Those PRIZM types are hilarious and fascinating. I found a similar site a few years back, and plugged the zips of friends and relations into it.

Growing up, I figure I was in one of the Blue Chip Blues families in 11560.

I'm guessing I'm a Young Literati now, uncomfortably dwelling in suburbanified 97124.

* * *

Rural traffic . . . tell me about it. One Tuesday, Oregon's second-ever Krispy Kreme opens up around the corner from my workplace in 97006, on a two-lane road.

The police have announced preemptive road closures and detours.

We've been told to consider walking or biking to work.

Posted August 24, 2003 5:21 PM by Stefan Jones

PRIZM is useful, and it's been around long enough to have some of its original problems fixed. The descriptions are clever, but don't always work in parts of the country very different from the original midwest/northeast market. Such as here in the San Joaquin Valley.

There is a good reason not to list average communte times -- they don't mean anything anymore in many areas. For example, 20 years ago you figured commute times in the Bay Area based on disance from San Francisco. Now, if you are in San Ramon, where I used to work, you might work in a corporate headquarters there, drive west to Oakland or San Francisco, or even further south to Silicon Valley. Or, these days, turn east to work in a business park in Livermore.

I don't know what is happening in New York, but one of the bigger trends of the past couple of decades is the move of jobs out to the suburbs. In the South and West, it is often hard now to predict what a typical commute will be for some communities. Generally, everybody is spending a half hour or more in their cars each way, unless they are lucky.

Posted August 24, 2003 6:10 PM by Claude Muncey

Couldn't they ask a whole bunch of people how much time they spend commuting, and take an average?

But no, come to think of it; that wouldn't quite work, because people cite one time when they're talking about how smart they are to live where they do, and a different one when they're explaining to their officemates why they're late to work.

Posted August 24, 2003 6:32 PM by Teresa Nielsen Hayden

"These families are poor only in money."

Which, with $3.75, will get you a latte at the Fleet Farm store. Crackers from the barrel are extra.

Posted August 24, 2003 6:47 PM by John M. Ford

There's something seriously wrong with the compare cities thing. It says our town in CA, San Ramon, was 5 degrees cooler, on average, than Seattle. Um, no. San Ramon has a range of hills between it and the Bay and much more between it and the ocean and averaged a hell of a lot higher than 69 degrees in the summer. Air conditioning was a necessity. Unless they're taking nighttime temps in CA and day time ones in WA. It could get pretty darn chilly there at night. Not a desert, but you could drive to one from there.

On the other hand, they characterize our WA neighborhood pretty well. About evenly divided between Money & Brains, Young Literati, and Bohemian MIx. All of which are in the Uptown Urban group and fit well.
MKK

Posted August 24, 2003 7:44 PM by Mary Kay

Mary Kay, the problem with PRIZM is that it is built around blocks of zipcodes that were often drawn up 20 years ago. The whole Pleasanton/Dublin/San Ramon area was very different back then (as you know better than I). The block might include the Danville-Moraga area or actually spread all the way across the hills.

I enjoyed working in San Ramon (Bishop Ranch 15) but I'm not sure how I would like to live there . . . too many years out here in the sticks.

Posted August 24, 2003 8:11 PM by Claude Muncey

"blocks of zipcodes that were often drawn up 20 years ago"

Interesting thought: A historical PRIZM database. Drag the "current year" scroll bar and the listing changes to reflect the demographic composition.

My immediate neighborhood was literally farmland perhaps seven years ago. In fact, there's still a working farm field across the street; the land is owned by Intel but some farmer or other plants and harvests clover there.

A PRIZM analysis that doesn't take into account the 750 apartment units and hundred-odd upscale homes in the New Urbanism project just West of here that appeared within the last five years would be seriously, seriously skewed. A very large percentage of the occupants work at the giant Intel plant around the corner.

