Friday photos: Jane Jacobs in Georgetown

 

Farewell Georgetown, C&O Canal

Grace Street, between Chaia and Dog Tag Bakery

There’s one passage in Death and Life where Jane Jacobs singles out the District for praise. Not surprisingly, it’s for the back streets of Georgetown, which (of course) house some of my favorite little eateries.

In city districts that become successful or magnetic, streets are virtually never made to disappear. Quite the contrary. Where it is possible, they multiply. Thus in the Rittenhouse Square district of Philadelphia and in Georgetown in the District of Columbia, what were once back alleys down the centers of blocks have become streets with buildings fronting on them, and users using them like streets. In Philadelphia, they often include commerce.

Cady's Alley shared street

Cady’s Alley, a shared street, with Leopold’s Kafe

Georgetown passages

C&O Towpath at 31st St., with Sushi to Go. Baked & Wired, Il Canale, and Snap Cafe are around the corner.

Georgetown passages

The newly opened Sovereign lies just behind the 100% corner at Wisconsin & M, at the end of an unnamed alley. More retail is planned for the interior of this block, between M, Wisconsin, Prospect, and Potomac.

What Would Jane Jacobs Do about zoning?

Tomorrow would have been Jane Jacobs’ 100th birthday, and so it’s a fine time to reflect upon her magnificent legacy of (empirically correct) ideas. Unbeknownst to many of her fans, she has a significant built legacy. 20 years ago, Toronto asked no less than Jane Jacobs about how to rezone two renewal areas on either side of downtown.

Distillery District

Toronto’s Distillery District, within the King-Parliament area that Jane Jacobs had a hand in rezoning.

The Kings Regeneration Initiative” targeted 400 acres of land along King Street, an east-west arterial with a streetcar. King-Spadina on the west side of downtown and King-Parliament on the east side were both declining CBD-adjacent industrial areas. Then-mayor Barbara Hall invited Jacobs to an advisory group on the regeneration project. “Paul Bedford, Toronto’s chief planner during Mayor Hall’s term, said that Jane kept encouraging him to take risks and to experiment,” writes Barry Wellman. The resulting code was a tremendous departure from how Toronto, and most other North American cities, regulated development:

Jacobs described the process in remarks given at Boston College’ law school:

Yet if the zoning were to be changed to permit dwellings, the developers would be blocked by rules applying to apartments, most especially parking requirements. Land coverage was high and parking couldn’t feasibly go underneath these sturdy but old buildings. Under the guidance of our very intelligent mayor at the time, these and almost all other regulatory controls were removed, except for fire and building safety codes. One rule was added: a ban against destruction of buildings, to prevent aesthetic and environmental waste. You would be amazed at how rapidly those dying districts have come back to life and blossomed. The principle at work here has been the addition of what the previous mixture lacked…

In the case of Toronto’s dying districts of downtown that were revitalized by radically overhauling the regulations, the mayor’s hardest job was goading and re-educating her own planning department, including the youngish man who then headed it.

The results have been breathtaking — and might surprise those for whom Jane is a hero for stopping bulldozers. Not only have the “Two Kings” not lost jobs, as many industrial lands taken out of production have, but the number of jobs has increased by 58%. Even more impressively, 46,000 dwelling units have been permitted in the Two Kings, many of them in very large new high-rises.

Of course, this approach would be much more difficult — if not impossible — to enact in America. It’s not that America over-regulates development per se, it’s that we regulate entirely the wrong things about development. As Jay Wickersham writes in the Boston College Environmental Affairs Law Review, the result is “an extraordinary situation. There is no other area in environmental law where the goals of the regulatory program are not just indifferent, but actively hostile, to the best thinking in the field.” From his introduction:

To paraphrase F. Scott Fitzgerald, Jacobs shows us that Euclidean zoning has been hard where it should be soft and soft where it should be hard. Zoning has been hard, or overly rigid, in dividing our cities and towns into uniform, low-density districts, each dedicated to a single primary use. And zoning has been soft, or overly permissive, in its failure to set design standards for streets, and for how buildings front upon those streets, that would reinforce the fundamental character of streets as public spaces…

Supreme Court rulings restrict municipalities to just two regulatory tools* that can shape development: Euclidean zoning (regulating density and land use) and historic designation (regulating appearance, but only meant for very limited circumstances). Euclidean zoning’s fixation on limiting density and land uses enforces conformity; even when it permits change, it’s only towards a distant, built-out end-state set forth in a comp plan. Jacobs writes:

