A plow, now!

Remember that time last weekend when it snowed a whole bunch and we all waded through four-foot drifts to get out of the house, went skiing down city streets, or else enjoyed lots of bacon and beer in the comfort of home?

The geography of this event can be summed up as “snow everywhere,” but besides vague or anecdotal accounts there is some data to look at. Beyond the snow itself, we can track the progress of storm response. The city of Boston set up a web map that tracked the live location of snow plows and their ground covered, which would be excellent for a summary map, but in an excessive triumph the site was overloaded and had to be shut down because the demand was interfering with the city’s ability to track plows itself. (Otherwise, we could make a nice map like our friend Derek Watkins did for NYC.)

So instead of the plows, how about people clamoring for plows? The city compiled a list of snow-related calls from the ever fascinating constituent services request data, around 15,000 in total from Friday through Monday. Most of these calls are labeled as requests for snow plowing. While there are a couple of hot spots (er, cold spots?) geographically, the temporal aspect is interesting. Who was impatient on Saturday morning? Who still had a snow-clogged street on Monday afternoon? (To say nothing of the problems that have lingered for days more.)

Boston snowstorm complaints

And in animated form through the weekend (pretty much 100% copied from this map):
Animated map of Boston snow-related complaints

Conclusions? Well, none over time, really, except that peak complaint time everywhere was between Saturday afternoon and Sunday afternoon. Overall, with a couple of exceptions there appears to be something of an inverse relationship between population density and request density. In high-density areas, plowing a single street satisfies more people than in a lower density area, where plows have to cover more ground to reach everyone. Some of the highest density areas are also important business districts too, and may have been a priority. Also, residents in the high-density areas are probably less likely to drive regularly, and thus there are fewer urgent demands for clear roads. Or maybe something else is going on. What’s your opinion or experience?

Whatever happened with snow removal, thumbs up to Boston for keeping the complaint data open and up to date!

Posted in Seasonal | Tagged , , | 1 Comment


New Yorkers want to own everything. Even the sunset once in a while: you may have heard of the phenomenon known as Manhattanhenge, when the setting sun aligns with Manhattan cross streets. It occurs twice a year—really four times if you count the rising sun, which is apparently too early for anyone to be in the mood to talk about.

Here in Boston, like most everywhere else in the country outside New York, we’re at the disadvantage of not having the insane skyscraper street canyons that make Manhattanhenge remarkable. But perhaps our lack of any overall street system makes up for it; Manhattan gets only a couple of days each year, while the sun rises and sets over different streets on a number of dates here.

Boston November sunset

For one thing we have MIThenge, when the sunset aligns with MIT’s Infinite Corridor. A phenomenon involving the sun’s reflection on the Hancock tower has also been noted.

As for streets, well, our straight streets don’t tend to be very long, but here’s a map of a few of those that theoretically offer views of the sunrise or sunset. (Click it for bigger, better size.)


This map is only based on geometry and a sunrise/sunset calculator, so it’s no guarantee of a good show, but these streets and others may be worth checking out on the right days. I think something fun could be done with “Commhenge” and the Prudential observation deck—as you may have noticed, the street points right at the Pru. You may know some other good streets from your own experience, so please share!

Go watch those sunrises and sunsets, Boston. Life is short; enjoy the colors!

Boston streets analemma

Tim said there should be a cool analemma infographic. I’m not sure I succeeded here, so let’s hope he’ll come to the rescue.

Posted in General | Tagged , | 10 Comments

Blue and Bluer: Massachusetts and Boston 2012

2012 Boston and Cambridge presidential votes

You’ve seen tables and maps of counties and towns with the numbers for President Obama and Governor Romney in November’s election. But you may not have seen the detailed story in Boston. Well, that’s what we present above, in collaboration with friend, colleague, and scholar Kirk Goldsberry, a geographer currently at Harvard whose Texas election maps made the rounds recently (although lately he’s known for his NBA visual-spatial analytics). This is about as detailed as one can get: it’s a map showing a dot for every Obama and Romney vote in Boston and Cambridge. Dots are distributed randomly within precincts, minus areas that are unpopulated according to several data sources—note that although this is a 1:1 map, the dot locations do not reflect actual individual voter locations.