Posted August 24, 2003 8:54 PM by Stefan Jones

We've been living in a semi-rural tourist town for a few years now, and it definitely has its pros and cons. The pros are that, since it is a tourist town, we get a lot of amenities that our permanent population probably wouldn't warrant (two first-run movie theaters, lots of fast food, two full-size grocery stores). The trade-offs of Everyone Knows Everyone are definitely in place, but it works for me.

The biggest downsides are the lack of income possibilities (I supplement my wages working as a computer person for the school district with freelance web design and database construction, which can be done from home) and education (my husband has to drive down the hill to go to school, which means he's probably away 4-5 nights a week.)

It's still worth it, though, every time I look out and see the mountains.

Posted August 24, 2003 10:12 PM by Rachel Heslin

We've got a (hideously expensive) database at the library called iMark, used in the real world in marketing. It uses the PRIZM clusters, as well as census data and other research data. It's impressive. It can tell you, in theory, what kind of products the clusters use: fast food preferred, cars driven, how much spent on electronics, magazines read, tv shows watched, books bought, kitchen remodeled, where they went on vacation, and on and on and on. Then they can locate these groups from state to zip to county down to census tract. All so you can figure out how to market your whatever.

Two things they don't tell you:

1) the groupings aren't perfect. There are actually 64 groups of classification, not 62. The last two clusters are people who are unclassifiable. But they're a very small percentage of people, and therefore (theoretically) of no interest to marketing folk.

2) it's also garbage in, garbage out. Example: I once decided to see what clusters read science fiction (it's a search choice: "Buys science fiction"). The top three clusters most likely are: younger professionals who make good money and are on their way to the top; young college-age types in lower end or starter jobs living in smaller towns; and military people. Then I decided to see who's buying the most books. There's a search for "Buys 10+ books a year." The number one in this category are young singletons who live in places like New York City, make lots of money, are well educated, and do things like go to the theatre and watch Masterpiece Theater. It's cluster #6, Urban Gold Coast. All right. I then went back to my "Buys SF" list to see where UGC ranks, expecting it to be lower than average. It is. They don't, at all. Zero, zip, zilch, never ever do these people buy or read science fiction.

Problem is, it's not that they don't read science fiction, it's just not marketed that way. They're buying stuff like Atwood's Oryx and Crake, or Saragamo's Blindness.

Lesson learned: no model is perfect.

Posted August 24, 2003 10:22 PM by Jon Hansen

The other problem is that ZIP codes may not be granular enough for a meaningful comparison. My ZIP code (85016) currently has houses for sale ranging from $80K to $5.2M. Also, according to the Census, about 5.9% of households made less than $10K per year, and 5.8% made more than $200K.

Posted August 24, 2003 11:13 PM by Alan Hamilton

If they say that San Ramon, California, is cooler than Seattle, they're not using obsolete zip code data. They must be using the nearest regional weather figure they have handy, probably a station at Candlestick Park (as it used to be known) in San Francisco.

The factors search is anemic. "Violent crime"? So who wants to live in a place where that's "Worse than national average"? And "Number of sunny days"? Where's "Minimum need for air conditioning during the summer"? And what about cultural factors? I want search boxes for "Good bookstores," "Liberal local newspapers," and "Symphony orchestras."

Posted August 25, 2003 12:49 AM by Simon

I have real trouble believing that Park Slope has more violent crime than Bedford-Stuyvesant. The latter isn't the hellmouth that you might imagine if you only knew about it from the moral equivalent of Reader's Digest articles; but it just doesn't seem reasonable.

Posted August 25, 2003 1:24 AM by Teresa Nielsen Hayden

"Number of sunny days"? Where's "Minimum need for air conditioning during the summer"? And what about cultural factors? I want search boxes for "Good bookstores," "Liberal local newspapers," and "Symphony orchestras."

Simon: forget the search; just move to Minneapolis! :)

Posted August 25, 2003 1:32 AM by Stefanie Murray

"The last two clusters are people who are unclassifiable. But they're a very small percentage of people, and therefore (theoretically) of no interest to marketing folk."