[T]he greatest flaw in city zoning is that it permits monotony… Perhaps the next greatest flaw is that it ignores scale of use, where this is an important consideration, or confuses it with kind of use, and this leads, on the one hand, to visual (and sometimes functional) disintegration of streets, or on the other hand to indiscriminate attempts to sort out and segregate kinds of uses no matter what their size or empiric effect. Diversity itself is thus unnecessarily suppressed. (D&L, 237-238)…

Instead, Death and Life‘s chapter 13 argues for “zoning for diversity”:

The purpose of zoning for deliberate diversity should not be to freeze conditions and uses as they stand. That would be death. Rather, the point is to insure that changes or replacements, as they do occur, cannot be overwhelmingly of one kind. (D&L, 253, emphasis added)

Jacobs was not against regulation, but as an empiricist she held tremendous regard for the way cities had evolved as complex systems over the centuries — and fought the woefully simplistic (and completely ideological, perhaps even “faith-based”) Modern-era planning regulations and programs then in place. Alas, those regulations remain at the foundation of American planning today. Wickersham again:

According to Jacobs, “[a]ll zoning is suppressive,” an interference with the unfettered movements of the real estate market. But Jacobs is not attacking regulation, per se, or even the notion of government planning… she is attacking the functionalist presumptions shared by many city planners. In this view, a city is a functional, repetitive machine, rather than an ever-evolving organism… Her goal is to strike a middle course: to preserve and enhance diversity by avoiding large-scale, cataclysmic physical and social changes (which can be caused by rapid influxes of private investments, as well as by publicly sponsored urban renewal projects), without permanently freezing a community’s character.

Density-and-use zoning is the metaphorical hammer of urban land use: every potential problem ends up looking like a nail, and gets hammered to smithereens. It doesn’t matter if the problem has nothing to do with density or land use, and it doesn’t matter that density and land use are (as the Kings show) pretty darn incidental to the grand scheme of things. The only tool that we have is the wrong one, but we’re going to use it anyways. Wickersham notes that even the modest attempts to circumvent Euclidean zoning through discretionary approvals, or worse yet to somehow require diversity, are doomed to failure from Jacobs’ perspective:

Because these reforms are project-specific, and not comprehensive, the counter-productive, as-of-right requirements of Euclidean zoning have been sidestepped, not removed. To tempt developers into the project review process, regulatory systems will offer a density or height bonus to offset the increased time and costs that are involved. Such incentives can cause all parties to undervalue small-scale, incremental renovation and infill projects—the incremental reinvestments that Jacobs showed us were so important for the stability of an urban district. Thus, favoring large private investments can cause the same kinds of cataclysmic change that Jacobs decried in the public urban renewal projects of the 1950s.

Update: Shawn Micallef has a fantastic summary of Jane Jacobs’ Toronto legacy on Curbed today. The headline of this post spoofs Spacing Store’s #WWJJD t-shirt.

* Non-regulatory tools, like redevelopment, are also legal but are so difficult and fraught with such complexity that they’re unlikely to have a substantial impact on regional-scale land use challenges. Form-based codes are the most promising alternative to Euclidean zoning in the US, but in practice require such a radical overhaul of the planning-and-zoning process that they have yet to achieve wide adoption; Miami is the notable exception that rewrote its plan and zoning code all at once.

DC has “parking craters,” just not downtown. Here’s why.

Most American downtowns are surrounded by “parking craters,” as Streetsblog has termed them. Here in DC, downtown’s successful redevelopment has almost eliminated downtown parking craters, with one key exception. This success hasn’t completely eliminated parking craters from DC, though — it’s just moved them outside downtown.

parking crater at CityCenterDC

DC’s last privately-owned parking crater has a very unusual backstory. Gould Property owns the site free and clear, but only due to a land swap to get the Marriott Marquis built two blocks north. Gould had purchased part of the Marriott site back in the 1990s, when prices really were cheap enough to justify parking craters. The land basis and opportunity cost on this site is unusually low, especially since the former building on the site could not have remained.

Most surface parking lots are built as what zoning calls “an accessory use,” which means they’re an “accessory” to something else on the same lot. The parking lot at Sam’s Park & Shop in Cleveland Park or the Capitol’s parking lots, are “accessory” parking lots.

Parking craters, on the other hand, are usually not accessory parking directly tied to another land use; they’re paid parking lots whose owners are holding onto land that they speculate could be a future development opportunity. A parking lot requires minimal maintenance, but pays out some income in the interim. Most importantly, a parking lot is “shovel ready” — unlike a building with tenants in place, whose leases might or might not expire at the same time, a parking lot can be emptied and demolished on short notice when opportunities arise.