The first thing you’ll notice is that these cities are very, very blue (i.e., Democratic). The split in Boston was 79% to 19% in favor of Obama; in Cambridge it was 86% to 11%. Given the blue baseline, it’s interesting to pick out the areas of Romney support, but even there you won’t find much in the numbers. According to early data, Romney carried only two precincts in Boston (in South Boston), one of which gave him a slight majority and the other of which he won with a plurality on a margin of four votes. He outperformed his national vote share (about 47%) in only one additional precinct, and he lost by less than a 10% margin only in a further three. In other words, you might say that Romney performed reasonably well in only six of 253 precincts in Boston. Across the river in the People’s Republic he didn’t have a prayer anywhere: his best precinct gave him only 18.7% of its vote.

Top Obama and Romney precincts in Boston/Cambridge

The “strong” Romney areas were in South Boston, and you can see some moderate support in places like the North End, Charlestown, Beacon Hill, Back Bay, West Roxbury, and part of Dorchester. Looking at the dot map you may start to think of race and ethnicity patterns in the city, and indeed there are very real correlations. As was the national trend, Romney’s strongest support was in areas with a mostly white population. Obama’s support, meanwhile, was exceptionally high in areas with mostly black population. It’s harder to see on the map because of the overall blue-ness, but among the 61 precincts with a majority black population, Obama averaged 96% of the vote.

White population vs Romney vote, Boston and Cambridge by precinctBlack population vs Obama vote, Boston and Cambridge by precinct

Compare the vote map with a map of a few race and ethnicity categories below. Whether we like it or not, voting patterns are inextricably linked to—and to some extent predictable by—demographic patterns, and not just the super basic racial patterns we’ve mentioned here. To fully understand the electoral landscape of the city, there are many other correlations to investigate.

Boston and Cambridge race & ethnicity

Stepping back to the state level, Massachusetts is near the top of the list of “blue states,” reliably voting Democratic in national elections and being home to the “Massachusetts liberal” bogeyman. Take a look at a county-level map of the 2012 presidential election and you’ll see that Massachusetts was one of only four states in which President Obama won every county. (Two of the other three are our neighbors, Vermont and Rhode Island.)

But counties—largely meaningless non-entities here—in one race paint a simplistic picture. What are the actual partisan voting patterns within the state? Here’s one way of looking at it:

Massachusetts 2012 election composite

This is a visual technique I first tried on 2006 elections in Ohio. It combines red–blue spectrum maps of several different races into a single, averaged image. In this case we have the major national and state elections of 2012: President, US Senator, US House, Massachusetts Senate, and Massachusetts House. Essentially, maps were overlaid on one another and average blue and red pixel values were calculated. (They were then converted to something more like cyan and magenta for aesthetic and legibility purposes.) It’s not a proven technique and we can’t claim that it accurately represents regional political sentiment, but it does roughly represent voting patterns in the recent election.

What probably strikes you first is a few particularly red areas. These are home to the several state races that had an unopposed Republican candidate, and they only stand out because the rest of the state is so blue. If for some reason you doubt Democratic strength in Massachusetts, consider this: in the recent election 3 of 9 US House districts, 24 of 40 state Senate districts, and 90 of 160 state House districts had no Republican candidate. That’s right, a majority of state legislative districts had no Republican candidate. (Not all of these had unopposed Democratic candidates; some had third party or independent challengers.)

Bottom line? Whether it’s the city or the state, this is blue country, my friends. But you knew that already.

Data sources:

Posted in Political | Tagged , , , | 5 Comments

Handy turkey maps of Boston

(Or, “Bostonography goes back to kindergarten.”)

Sidewalk turkey
This town is no stranger to turkeys. Above is a rude, slow pedestrian near Harvard Square, for example.

Wait, maybe Harvard Square is a turkey! We tried our hand at mapping this thought:
Harvard Square turkey hand map

Actually, the whole region is a turkey. And look, its head is right where the original Thanksgiving was! As they say on a different North Shore, shaka brah:
North Shore/South Shore turkey hand map

Step back and we can see that the whole of Massachusetts, in fact, is a turkey:
Mass turkey hand map

The Boston wharves turkey sports proudly struts around in Boston Common underpants and, naturally, needs corrective lenses to see Spectacle Island:
Boston wharf turkey hand map

And what do you see when you look at the Back Bay skyline? Why, a turkey of course:
Back Bay skyline hand turkey

Where do everyone’s favorite turkeys play baseball? In America’s Most Beloved Turkey:
Fenway Park turkey hand map

Please, show us maps (in turkey form) of your favorite places!
We’re looking at you, Mr. Sullivan!