Wow . . . are they immune from telemarketers and junk mail?

Of course, they might be subject to random detention by Homeland Security, because they don't act according to established models.

Posted August 25, 2003 2:12 AM by Stefan Jones

A neighborhood with a higher number of people per household is, I understand, going to have a higher crime rate all-else-equal, because domestic violence is common enough to shift the average. Domestic violence doesn't stick in people's memories as a neighborhood trait as much as street or stranger violence does. Is that a plausible explanation for Park Slope vs Bed-Stuy?

They do have a category for "Commute by bus", maybe that absorbs the subway and light rail numbers. Ill-named, if true.

"Dominant Race:a0 White, High Asian" - !! - is that short for "high proportion of Asians", or are they explicitly describing some Asians as higher than others?

Posted August 25, 2003 2:20 AM by clew

Also, it notes that Phoenix has a much higher than average UV index, but doesn't flag it as a "worse than average" value.

Posted August 25, 2003 4:02 AM by Alan Hamilton

When they talked about "The Sweet Smell of Country Life" they didn't mention farmers fertilizing their fields, or downwind from a dairy barn.

But be that as it may, where I live clocks in like this:

Blue Highways
Moderate Blue-Collar & Farm Families
38.72%

Rustic Elders
Low-Income, Older, Rural Couples
34.15%

Back Country Folks
Remote Rural/Town Families
26.99%

Which I suppose is fairly accurate. It's also 45 minutes of hard driving (assuming no snow) to the nearest McDonald's (same to the nearest movie theatre), and hitting a moose with your car is a major hazard.

However, the little map showing where things are (post office, fire station, etc.) is totally out to lunch, including showing some things that don't exist and others in places widely divergent from their true locations.

Posted August 25, 2003 8:03 AM by James D. Macdonald

Jim, I find that reassuring. I'd just as soon marketers didn't have us all pinned down too accurately.

The map for Park Slope is a crock. It's a chunk of the standard Brooklyn highway map, with a red pushpin marking the location of our neighborhood. Trouble is, the pushpin's been stuck in somewhere around downtown Brooklyn, DUMBO, and the zombie factory.

Posted August 25, 2003 9:10 AM by Teresa Nielsen Hayden

Well, I clicked through to get it to pick a place for me, and discovered it has absolutely no factors that I give two sneezes about, to use as narrowing factors. I don't care about 'average house price,' I care whether I can afford to buy *A* house that fits my requirements. There was no choice for 'small city and up.' I don't care about 'average future job growth,' I care whether I myself and my husband himself can find jobs people are willing to give us specifically. Everything else on there is so irrelevant to my choice of domicile, when 'averaged' over a whole region, as to be laughable.

At least the last one of these 'where should you live to be happy?' questionnaires I saw asked upwards of sixty specific questions about how I like to live and what I like to do (for the record, it thinks I should move to Connecticut, but put Chicago -- my current home -- third on my list).

Posted August 25, 2003 9:16 AM by Eloise Mason

I put in my neighborhood-- 11435, Queens-- and it came up with this, for the Prizm breakdown:

-Old Yankee Rows- Empty-Nest, Middle-Class Families 22.95%
-Urban Achievers- Mid-Level, White-Collar Urban Couples 21.48%
-Mid-City Mix- Urban Mix, Singles & Couples 15.95%

Three things about that breakdown: 1) it just ain't right, going by the people I see on the street every day. 2) The percentages only add up to 60.38; 2 or 3% missing, okay fine, but 40? No. It's just not done. And 3) the first time I put in my zip code, it came up with a very (completely? might have been) different list of types.