High rents and short buildings limit speculation

The opportunity that many “parking crater” developers are waiting for is the chance to build a big office tower. Offices pay higher rents to landlords than apartments (although in the best locations, retail or hotels can be even more valuable). However, the banks who make construction loans to developers rarely allow new office buildings to be built before a large, well-established company has signed a long-term “anchor tenant” lease for much of the new building’s space. If the building isn’t pre-leased, the result can be a bank’s worst nightmare: a “see-through tower” that cost millions of dollars to build, but which isn’t paying any rent.

Within downtown DC, robust demand and high rents mean that landowners face a very high opportunity cost if they leave downtown land or buildings empty for a long time. Instead of demolishing buildings years before construction starts, developers can make room for new buildings by carefully lining up departing and arriving tenants, as Carr Properties did when swapping out Fannie Mae for the Washington Post.

Less often, a developer will build new offices “on spec,” or without lease commitments in place. A spec developer usually bets on smaller companies signing leases once they see the building under construction. Downtown DC has a constant churn of smaller tenants (particularly law firms and associations) that collectively fill a lot of offices, but few are individually big enough to count as anchor tenants.

Because office buildings in DC are so short, they’re relatively small, and therefore the risk of not renting out the office space is not that high. In a city like Chicago, by contrast, few developers would bother building a 250,000 square foot, 12-story office building to rent out to smaller tenants. Instead, they could wait a few more years and build a 36-story building, lease 500,000 square feet to a large corporation, and still have 250,000 square feet of offices for smaller tenants.

While height limits certainly constrain the size of offices in DC, other cities with much less stringent height limits have also managed to eradicate most of their parking craters. Boston and Portland are similarly almost bereft of parking craters within their cores, not because of Congress but because other planning actions have maximized predictability and minimized speculation. In both cities, small blocks and zoning-imposed height limits of ~40 stories (!) encourage construction of smaller office buildings

Another factor common to these cities are policies also encourage non-car commutes — Boston even banned new non-accessory parking downtown — and rail transit that distributes commuters through downtown, rather than focusing access along a freeway or a vast commuter rail terminal. Metro’s three downtown tunnels, and DC’s largely freeway-free downtown, help to equalize access (and property values) across a wide swath of land. In retrospect, it’s impossible to identify which one factor had the greatest effect.

This customer is always right

There is one big anchor tenant in DC’s office market: the federal government. The government has some peculiar parameters around its office locations, which also help to explain where DC does have parking craters.

Private companies often don’t mind paying more rent for offices closer to the center of downtown, which puts them closer to clients, vendors, and amenities like restaurants, shops, or particular transit hubs. The government, on the other hand, has different priorities: it would rather save money on rent than be close-in. The General Services Administration, which handles the government’s office space, defines a “Central Employment Area” for each city, and considers every location within the CEA to be equal when it’s leasing offices. It also usually stipulates that it wants offices near Metro, but never specifies a particular line or station.

As rents in prime parts of downtown rose, the government began shifting leased offices from the most expensive parts of downtown to then-emerging areas. Large federal offices filled new office buildings in the “East End,” helping to rejuvenate the area around Gallery Place and eliminate many parking craters.

This one rule scattered “parking craters” all around DC, but they’re steadily disappearing

Over the years, DC noticed the success it found in broadening the federal government’s definition of the Central Employment Area, thereby spreading federal offices to new areas. It successfully lobbied GSA to widen the CEA further, encompassing not just downtown but also NoMa, much of the Anacostia riverfront, and the former St. Elizabeth’s campus. Because the latter areas have much cheaper land than downtown DC, and lots of land to build huge new office buildings, federal offices are now drifting away from the downtown core.

A developer with a small site downtown usually won’t bother to wait for a big federal lease: the government wants bigger spaces at cheaper rents. It’s easier to just rent to private-sector tenants. However, a developer with a large site within the CEA and next to Metro, but outside downtown, has a good chance of landing a big federal lease that could jump-start development on their land — exactly the formula that can result in a parking crater.

One recent deal on the market illustrates the point: the GSA recently sought proposals for a new Department of Labor headquarters. GSA wants the new headquarters to be within the District’s CEA, within 1/2 mile walking distance to a Metro station, and hold 850,000 to 1,400,000 square feet of office space.

The kicker is the timeline: GSA wants to own the site by April 2018, and prefers if DC has already granted zoning approval for offices on the site. It would be difficult for a developer to buy, clear, and rezone several acres of land meeting those requirements within the next two years, so chances are that the DOL headquarters will be built on a “parking crater” somewhere in DC. Somewhere outside downtown, but within the CEA, like:

Parking crater at Spooky Park, Yards

Parcel A at the Yards.

