comment about the population density relationship is not a conclusion that I drew from the graph, but is somewhat reinforced by it. It is one formed by reading the literature. Is it important, yes, in the most obvious of ways. At the same time it can also be a non factor. In a mature city with a fixed border, you could not achieve a large enough population growth within a reasonable time frame to provide a useful conclusion.
Normally I wouldn’t join in a debate like this but I felt somewhat compelled by Glen’s namedrop of my thesis supervisor.
My two cents are that this discussion is somewhat moot in Toronto…almost no modelling of transit ridership is done using a single variable anymore (apart from extremely high-level – like national – scales, or for rough calculations). Transit ridership is closely related to mode choice; indeed it is the aggregate of mode choices made across the entire population. Modelling this is called microsimulation, which is one of Eric Miller’s research topics.
Glen, your following statement shows a very basic luck of understanding about statistics, even though your later comment seems to suggest that you kind of realized it and was trying to rephrase yourself to look not so ridiculous.
“If there was a strong correlation between ridership and population, as you seem to suggest, then ridership would have remained constant. ”
Let me give you a simple example to show why this is ridiculous: imagine that you run an ice-cream store, and you are using stats to help your business. You decide to study the correlation of two variables, temperature, and price of your ice-cream, on the number of clients you server daily. During the study you kept the price constant. After a summer, your study shows that there is a positive correlation between temperature and the number of clients. At the same time, you also found that there is no correlation between price and the number of clients, because, “if there was a strong correlation between price and number of clients, as you seem to suggest, then number of clients would have remained constant.” Equipped with this knowledge you confidently double the price of your ice-cream. Now, are you going to make twice as much money, or are you going bankrupt? Well, both are possible, but in normal market situation, the latter seems to be more likely.
Peter, you are right. This is a very complex issue with so many possible variables. My argument with Glen is indeed pointless.
YU,
Your analogy does not apply. Lets modify that scenario and raise ice cream prices by 5% per week. If sales continue fluctuate in relative amounts to temperature but a agnostic towards price, there is some information to be had. At some point the rise in prices will alter the previous relationship of temperature:sales.
WRT transit modelling, it is very complicated (hence the link to Amal’s thesis). One thing is simple though, unless transit can serve the origin/destination relationship, which is heavily influenced by employment levels and density (for economic reasons), it will not be utilized. Which was my point. I think the problem is that you are assuming I drew my conclusions from the graph. I most certainly did not.