Turns out I did it two different ways. On the main page (the linked page) there's a field to enter the zip code, and there's also a link called "Search for a neighborhood like the one I live in now", on which page you can enter the zip into a field. The listing above is from the first method; the second one yields this:

-Southside City- Solo Parents & Single Service Workers 35.17%
-Suburban Sprawl- Young Midscale Suburban Couples & Singles 21.62%
-Starter Families- Young Middle-Class Families 20.82%
-Hometown Retired- Low-Income, Older Singles & Couples 10.88%
-Gray Collars- Aging Couples in Inner Suburbs 10.65%

Which is a bit more accurate (though there's no mention of the elderly, who I see by the hundreds), and the numbers add up to 99. But if they're going to give two massively different breakdowns for the same zip code, something's wonky somewhere.

Oh, also, the definition for "Commute by bus" says that it's the % of people using public transportation. I have to assume that means the subway too. Why not just say "Commute by public transportation?" Unfathomable.

Posted August 25, 2003 11:04 AM by Adam Lipkin

11435? Which neighborhood is that really?

Posted August 25, 2003 11:36 AM by Teresa Nielsen Hayden

What does looks like pure whitebread mean?

Posted August 25, 2003 12:05 PM by Damien Warman

What I found interesting about my neighborhood was that the median income was a bit over $20,000. But the median household income was $49,000. Yet another confirmation that it takes two or more incomes to run a household these days.

(When I was a kid, during the 50s/60s, our family usually got along on my dad's income as a machinist. There were extra-tight periods when Mom would take a job -- usually as a hospital switchboard operator; remember switchboards? -- but the majority of the time we were a one-income family.)

Posted August 25, 2003 1:11 PM by Bruce Arthurs

Wow. 08016 really does have ultra-low-quality water. And I thought it was just me...

Posted August 25, 2003 2:32 PM by Daniel Martin

Zombie factory? Teresa, is there a reason for posting the FVZA link that you haven't told us. It would be good to know before planning travel to NYC . . .

Posted August 25, 2003 2:54 PM by Claude Muncey

What is it with my forgetting question marks? This is the second in two days.

Posted August 25, 2003 2:55 PM by Claude Muncey

ZIP code granularity -- my entire town, Hoboken, NJ, has only one ZIP code. Trust me that it has multiple neighborhoods. The people who live in the luxury apartments on River Street are not in the same neighborhood as the people who live in the projects on Jackson. I'm not even in the same neighborhood, and I only live a block away (on Monroe).

But then none of these mapping things seem to understand urban microenvironments.

Posted August 25, 2003 2:58 PM by Xopher

"11435? Which neighborhood is that really?"

11435 is Briarwood. Bounded by the Van Wyck Expressway, Union Turnpike, Hillside Avenue, and, er, some street on the north side (164 St? Maybe, but that feels too far north). It's the Briarwood/Van Wyck stop on the F train.

Posted August 25, 2003 4:35 PM by Adam Lipkin

At least, 11435 is *in* Briarwood. Could be other zip codes leaching in at the edges. I think it's one of those neighborhoods that people come from, but no one's ever heard of. No one that I've talked to has heard of it, at any rate; "You know Forest Hills? Down Queens Blvd a couple of miles" seems to work though.

Posted August 25, 2003 4:39 PM by Adam Lipkin

How sad: 11231's "Commute by carpool" percentage is "worse" than the national average. Oh, wait--could that have something to do with the fact that few people in my part of Brooklyn commute by car? How terrible!

My point being, why do they say "worse" rather than simply "lower"? Stupid.

Posted August 25, 2003 7:06 PM by India

Adam: Got it. 11435 is where I wind up if I turn the wrong way when I'm taking the Interboro to the Van Wyck to the Whitestone. Neither rich nor poor; by some definitions mixed, by others not; mostly, a place where people live.

Claude, "the zombie factory" is the impolite term for the cluster of industrial buildings near the foot of the Brooklyn Bridge that are all owned by the Jehovah's Witnesses.

Damien, I'll have to think about that. I've never had to explain it before.

Stefanie, I've decided that that database is short on all sorts of useful warnings, like "there are no bookstores in this borough", or "odds of getting a seat on mass transit in the morning", or ratings on how embarrassing your local politicians are.