Happy Thanksgiving, folks.

Posted in Holiday | Tagged , , | Comments Off on Handy turkey maps of Boston

Visualizing Hubway, for real this time

How convenient. As soon as we got around to our previous post, in which we mapped some Hubway bike sharing data, a new visualization contest started. This time, MAPC and Hubway invited people to visualize every Hubway trip to date (October 1, 2012), providing a record for each that included the start and end points, the date and time, the particular bike used, and some information on the user. Juicy, delicious data, folks.

There’s a ton one can do with this data, but one only has so much time. I took a swing at it on behalf of Team Bostonography by building an interactive map that allows you to filter trips by about ten different variables and view where trips occurred under the specified conditions. The map queries a database for the given combination of filter settings and returns trip counts for each station to every other station. On the front end it’s built with Leaflet and uses MAPC’s basemap tiles. (<greasy salesman>And the basemap tiles are a product of my company; we worked with MAPC to develop those styles back in the spring.</greasy salesman>)

Hubway Trip Explorer

You can view trip volumes (represented by line width and/or color) either for individual stations or for the whole system at once. The available filters are these:

  • Time of Day: A time range in which the trip either began or ended
  • Date of ride: Months and days of the week on which the trip occurred
  • Day or night: Whether the ride occurred during daylight
  • Duration of ride: How long the ride took
  • Member type: Registered annual member or Casual (1 or 3 day) member
  • Sex of rider: Uh, sex of rider
  • Age of rider: Yup
  • Home of rider: Where the rider lives, from a selection of several zones
  • Precipitation on day of ride: Whether there was any precipitation that day
  • Average temperature on day of ride: The 24-hour average temperature for the day

In order to account for the the fact that some stations (e.g. everything in Cambridge and Somerville) are much younger than others, the overview maps represent trips per day of station activity, and the individual station maps show volumes as a proportion of the total trips for that particular station. It’s easy to see the potential of some of those newer stations—one or two of the Harvard Square stations, for example, actually show up strongly enough even as raw counts despite being much shorter-lived than the original set of stations.

To highlight a few salient or interesting bits of information coming from the data, I also made a short series of infographic-like things with maps and charts. Some of them are admittedly not super informative or not charted in ideal ways, but they nevertheless provide glimpses of some avenues to explore in the data. See them below or compiled into a PDF.

Hubway Snapshots
Hubway Origins & Destinations
Hubway Daily Cycles
Hubway: Getting Green
Hubway Demographics

Those graphics include much of what I found interesting from the data, so I won’t try to talk through many of my own conclusions. I will say that looking at certain time slices (times of day and/or days of the week) tends to be the most revealing—especially combined with the home zip code of the rider. We’re all about geography here, after all!

Again, there are gazillions of questions to be asked and answered with the data, and we’ve only gone after a few of them. Be sure to look through all the submissions to this contest to see the awesome things people have done here and the information they’ve uncovered!

Posted in Transportation | Tagged , , | Comments Off on Visualizing Hubway, for real this time

MBTA + Boston Bikes Visualization

Back in January, the MBTA and City of Boston sponsored a developer challenge to develop new apps and visualizations making use of real-time bus locations and Hubway bike sharing station data. For the visualization challenge, they provided three days’ worth of bus GPS location reports as well as point-to-point trips in the Hubway system. Just before the data become the officially out-of-date age of one year old, which is too old especially in light of the recent Hubway expansions, we figured we should post what we put together for the contest.

We worked with Kirk Goldsberry—Geography professor at Michigan State University, current visiting scholar at Harvard’s Center for Geographic Analysis, and frequent proposer of cool map ideas, known best these days for his NBA spatial/visual analysis—who had also been interested in mapping this kind of data and who had some good ideas for how to approach it. Dr. Goldsberry’s plan was to focus not simply on the point data but rather on the broader paths and areas affected by these data. In what areas might we find activity associated with buses? What parts of the city see the most Hubway riders on their streets? We weren’t able to address the idea fully, but it was from this angle that we viewed the data, especially the Hubway trips.