  • “Spooky Park,” or Parcel A at The Yards, formerly the National Geospatial-Intelligence Agency offices across from the Navy Yard Metro. It’s zoned for 1.8 million square feet of offices, and is probably the largest single parking crater in DC.
  • Behind the Big Chair in Anacostia are several parking lots that could house a million square feet of offices.
  • The Portals, next to the Mandarin Oriental Hotel at 14th and D streets SW, has two empty lots left. A residential tower will soon sprout on one, but the other is being reserved for another office building, across from to another building that was built for the FCC, but which will soon be vacant.
  • The two blocks just west of the Wendy’s at “Dave Thomas Circle,” in the northwest corner of NoMa, are owned by Douglas Development and Brookfield Asset Management. Brookfield’s site could house 965,000 square feet of development, and Douglas’ site could have a million square feet.

High-rise residential seems like it would be an obvious use for land like the Yards, which is outside downtown but atop a heavy-rail station. Yet even there, where one-bedroom apartments rent for $2,500 a month, it’s still more valuable to land-bank the site (as parking, a small green area, and a trapeze school) in the hopes of eventually landing federal offices.

Many federal leases are also signed for Metro-accessible buildings outside the District, which helps to explain why prominent parking craters exist outside of Metro stations like Eisenhower Avenue, New Carrollton, and White Flint. (For its part, Metro generally applauds locating offices at its stations outside downtown, since that better balances the rush-hour commuter flows.)

One reform could fix the problem

One esoteric reform that could help minimize the creation of future parking craters around DC is to fully fund the GSA. Doing so would permit it to more effectively shepherd the federal government’s ample existing inventory of buildings and land, and to coordinate its short-term space needs with the National Capital Planning Commission’s long-term plans.

Indeed, GSA shouldn’t need very many brand-new office buildings in the foreseeable future. Federal agencies are heeding its call to “reduce the footprint” and cut their space needs, even when headcount is increasing. Meanwhile, GSA controls plenty of land at St. Elizabeth’s West, Federal Triangle South (an area NCPC has extensively investigated as the future Southwest EcoDistrict), Suitland Federal Center, and other sites.

However, ongoing underfunding of GSA has left it trying to fund its needs by selling its assets, notably the real estate it now owns in now-valuable downtown DC. GSA does this through complicated land-swap transactions, like proposing to pay for DOL’s new headquarters by trading away DOL’s existing three-block headquarters building at Constitution and 3rd St. NW.

In theory, it should be cheaper and easier for GSA to just build new office buildings itself. In practice, though, they’ve been trying to do so for the Department of Homeland Security at St. Elizabeth’s West — a process that Congressional underfunding has turned into a fiasco.

Parking craters will slowly go away on their own

In the long run, new parking craters will probably rarely emerge in the DC area. Real estate markets have shifted in recent years: offices and parking are less valuable, and residential has become much more valuable. This has helped to fill many smaller parking craters, since developers have dropped plans for future offices and built apartments instead.

Goodnight, parking crater

A parking crater in NoMa that’s soon to be no more, thanks to apartment development.

Even when developers do have vacant sites awaiting development, the city’s growing residential population means that there are other revenue-generating options besides parking. “Previtalizing” a site can involve bringing festivals, markets, or temporary retail to a vacant lot, like The Fairgrounds, NoMa Junction @ Storey Park, and the nearby Wunder Garten. This is especially useful if the developer wants to eventually make the site into a retail destination.

Broader trends in the office market will also diminish the demand for parking craters, by reducing the premium that big offices command over other property types. Demand for offices in general is sliding. Some large organizations are moving away from having consolidated headquarters, and are shifting towards more but smaller workplaces with denser and more flexible work arrangements.

Unlike the boom years of office construction, there’s now plenty of existing office space to go around. Since 1980, 295 million square feet of office buildings were built within metro DC, enough to move every single office in metro Boston and Philadelphia here. While some excess office space can be redeveloped into other uses, other old office buildings — and their accessory parking lots — could be renovated into the offices of the future.

A version of this appeared in Greater Greater Washington.

How Chicago’s zoning excludes small apartments from the neighborhoods

chizone

Chicago’s zoning code has a built-in bias against smaller apartments – except in a few high-density zoning districts, which cover a vanishingly tiny slice of the city.

The zoning ordinance regulates building size and density in three ways: through floor area ratio (FAR), “minimum lot area per unit” (MLA, a backwards way of saying dwelling units per acre), and through various setback regulations. Yet these don’t follow a linear relationship at all; instead, the interaction between FAR maximums and MLA minimums encourages larger buildings with fewer apartments in lower-density districts, and more apartments per building in higher-density districts.