I'll know they've figured out urban micro-neighborhoods when you can find the Hell's Angels' territory on their maps, with a note saying that if you don't mind the sound of exhaust pipes, it's a remarkably safe block.

I can imagine further helpful notes: "super-observant neighborhood; no business done on Saturdays", or "average days per year plagued by movie shoots", or "highest per capita dyke population in NYC", or "falls within late-afternoon high-water mark of San Gennaro Festival traffic backup".

Bruce, the changeover to a non-elective two-income lifestyle has been an underreported story, as has the gradual (but significant) erosion of our disposable free time. One the days when conspiracy-mindedness gets the better of me, I figure it's part of the general plot to bring back the class system.

India: You're in Red Hook?

Listing us as "worse than the national average" on carpooling is pretty dumb. I note that my neighborhood is also worse than average on 2-year postsecondary degrees -- 4.68% to the national average of 8.19%. On the other hand, our percentage of graduate degrees is 27.64%, as opposed to the national average of 7.01%.

What appalls me are the national stats on commuting via mass transit. Your neighborhood and mine are pretty comparable, 65.18% and 65.36%. Rest of the country? 1.95%. Pathetic.

Posted August 25, 2003 8:16 PM by Teresa Nielsen Hayden

Fascinating. And a great time-waster.

Did you notice that one of the "Lifestyle Preferences" for the category "Urban Achievers" (one of the notable population groups, according to this site, in our old neighborhood in SF: 94110) is "Buy science fiction books"?

I've wondered for years why so many US statistics break down by Zip Code, but only by the five-digit Zip Code. In both our Seattle neighborhood (Capitol Hill, in 98112) and our former SF neighborhood (Bernal Heights, in 94110 -- which includes a good chunk of the Mission District), the Zip Codes are quite misleading about the actual microneighborhoods where we've lived. Why not make use of the much more specific nine-digit Zip Code for purposes like this?

-- Puzzled in Seattle

Posted August 25, 2003 8:36 PM by John D. Berry

Simon: If they say that San Ramon, California, is cooler than Seattle, they're not using obsolete zip code data. They must be using the nearest regional weather figure they have handy, probably a station at Candlestick Park (as it used to be known) in San Francisco.

Isn't there a weather station on top of Mount Diablo? :-S

Posted August 25, 2003 8:37 PM by Alan Bostick

I dunno, Ms. Hayden, the only two times I've ever gotten mugged is in Park Slope. My friend from Bay Ridge jokes that my neighborhood is the most crimeridden in the city, since his only time ever mugged was also in Park Slope. There are certainly some bad apples hanging about, IMHO.

Posted August 25, 2003 9:47 PM by The Littlest Cynic

I find it quite interesting that this website ignores particular details of ethnicity and religion. Type in "60455," and you'll have no way of knowing this ZIP code includes south Chicagoland's most prominent mosque, as well as a stretch of Harlem Avenue in which a third to a half of the storefronts have signs in Arabic. Type in "60477," and it won't tell you that the young families are mostly South Side Irish.

Is this just a matter of ignoring politics & religion because they're "impolite"?

Posted August 25, 2003 10:32 PM by Isabeau

Just to remind everyone: the only "Ms. Hayden" I know is my mother. And a fine person she is too, but her first name isn't Teresa.

As ever, our difficult last name is explained here.

Posted August 26, 2003 1:27 AM by Patrick Nielsen Hayden

Cynic, I got mugged in Inwood. You can get mugged anywhere.

Posted August 26, 2003 1:32 AM by Teresa Nielsen Hayden

The English equivalent, upmystreet.com uses UK postcodes, which translate to around 10 houses. It is pretty accurate for the various family addresses I tried.
One of the caps who runs it told me how he improved his neighbourhood rating by lobbying his neighbours.