Hubway and MBTA bus map

One part of our submission was a set of overview graphics. (Download it in higher resolution as a PDF if you wish.) Above is an overview map of the bike and bus trips in the center of town, in our classic style of crazy jumbled lines. The bus locations, shown in yellow, are straightforward plots of the GPS coordinates. The Hubway trips, each one shown in semi-transparent blue, are speculations about the routes riders may have been, based on the origin and destination stations. (More interesting maps and explanations of that in a moment!) The final thing on this map is orange points indicating where buses reported doors open near Hubway stations, presumably where passengers boarded or alighted. It’s a feeble way to get at possible connections between the bus and bike systems, indicating the opportunities to transfer from bus to bike or bike to bus. Altogether it’s hard to say whether this map reveals anything in particular, but it shows the coverage of the systems in question, anyway.

Hubway and MBTA bus trips: Oct 9–11, 2011

Hubway trip durations

Following the overview map were two quick charts, above. The first shows the number of Hubway trips and bus location reports per hour over the three days. There’s a different pattern on each day: one is a Sunday, the next is a holiday, and the third is a regular weekday. The third shows a predictable bimodal distribution with intense morning and evening rush hours, while the others peak in the middle of the day. Our favorite (but troubling) thing revealed by this chart is the small Hubway spike around 2 AM Sunday morning: presumably people are pouring out of bars and hopping on Hubway bikes. The second type of chart is a matrix of all the Hubway stations, showing the median trip duration from each station to every other station. The two take-home points here are that yearly registered Hubway members take shorter trips than the “casual” (one or three day) members and that the longest trips tend to be the ones that start and end at the same station. In both charts we can see a lot of non-commuting usage of the system, which raises interesting questions of exactly what role bike sharing plays in the overall transportation system.

A final question we wished to address was that of common routes taken by Hubway users. Without knowing the exact location of each bike, via GPS or RFID, we had to extrapolate potential routes based on reported “departure” and “destination” locations. This is an important point—these routes are “potential” or “possible”, not actual. They are based on Google Bicycling directions via the Google Directions API. After calculating all potential routes, we sliced the data up by routes “to” and “from” each station. Once we had all of these routes, we could count the number of potential routes at each road segment. The last step of production was made much easier by using Matt Ericson’s MultiExporter .jsx script for Adobe Illustrator. The result is a set of 57 small multiple maps highlighting areas of Boston with the highest potential for Hubway activity. Some areas, like the corner of Beacon & Arlington Streets, could have had as many as 650 bicyclists pass through over the course of the three days when the data was collected. Meanwhile, the largest number of bicyclists to leave any location headed in the same direction was 230. That number of people likely headed southwest from South Station. Take a look at the maps below or (warning: gigantic file!) download the full resolution versions here.

[portfolio_slideshow pagerpos=disabled size=full include=”1844,1845,1846,1847,1848,1849,1850,1851,1852,1853,1854,1855,1856,1857,1858,1859,1860,1861,1862,1863,1864,1865,1866,1867,1868,1869,1870,1871,1872,1873,1874,1875,1876,1877,1878,1879,1880,1881,1882,1883,1884,1885,1886,1887,1888,1889,1890,1891,1892,1893,1894,1895,1896,1897,1898,1899,1900″]

There are a lot of other ways to look at the data, some of which we tried but didn’t publish, so we’d be happy to whip up other maps if there are requests and we can remember how we did any of this. Otherwise, dear readers, do you see any especially interesting patterns?

Credit goes to Tom Auer for first suggesting the approach of looking at volume by street segments.

Posted in Transportation | Tagged , | 4 Comments

Street, Road, and Avenue

Our friend Derek Watkins recently made a map which he claimed copied something I once did (although he did it much more beautifully), so why don’t we return the favor? He also made a few other maps of Portland, Oregon for the new extent(PNW) blog he’s running along with Nick Martinelli—you should check it out because it’s kind of like the Bostonography of the Northwest—and although those additional maps seem to be inspired by formerly local cartographer Bill Rankin, they have in turn inspired a few quick maps right here. (Because honestly we haven’t found much time lately for maps that involve more than copying someone else.)