What this graph shows is: If I have a standard city lot in a given zoning district, and I build the biggest possible building with the most possible units, how large would those units be? The answer varies tremendously across the city, from a low of ~600 square feet in high-density districts like RM-6 and DR-7, to “impossible” in the lowest-density districts (note 1).

Most of the city’s neighborhoods, from the Bungalow Belt through the Zone of Two-Flats (mostly zoned RS-3) and into the Zone of Three-Flats (mostly RT-4), is zoned for the lowest-density (left-most) third of this graph. From RM-4.5 on down, the average apartment that can be built at the maximum density must be 1,250 square feet or larger (the size of a large two-bedroom apartment). Sure, you could build studio and one-bedroom apartments, but then you’d have to build huge three- and four-bedroom apartments, too.

Only for RM-5 and above, high-density zoning classifications pretty much only found in a narrow band along the lakefront, do the average apartment sizes permitted begin to dip into one-bedroom territory.

Someone who wishes to build new smaller apartments, like one-bedrooms or studios, in order to accommodate shrinking households can only do so in neighborhoods like Lakeview or Logan Square by either (a) under-building the FAR, at a considerable opportunity cost, or (b) getting rezoned to a denser category. Anyone who goes the latter route might as well build a lot bigger and higher, too.

Methodology notes:

  1. I assumed a standard 25′ x 125′ (3,125 sq. ft.) city lot. In RS-1 and RS-2, you cannot build on a lot that small, hence those values are excluded.
  2. The MLA chosen is the number specified for efficiency apartments, a distinction made in higher-density districts which would skew their figures down.
  3. “BCD” here refers to mixed-use zones that can be designated Business, Commercial, or Downtown depending on the use. What’s important for these purposes is the numeral in the zone name, which defines the density allowed. These zones all permit (and indeed, require) substantial amounts of commercial floor space, which counts against FAR but not MLA; for these zones, I assumed an apartment building with 0.5 FAR of commercial space, and the rest residential. This skews the figures upwards for downtown districts, where one would reasonably expect more than 0.5 FAR of commercial space.

Friday photos: Tokyo’s small scale, and how I learned to love Modernism

Cube

No less an authority on great streets and great cities than Holly Whyte, guiding spirit of the Project for Public Spaces, wrote in City: “But the most fascinating of Tokyo’s streets are its ordinary ones. Mile after mile, they are consistently more interesting than ours. In the Shinjuku district there are more streets to savor than in most U.S. cities put together, and for sheer sensory impact there is nothing to match its back alleys…” (And Whyte’s comparison set was NYC in the 1980s, which hardly lacked for “interesting” streets that packed in “sensory impact.”)

Narrow flatiron

Tokyo disabused me of any notion that classical architecture was integral to creating walkable, human-scaled places, or that newness per se is the problem. Japanese cities show that fascinating places can exist with mediocre 20th-century architecture. And because they’ve shown that it’s possible to build fantastic, prosperous, massive, and efficient cities in the 20th century, they’re much more powerful examples for city-builders in our time than European antiquities.

Tokyo was a megacity in the pre-industrial era: it boasted a million residents back when New York was a mere town of 50,000 cowering on the edge of a vast wilderness. But Tokyo was leveled by an earthquake in 1923, then firebombed in World War II. Unlike other societies, which would use those catastrophes as excuses to re-plat and re-organize, the Japanese rebuilt their cities with largely the same warren of narrow streets and the same itty-bitty lots, but modernized the architecture with the very latest in cement-block schlock. It’s as if Rome were rebuilt by Angelenos in the 1970s. And it’s fantastic.

Here’s just a random intersection I walked through near Fudomae station, southwest of the city center and outside the Yamanote Line “beltway,” where a friend of mine happened to let me stay in her inexpensive flat. Yet it’s quiet, livable, affordable, fascinating to walk around, and completely illegal in America due to its narrow streets, mixed uses, Byzantine street pattern, high lot coverage, and high density — in other words, its small scale.

Many smaller buildings will create a much more lively streetscape, and more opportunities for cool urban spaces, than fewer but larger buildings. The critical deficiency in American cities, by comparison, is not limited to fixing a single profession, like architecture or even engineering. Rather, our entire system is tilted towards gigantism: our suburban-era parcel sizes, our High Modern-era zoning (which almost require large buildings), our mortgages, our street design manuals, our nonsensically upside-down parking requirements.

inside the void

If the stupendous urbanism yields such terrific results in the hands of ordinary architects, it becomes sublime in the hands of masters like Tadao Ando.