Posted August 26, 2003 2:45 AM by Kevin Marks

"Ms. Hayden"? Is Melissa Hayden(*) posting here? KEWL!

I've been mugged twice. Once in Newport Beach, California, and once in Pasadena, two blocks from the Caltech campus.

(*) Book designer; photographer; world-class poker player.

Posted August 26, 2003 12:05 PM by Alan Bostick

Thanks for the note on the zombie factory, Teresa -- I'll make sure to miss it if I ever get a chance.

Along the general thread here, there is a good article up today on SFGate.com, the weekly installment of Surreal Estate by Carol Lloyd. (If you live out here on the Left Coast you know the title is not a joke.) It is a comparison of the SF and NYC real estate scenes, pointing out that the NY "halo" has better house prices than northern California right now -- to the surprise of many who live here. One reason Lloyd presents is that the jobs in NYC are still centerd in Manhattan, as opposed to the wider distribution cities like LA, the Bay Area, or Seattle.

I live in a city of roughly 60K, two and a half hours from San Francisco. It has been rather affordable for years, but is starting to get more expensive (well, our equity is increasing with that, so it's fine so far) because the next University of California campus will be opening here in the fall of 2005.

Now is the time to start planning the combination science fiction bookstore/comics shop/coffee shop near the campus. I can show you some good sites . . .

Posted August 26, 2003 5:48 PM by Claude Muncey

Late to the party...

"...inexpensive, pleasantly semi-rural areas within commuting distance of lots of jobs don92t retain their character for long."

They do if

1) They're protected by long, winding mountain road commutes through state park and forest lands, with no viable alternate routes.

2) They're surrounded by state park lands, protected watersheds, and logging preserves (at least in California).

Case in point: the San Lorenzo Valley in California (Boulder Creek, Brookdale, Ben Lomond, and Felton). The northern end (Boulder Creek) where I live is only about 50 minutes from most parts of Silicon Valley. Yes, there are more and more expensive homes, but the fact that there are only a couple of roads in and out of the valley, combined with the surrounding state lands that cannot be developed and the engineering impracticality of expanding those winding mountain roads, makes it impossible to grow the area beyond a certain density. I'm relatively certain that Boulder Creek will maintain much of its present character for a long time to come. It's a place where you can throw yourself into small town, intimate life, or keep aloof.

And, yes, housing is getting more expensive here--it's certainly shockingly high by Midwest standards, but it's still 25% under Silicon Valley, and shows no signs of catching up.

So perhaps one could say such areas don't stay inexpensive for long, and their character changes accordingly. Fair enough. It's still a wonderful place to live--and there's something about redwoods that takes the edge off.

Posted August 26, 2003 6:16 PM by Dave Trowbridge

Ahh, I apologize for my ignorance. I will address you correctly from now on. My bad.

Posted August 26, 2003 8:40 PM by The Littlest Cynic

Teresa:

Bruce, the changeover to a non-elective two-income lifestyle has been an underreported story, as has the gradual (but significant) erosion of our disposable free time. On the days when conspiracy-mindedness gets the better of me, I figure it's part of the general plot to bring back the class system.

Did we ever really lose it, though? Or just weaken its hold on the institutions of power for a while?

-l.

Posted August 26, 2003 11:33 PM by LauraJMixon

> Up popped a complete list of American society,
> classified by marketing category, from #01 Blue
> Blood Estates to #62 Hard Scrabble

Lovely! Reminiscent of Stephen Bury's _Interface_, which had some gorgeous lists of marketting demographics. You just can't trust fiction to be stranger than truth.

Posted August 27, 2003 1:32 AM by Steve Taylor

Don't feel bad, Anaktuvuk Pass, AK (99721), population ONE (1), 100% female, also rated an 8 for crime risk.

Emmonak (99581), also pop 1, has a 100% vacancy rate. Guess the poor girl is homeless. She's also 54.17% Hard Scrabble and 45.83% Blue Highways.