Anyway, what we’re interested in is street names again. This time it’s the type of street: Street, Avenue, &c. Some cities, like the Portland that Derek mapped, have very orderly patterns of streets and avenues, but “orderly” is not a word that is often used with Boston streets outside the Back Bay or Southie. Still, there are patterns. Have a look:

Boston Streets

Boston Roads

Boston Avenues

Streets, Roads, and Avenues seem to account for the vast majority of street names in the immediate Boston area. Streets clearly occupy most of the central cities, while Roads are a bit more prevalent in the suburbs, especially toward the west. Avenues, meanwhile, dominate some small pockets and comprise a few notable long roads, but otherwise are not nearly as popular.

Having looked at enough maps of the Boston area day after day, something that struck me was how well the “Streets” map seemed to correspond to population density patterns, which can also be seen in aerial images. While we haven’t gone so far as to draw any statistically meaningful conclusions, a visual glance does suggest decent correlation. Here you can see the Streets again with population density greater than 10,000 per square mile shown in yellow in the background:

Boston "Streets" vs population density

It seems clear that “Street” is the type that dominates urban areas much more than elsewhere. To me this makes some sense. The several types suggest different characters: Roads are long paths between disparate places; Avenues are broad, less organic thoroughfares within a city; and Streets are the smaller urban ways. What do you think? Does this hold true in the Boston area?

Posted in General | Tagged | 11 Comments

You can’t get here from there

There’s an excellent Far Side cartoon of which I’m often reminded around here: a man in a car has asked a farmer for directions from Point B to Point A, as indicated by the map in the man’s hand. The farmer is stumped, saying (to paraphrase) “most folks want to go the other way.” This seems to happen to me a lot when in a car or on a bicycle, i.e., when subject to traffic regulations. I have a clear picture of how to get somewhere but not how to get back, thanks to Boston’s uniquely bizarre system of twisting roads and one-way streets. Sometimes you can’t get there from here, but sometimes you can—you just can’t get here from there.

Here, for example, are a couple of routes to and from a point near my home in Cambridge that have given me pause more than once.

It’s super quick and easy to get to the main Cambridge library on a bike. Getting back takes some thinkin’. Google’s suggested routes from A to B and B to A don’t share any road segments at all.
A to B versus B to A: Cambridge Public library

“Sure. I’ll drive you to the South Station bus terminal,” I say, helpfully. Then I drop off my friend and sit amid the taxis and realize I don’t know the best way to get back home.
A to B versus B to A: South Station

It’s not that the routes are objectively difficult; it’s that unlike many other cities where you can mostly retrace your steps, the return trip here often requires knowing a whole different set of directions. It’s not hard to figure out these days, either, what with smarty-pants phones. But to know offhand how to get to any one point from another AND back again is something to be proud of, I think, as it demonstrates a mastery in local geography, which doesn’t come easily. (If you’re a cartographer, you crave this point of pride.) This should also indicate what a difference there is in this town between walking and using wheeled vehicles, at least if you more or less obey traffic patterns on a bicycle, and it explains why I, for one, am so bad at giving directions to drivers like the man in the Far Side cartoon even if I know exactly how to walk there. Boston-area navigation is difficult enough on foot; throw in one-way streets and turn restrictions and you’ve got a vehicular nightmare. (Don’t get me wrong: I’m as hip as the rest of you and almost never drive a car in town, but these problems occur once in a while.)

Anyway, above are just two of my personal examples. The real question is: what are your own favorite examples of knowing how to get from Point A to Point B, but not Point B to Point A? Do share with all of us, and maps are encouraged!

Posted in Transportation | Tagged , | 35 Comments

Crowdsourced neighborhood boundaries, Part One: Consensus

UPDATE! We’ve got a new and better version of the neighborhood mapping project! Head on over to bostonography.com/hoods!

OLDER UPDATE: There’s also a newer map of what’s described below.

As you may recall, we’re running an ongoing project soliciting opinions on Boston’s neighborhood boundaries via an interactive map. We want to keep collecting data, but we’ve already received excellent responses that we’re itching to start mapping, and when we hit 300 submissions recently it seemed like a good enough milestone to take a crack at it. (That’s actually 300 minus some junk data. If you offer the ability to draw freeform shapes, some people draw random rectangles and triangles, and some people draw… er, other long, tipped objects.)

There are many questions to be asked here. Where are the areas of consensus? Where are the disputed zones? Where are the no-man’s lands? &c.? Let’s tackle these one at a time in a series of posts and maps. Today we look at consensus.