DC’s branch campuses

Cornell University Wolpe Center
Cornell University’s Wolpe Center. Photo: Josh/NCinDC, via Flickr

Quite a few universities have branch campuses in Washington, DC, but it seems like there isn’t a definitive list online. This seems odd: not only is higher education is one of the metro area’s largest non-federal industries, but in international economic development circles branch campuses seem just as highly sought after as corporate headquarters. After all, not everyone has the chance to import smart, motivated, mobile, impressionable kids from elsewhere during the most memorable years of their lives. Even if they don’t stay afterwards, there’s at least a chance to leave a good impression on future global leaders.

That’s why countries in Asia and the Middle East are spending lavishly to attract branch campuses of prestigious universities: Abu Dhabi and Shanghai both custom-built free campuses and gave cash grants to get NYU started there. NYC has also embarked upon the same effort through its three Applied Science Campuses.

Contrast that approach to here: NYU paid for its DC building through their own fundraising.* DC’s economic development officials have been dragging their feet for years on opening a college campus at St. Elizabeth’s, instead spending their time moving pro sports facilities there from elsewhere in the District. Not only is there not a concerted effort to attract or retain facilities, or the go-getter students who attend these campuses – there’s not even a list of these facilities. This seems like a missed opportunity.

So, crowdsourced by Jason Terry and his friends, universities that have a full-time physical presence within the counties traversed by the Beltway and that aren’t the locally headquartered members of CUWMA:

Consortiums that serve multiple universities from elsewhere

Full, undergraduate degree-granting, residential branch campus

  • None that I know of

Has classrooms, offices, and dorms

Classrooms and offices, subleased dormitories

Classrooms and offices, no dorms (includes postgraduate-only and commuter schools)

Office with supportive programming for interns

Additional crowdsourced entries welcome via Twitter. After all, both USCs and both CMUs are on this list, so I’m sure that numerous other universities with completely different initials are here.

* Not to be misconstrued as a statement in favor of economic development incentives, which are largely useless whether they’re given to for-profit or non-profit corporations.

Friday photo: Shared space, residential parking-lot edition

Sofia Lofts, San Diego, CA

The ULI Case Study that I wrote about Sofia Lofts, a 17-apartment development in a neighborhood in eastern San Diego, was recently posted. I was particularly intrigued by how the developer/architects used shared space to maximize usable open space while meeting the letter of the law with regard to parking requirements:

For the balance of the 21 [parking] spaces planned, six more garages are accessed via the courtyard. The plans show seven more spaces within the courtyard itself, immediately in front of the garages — but to keep the space inviting, the spaces in the courtyard are off limits for tenant parking. It’s a small net loss for NDD, which discounts the rent for some tenants who forgo on-site parking, but results in a common area that looks like an expansive garden, rather than a parking lot.

The materials used within the courtyard were also chosen to make the garden feel like a place where cars just happen to be allowed. Part of the driveway near the alley is paved with concrete, but the center of the property is covered in pea-sized gravel. The texture of the gravel slows down cars to walking speed, adds visual and auditory interest, and permits rainwater to filter through. Its inspiration comes from a house in Italy that the Nakhshab family lived in; as Soheil recalls, “The courtyard we would always play in had that type of gravel, and we are seeing the tenants’ children doing the same thing.”

Friday photo: Why did Chicago courtyards disappear?

anti-courtyard

Here’s my old block in Wicker Park (and namesake for this blog): the 2100 block of West North Avenue. It’s also an interesting illustration of how Chicago’s 1957 zoning ordinance made courtyard apartment buildings illegal — even though this eminently livable, passively vented building type defines high-density Chicago neighborhoods like Rogers Park and South Shore.

What did courtyards in were backyards. The 1957 ordinance (mildly updated since) requires a 30-50′ rear yard for residential units in all but the densest of downtown zones; functionally, this is where the garage went. This mandate shifted open space to the rear of the lot, and the entire point of a courtyard building is that the open space is consolidated at the center of the lot.

The 1911 C-shaped courtyard shown at lower right has minimal setbacks along the sides and rear, which results in a generous central courtyard. The 1901 3-shaped courtyard at upper left was required (by a subdivision-specific covenant and by a ‘L’ line easement) to retain front and rear setbacks, and as a result its courtyard shape is so constrained as to be almost illegible. The new buildings at lower left were required to have a rear setback (the ones at far left got a slight variance) and thus consolidate their open space at the front and rear of their lots, to the great detriment of the tiny sliver of open space that the architect sought to insert in the middle.