Ambler (99786), pop 4, is -0.51% female and 100.51% male. Guess those Alaskan types are manly, indeed.

Enough for now, back to work. :)

Posted August 27, 2003 7:48 PM by B

When was the data last updated? I lived in Bed-Stuy for a short time while freelancing, and the neighborhood is, well, "turning."

I saw a photograph of my friends' brownstone (two blocks south of Pratt Institute) from twelve years ago: the stoop is invisible, as it's covered with rats. Across the street (in another photo) you can see where everyone was dumping their garbage. It's a pile that reaches to the windows of the second story of the burnt-out shell next to it.

And last summer, the building on the corner was in the process of being gutted and restored to original condition. The apartments inside were already sold, to the tune of over $100,000 each.

Change marches on...

If there isn't a database that shows the progression of change through neighborhoods over time, there should be.

Posted August 28, 2003 9:59 AM by Jazz

My youth was in a privintial town in a provintial part of the country & boy was it dull. Factorthat into the equation.The cheap thrills of a city are well fun when your 17 and I'm not sure I could do without them at 47

Posted August 29, 2003 11:33 AM by DM SHERWOOD

Great link.
I think the accuracy is dependent on the size and homogeneity within your zip code (obviously). It nailed mine. Interesting stuff:

Nationally
- the median income (indiv.) is $18,000
- percentage w/ college degrees are 15%

My zip 94707
- the median income (indiv.) is $15,000
- percentage w/ graduate degrees is 33%
- median house price is almost $500,000!
- median age of house was 50 years!
- crime is above the national average
- population density is 9X the national avg.

Year-round beautiful weather, no need for air conditioners...ahhh, the Bay Area.

Posted August 30, 2003 3:03 AM by andrew

We did this in Plokta some issues ago; or, more correctly, we did it on the envelopes, with the explanation 'Dr Plokta Knows Where You Live' and social classification codes for each of our readers. Or at least, each of our readers who lived in the UK or US; the Canadians got one between the lot of them, as did the Australians and the people with post office boxes, and I think we made one up for each of the remaining ones. I must still have the database somewhere. Looking them all up was good fun, though exposed many of the inadequacies of our mailing database. Brief explanation, including wholly invented demographic code for Tobes, here; subsequent letters, mostly denying everything, here.

In the UK, the prevailing codes are much smaller than first-half-of-the-postcode, but bigger than entire-postcode (which normally equates to one side of one street). There were about 60 in total I think, for each of the countries, and the different countries were very different in terms of what they emphasised.

Posted September 1, 2003 7:34 PM by Alison Scott

Teresa on traffic: However, this is not actually going to help, because the newly widened roads will make it possible for people to commute from even greater distances. The broader the road, the more commuters it can feed back and forth every day between the city and the expanding suburban frontier.

The math behind this is interesting. Additional lanes increase capacity linearly, but a road with more capacity collects traffic from a wider area. Wider roads are also more likely to be used for longer trips, so they collect even more traffic. These factors mean that traffic increases exponentially relative to the capacity. Not that roads are inherently bad, just that if a major road was worth building at all, it probably will be jammed at least once a day.

Mass transit does not escape the exponential traffic / capacity relation. Transit drives growth, and the first major wave of urban expansion was rail-based. The saving grace of transit is that when the system fills up to capacity, it keeps running at capacity instead of jamming and breaking down (unless you are on MUNI). Having to wait for another bus is a drag, but if you get on the freeway, and you are that one last car that caused the traffic flow to change phase into a solid, you will wait a lot longer.

The fractal dimension of a road network is greater than one, but you can't pave everything, so it will always be much less than two. The fractal dimension of a developed area is close to two for urban sprawl, and less for transit oriented development (because of the clustering). The difference between the dimensionality of the developed area and the dimensionality of the road network, gives you the traffic growth factor. The higher the traffic growth factor, the faster roads will become saturated again after capacity is added.

Posted September 7, 2003 11:49 PM by Tom Becker