But first, let’s check out a raw picture of the boundaries everyone has drawn. This map, which is clipped to the city limits, shows that the word “tangled” is apt in reference to neighborhood boundaries. Some lines are strong as a result of many people drawing in the same place, but many others are all over the place.

Crowdsourced neighborhood boundaries

Based on those strong lines we can kind of see where there is decent agreement on boundaries, but we can quantify this for all of the neighborhood extents. What we have is 21 sets of polygons, each set representing the shapes people drew for one of the neighborhoods in our survey (the main neighborhoods except for the physically isolated Charlestown and East Boston). Within the set for each neighborhood we can look at the amount of overlap between polygons and consider overlapping areas to be areas of agreement between two or more respondents—as many respondents as there are overlapping polygons in that area. In this exercise we’re interested in those counts. We could find the area in which all polygons overlap, for instance, which would represent the area that 100% of people agree is part of the neighborhood. Total consensus usually doesn’t show us much or doesn’t exist, though, because of some especially bad data (some people just didn’t have a clue, although we tried to weed those out) and, well, because definitive agreement probably is impossible anyway. But we can still find the areas where there is moderate to strong agreement—at least among the self-selected people who responded to the survey.

We’ve chosen to look at areas of greater than 25% agreement, greater than 50%, and greater than 75%. They’re handy numbers that seem represent a notable minority, a majority, and a strong majority. So let’s get to the maps. We’ve laid down an arbitrary hexagonal grid across the city and for each cell tallied how many respondents’ polygons it intersects in each neighborhood, then divided that count by the total number of respondents for the neighborhood, which ranged from a little over 40 to 150ish. Here, then, are the neighborhoods in almost alphabetical order, with the number of submissions noted for each.
(Note: when we started these, hexagons had slightly more carto-hipster cachet. These are not really the hexagonal bins that have been recently popular—albeit legitimately useful—but rather kind of a coarse raster approach that looks nicer than squares.)

Crowdsourced Allston extent

Allston (n=120). The first map, I think, demonstrates how closely the categories tend to follow certain streets in all the neighborhoods. Market Street and Everett Street mark the western edges of the 25% and 75% categories, respectively, with 50% a little less defined on that side. Each level then cuts east and south in a couple of steps. It’s interesting that the commercial center of Allston, the Harvard Ave corridor, is only barely within the area of strongest agreement. A neighborhood “center” isn’t always in the middle!

Brighton (n=109). Not surprisingly, Brighton’s eastern edge matches up fairly well with Allston’s western edge. They’re not exactly interlocking jigsaw puzzle pieces, however, because not everyone who drew one neighborhood drew the other, nor were they necessarily careful to make the borders match precisely.

Back Bay (n=139). There are some hard lines here. Nobody goes past the Charlesgate, or past Mass Ave south of the Pike. An interesting section is that south of Huntington Ave. It has the character of the South End but Back Bay parking permits, presumably because the Southwest Corridor partially cuts it off from the former. Still, it doesn’t enjoy majority Back Bay status in this map.

Bay Village (n=80). Not much room to mess around in this tiny neighborhood. The majority stays south of Stuart Street, with the supermajority between Arlington and Charles. The part north of Stuart, although small relative to the whole city, will be interesting to look at later as a no-man’s land; it seems hard to classify even though it has decent Bay Village representation here.

Beacon Hill (n=145). This may be the neighborhood with the most consensus. Everyone agrees on the southern, western, and northern boundaries. On the east side, the question is only whether Bowdoin, Somerset, or Cambridge/Tremont Street is the edge.

Chinatown (n=113). It’s a little surprising how small the greatest agreement is here, at least to me. I’d probably draw something like the 50% shape. Note that even the Chinatown gate is just outside the 75% region. Perhaps it’s just hard to know which street is which on a map of this dense area, leading to more uncertain shapes from our respondents.

Dorchester (n=54). Dorchester is large enough that we probably need more responses in order to see more agreement. (This is the most populous neighborhood, of course, but it doesn’t have as many responses as some others, probably in part because it’s harder for non-residents to grasp.) Still, we have some good edges here: Blue Hill Ave for 25% and the Fairmount line for 50%. We’ll see those again in the Roxbury map.