Other contributing factors to the demise of courtyard apartments include:

  • Parking requirements, which are difficult but not impossible to accommodate in basements. The large courtyard shown here is raised, and could have been built with parking underneath.
  • The fashion for high-rises and for deeper floorplans in general, fed by postwar technological changes — cheaper steel and concrete, improved lighting, air conditioning, elevators, etc.
  • By the time 1929 came around, appropriate sites for courtyard buildings, which required contiguous lots zoned for medium density (at the time, Chicago had three density districts, although then as now the low-density district was pretty capacious). In 1945, when construction resumed, the city’s builders had shifted their attention away from the relatively built-up lakefront.
  • Market distortions that encouraged detached single-family houses instead of apartments.

 

Ivy City: who goes there?

Hecht Warehouse, Ivy City, DC

The view northeast from the Hecht Warehouse’s parking garage, towards “NewCityDC.” 

Last month, Douglas Development filed plans for NewCityDC, which will bring more than half a million more square feet of retail space to the New York Avenue NE corridor, adjacent to the substantial residential and retail investments it’s gradually opening around the old Hecht Company warehouse in Ivy City.

NY Ave, aka US 50, is the only full-on traffic sewer in DC, with six through lanes, a speed limit up to 50 MPH. For a three-mile stretch between the Maryland line and Florida Avenue (the boundary of the L’Enfant city), it’s paralleled on the north by a trench holding the Northeast Corridor railroad and cut off to the south by a variety of institutions (the arboretum, Gallaudet University, cemeteries), and thus has only a handful of intersections with the street grid. That proximity to the railroad brought both low-density industrial buildings and a Skid Row feeling to the blocks surrounding it. The street hardly has sidewalks, definitely does not have bike lanes, and doesn’t even have a city bus route.

Yet despite all that, Douglas — who has made a fortune turning around the East End of downtown DC — thinks there are customers for 300,000 square feet (a regional mall’s worth, net of the anchor stores) of specialty retail in this isolated location. And they’re sinking lots of money into the area; this is some very heavy-duty and expensive work to do for single-level retail:

After Hecht, Douglas' next retail building

Douglas’ marketing would have retailers think that there are lots of customers right at their doorstep, thanks to dubious maps like this “trade area analysis”:

hecht2

The map gooses up the demographics by drawing a “15-minute drive time” radius that brings everything from College Park to Georgetown to Pentagon City into the mix — even though

  1. Georgetown is almost never a 15-minute drive to Ivy City;
  2. More than half of households in the Census tracts surrounding Ivy City do not own cars, along with about 40% of central-city households;
  3. Most residents west of this site may be only scarcely aware that New York Avenue, much less Ivy City, exists.

“Average household income” is basically meaningless, especially in prosperous (and expensive) metro DC, since all three of those figures are substantially below the city average of $106,000.

A site-and-vicinity map is even more misleading:

hecht1

This map highlights thousands of apartments that are being delivered around NoMA. Never mind that many of those new units don’t have parking spaces (since most of the city’s new households don’t have cars), which will make it nearly impossible for their residents to get to Ivy City.

Sure, Hecht is a 4-minute drive from NoMA, and a mere 3/4 mile for any birds who are roosting atop its new high-rises. But for most anyone who actually lives in NoMA, it’s nearly half an hour away (by bus and foot) — during which time that resident could have just gone to Metro Center (with 15 minutes to spare) or Pentagon City or Silver Spring. In short:

The real market for Hecht, NewCityDC, and Fort Lincoln’s retail is exactly what you’ll see in the parking lot at the Costco at the latter: lots of Maryland license plates. All of these are set up for easy right-in/right-out access for drivers headed outbound on US-50, who don’t have many other shopping choices until Bowie or Annapolis. That market is certainly underserved — but it’s much smaller than the one that Douglas’ maps promise. Economic development officials in Prince George’s County should take note.

Help, a mall ate my walk shed

Mall entrance

The entrance to Hong Kong’s airport express train is somewhere within this mall. The cross-border bus terminal entrance is somewhere else entirely, and the actual “public” spaces are completely dispiriting.

When I was a small child, John Portman-style complexes were architecture’s futuristic vision: Someday, we’d all live in hermetically sealed downtown compounds of office and hotel skyscrapers set atop multi-story podiums.

My family stayed in several of these hotels on trips; sometimes, my mom would tell me about how she, as a child in Hong Kong, had dreamed of a city of skyscrapers and layers of indoor shops, all suspended above grade so that buses, boats, and cars could fill the ground plane.

That hyper-dense Modernist vision eventually came to pass in much of central Hong Kong, fed by not only its unique geography but also by its unique private-provision model for both rail and property. There’s danger, it turns out, in putting developers in charge of your rail stations; the “gift horse” of free infrastructure comes with strings attached.