Downtown/Financial District (n=118). “Downtown” is fun to try to define, but we’ve wrapped it up here, perhaps not entirely appropriately, with the Financial District. Besides the easier-to-locate Financial District, to some degree this looks like a process-of-elimination neighborhood: if it isn’t Chinatown, the North End, &c., it must be downtown. Maybe that’s because it’s least residential neighborhood of the bunch, so there’s not really a strong identity.

Fenway/Kenmore (n=128). I’m a bit surprised by how many people consider the Fenway neighborhood to end, well, at the Fenway itself on the west side. A minority of respondents here consider some of the Colleges of the Fenway to be actually in the neighborhood. What they’re thinking, presumably, is that this is the Longwood Medical Area and that the LMA is not part of the Fenway. It’s an intriguing zone because we left it off our neighborhood list, forcing people either to include it in an adjacent neighborhood or leave it blank. But we can talk more about that in another post.

Hyde Park (n=42). We had few enough people drawing Hyde Park (or enough of them were clueless) that there is nowhere with more than 75% agreement. And to be honest I’m ignorant enough about Hyde Park that I can’t comment much on this one. It’s as outlying as outlying neighborhoods get in Boston. Anybody see anything interesting?

Jamaica Plain (n=74). There are ome fascinatingly well defined edges here. On the east side, people thing the neighborhood ends either at the Southwest Corridor, at Washington Street and then in a line up to Jackson Square, or at Franklin Park et al. and Columbus Ave. The 75% line on the west and south nicely follows the parkways, too (otherwise it’s just the city limits).

Leather District (n=76). Tiny neighborhood #2, easily lost between Chinatown and downtown. But hey, it has its own parking permit! There’s not a lot to talk about at the scale of our maps this time; some folks have it bleed over to Fort Point Channel but otherwise it’s mostly confined to a few blocks.

Mattapan (n=45). The 75% agreement is a very odd and small sliver, probably again due to a lack of data, or at least good data. But there are some clearer definitions at the other levels, for instance along Morton Street. As with the other southernmost neighborhoods, I can’t say much about this one personally. Comments are welcome!

Mission Hill (n=82). There’s a lot of agreement here below 75%, with the only real difference being the section below Francis Street. 75% agreement is only in a small area for some reason, not extending far from Tremont Street. Note that fewer than 25% consider the Longwood area to be part of Mission Hill.

North End (n=151). Along with Beacon Hill this is one of the best-defined neighborhoods, being easily recognized by residents and outsiders alike. The only uncertainties seem to be parts of the Bullfinch Triangle and Christopher Columbus Park (where, it’s worth pointing out, there is a sign welcoming you to the North End), and all the piers. But even those places are mostly above 50%.

Roslindale (n=41). Wait, is Roslindale a real place or just the name of an RMV branch? Poor Rozzie had the worst data of the lot. It’s just, you know, down there somewhere in between other neighborhoods. What you see here is after some purging of especially bizarre data that were throwing everything too far out of whack. No strong majority here, although the 50% and 25% area are reasonably clearly defined in places. Some of the rest, though… who knows. There’s even a weird floating piece of 25% territory.

Roxbury (n=64). I personally think Roxbury is the most interesting neighborhood to watch in all this. It has uncertainties on at least three sides, and it and its neighbors seem to have reasonably strong identities to both their residents and outsiders (in contrast to some places that only have the former). The South End-Roxbury border debate is well-publicized, and the JP border gets fuzzy too. Mission Hill isn’t always even considered separate from Roxbury, and the border with Dorchester varies widely, as already mentioned. Apart from some weirdness in the south part of 75% land, there are some interestingly clear-cut camps here.

South Boston (n=92). There seems to be a lot of unity in Southie, apart from the Seaport area and a bit of the other end. Maybe it’s just because there’s not much to compete with: there’s water on most of three sides and something of a dead zone on the other. South Boston is a neighborhood that has grown considerably over the years thanks to landfill, and this map is suggesting that people’s idea of the neighborhood has mostly, but not entirely, grown with it.

South End (n=116). Well, a lot of this has already been touched on by way of talking about surrounding neighborhoods. There’s a lot of agreement on Mass Ave as a border, despite the parking permit zone that crosses it (I think). Otherwise, the edges seem to match what we see in the adjacent neighborhoods, and within there is strong agreement.

West End (n=114). I suspect that to old timers and casual scholars of Boston, the West End doesn’t even exist; it was torn down in the 1950s. But something is there: a weird land of hospitals and towers in the park. Everyone agrees on a tight loop of roads defining the West End, but beyond that it’s harder to say.

West Roxbury (n=43). The last alphabetically and somewhat fittingly one of the most distant neighborhoods, this one again suffers from a lack of good data. It has the fortune of meeting the city boundary for much of its edge, but away from that there are some funny shapes. And once again I don’t have much to add. We need more data and your thoughts!

Boston neighborhood certainties

What does it mean?

Although we talk a lot about boundaries, this post included, the maps here should also remind us that neighborhoods are not defined by their edges—essentially, what is outside the neighborhood—but rather by their contents. And it’s not just a collection of roads and things you see on a map; it’s about some shared history, activities, architecture, and culture. So while the neighborhood summaries above rely on edges to describe the maps, let’s also think about the areas represented by the shapes and what’s inside them. What are the characteristics of these areas? Why are they the shapes that they are? Why is consensus easy or difficult in different areas? What is the significance of the differences in opinion between residents of a neighborhood and people outside the neighborhood?

We’ll revisit those questions in further detail in future posts, and also generate maps of other facets of the data. Next up: areas of overlap between neighborhoods. Here we’ve looked neighborhood-by-neighborhood at how much people agree, so now let’s map those zones that exhibit disagreement. Meanwhile, thanks so much for all the submissions for this project; and if you haven’t drawn some neighborhoods, what’s your problem? Get on it!

Posted in Geography, Projects | Tagged , | 90 Comments

Typographic Boston and Massachusetts

Typographic maps—that is, maps constructed mostly by typography—are quite the rage these days. I think I got in on, well, if not the ground floor then a fairly low level of the fad some four years ago with a small type-based map of the Prudential area in Boston (shown below), which later grew into a whole line of typographic map posters. Besides those posters, a number of other typographic maps of Massachusetts and/or Boston have since shown up on our radar, so here they are. Click ’em for the source pages, where there are larger images and usually prints for sale. Do let us know of any that were left out!

Ork Posters - BostonAs far as posters go, probably the “original” city typographic map is that of Ork Posters, which fills neighborhood areas with text bearing the neighborhood names. I vaguely recall that an early version of this used an “Allston-Brighton” label that essentially meant that their labels were switched, because “Allston” is first in the name but Brighton is first on the map. Or maybe I imagined that.

Typographic Boston party invitation mapThis is the first one I did, in 2008. It’s an invitation to a departmental party during the Association of American Geographers conference, which was in Boston that year. My roommate, who was in charge of the planning, commissioned a map and typographic map was my idea. Later I expanded this style into more of a full city map.

Welcome to Boston by Michael WensteinBoston Marathon map by Michael WeinsteinLocal designer Michael Weinstein has a couple of text-based Boston maps. The first, called Welcome to Boston, shows inner neighborhood names (plus Brookline and Cambridge) in different stylized ways, along with the T lines. Prints are for sale. The second shows the towns and streets along the Boston Marathon route, in a similar style.

Boston typographic map by UrbanFootPrintDesignThis map from UrbanFootPrintDesign is solidly filled with horizontal and vertical text of several different styles (I haven’t taken a close look at how many typefaces are here), showing streets names, place names, and landmarks.

Boston typographic map by Jessica HarringtonUsing a similar technique but somewhat different style is this one by Jessica Harrington. She’s got a few other maps of regional interest, too, so poke around that Etsy shop linked in the image.

Typographic transit map by Fadeout DesignA typographic transit map from Fadeout Design shows nothing but the MBTA lines (yeah, Silver Line included) as text. The text is the names of stations. I think I’ve seen something like this but including a bunch of bus routes, too. Was that also just imagination?

Massachusetts towns text map by Molly MattinHere’s colorful one of the state filled with town names, by Molly Mattin. This and some of the other examples are not exactly typographic maps, of course, since they are hand-drawn illustrations.

Massachusetts counties text map by  Crystal PowellFinally, a county name map illustration from CAPow (Crystal Powell). I like the way that some of the coastal letters have the rugged edge of the actual coastline.

Posted in General | Tagged | 3 Comments