When private firms are put in charge of designing pedestrian circulation networks, they will place their values — primarily shuffling eyeballs past storefronts — over the public’s need for legible, direct links from A to B. The tremendously high value of rail access behooves developers to elbow their way closer to the station; the location imperative isn’t to be near transit, it is to be at transit; to make it not just easy, but necessary to traverse their property. And, once the development has cornered the transit station, it will seek to entrap the resulting pedestrian flows within a spiderweb of passages. As Chris DeWolf writes:

Last month, a survey of 657 Tsim Sha Tsui pedestrians conducted by urban design watchdog Designing Hong Kong revealed that 77 percent prefer using street-level crossings over footbridges and subways… “The problem is that bridges and tunnels force you into particular routes that limit your ability to take the shortest path,” says Designing Hong Kong convenor Paul Zimmerman. “People also pick attractive routes. That’s a very qualitative statement, but part of what makes a route attractive is being able to see other people, to window shop, to have an experience. With subways and footbridges that becomes quite limited.”

An early version of this phenomenon can be seen in Montreal’s underground city, the most valuable retail frontage is as close to the train platforms as possible. Thus (almost as in casinos) the developers twist and turn the corridors to herd people past the shops. Future iterations of the phenomenon will soon be unveiled at the World Trade Center mall and at Hudson Yards.

Edit 7 Jul 18: author Marion Girodo has a book of “urban mangroves,” the multilevel urbanism that crops up around metro stations in Montreal, Paris, and Singapore.

Edit 17 Jan 17: Henry Grabar in Gothamist writes of “the oculus”: “If Grand Central is a train station with some shops, the Oculus is a shopping center with some trains.”

New PATH - WTC underground passage

Want to get to the PATH, or cross West Street to get to the Hudson? You’ll have to walk down this hall, and oh by the way there are plenty of shopping opportunities.  

A similar landscape may be emerging at Tysons Corner, where the in-process retrofitted suburbia — now and forevermore ridden with alienating highways — shows little sign of ever becoming a Greenwich Village sort of urbanism with small blocks of public streets lined with small-but-tall buildings. Instead, the spatial complexity that’s emerging is of a very different, much more Portman-esque sort. Philip Kennicott writes in the Post:

The decision to elevate the stations — a far less expensive approach than burying them — may well presage this sleek new world of elevated plazas and public areas, disconnected from the ground. A new office building across from the Tysons Corner station is built atop a parking garage, so that at ground level one faces a seemingly impenetrable plinth. Already, a web of pedestrian bridges — some built by Metro, others by private developers — is emerging, keeping us safely above the world of machines and hydrocarbons and asphalt…

One wonders if you will emerge from these stations with [a] sense of pleasant surprise and rootedness in the urban landscape… Likely not. Rather, you will emerge, slightly disoriented by the ever sameness of the commercial and physical space around you, wondering for a moment if you have arrived at the right station, before your basic sense of purpose — to get home, to find a restaurant, to locate a shop — kicks in, and you begin to move by habit and instinct through a pleasantly unobtrusive world of concrete and glass that could be anywhere.

Tysons Corner Center expansion

Tysons Corner Center’s “Metro Plaza” under construction. At left, the bridge to the station, at right, the bridge to the mall.

At Tysons Corner Center, the megamall at the heart of Tysons, building a bridge keeps the distance from station to mall is 300 feet — a 1.6 minute walk. The bridge siphons customers directly into the mall, creating tremendous value in three dimensions: “You’ve got a first floor on the first floor and a first floor on the second floor, so you’ve solved the verticality problem” [of pulling mall foot traffic from the entry to different levels], Timothy Steffan, an executive for mall owner Macerich, told the Post’s Jonathan O’Connell.

This Disney-esque strategy of spiriting people directly into an immersive environment has ample precedent: in fact, Roppongi Hills, a fantastically successful redevelopment in the heart of Tokyo, also uses an elevated plaza to deliver customers from the subway onto the development’s podium. The experience for customers who drive in is akin to what Rick Caruso’s malls or skyway’d downtowns provide: parking garages deadening the street environment all around, but a fantastic public space within. Sure, bus and bike customers have to deal with an ugly exterior, but the privileged modes (driving and heavy rail) get the red carpet.

Amazingly not Portman: Plaza of the Americas

Skyway urbanism in Dallas

My first instinct is to warn “creative” policymakers to be careful what you wish for: these projects result in such high cost and complexity that the only financially worthwhile result is a giant mall. They’re so huge, boring, and bland because of private value capture. Finding room in a private developer’s pro forma to build expensive underground rail infrastructure requires selling stupendous quantities of expensive corporate real estate, which will never be cool and lively. It also requires generous, greenfield-esque parcel sizes. In the worst case scenario, the project fails, and there’s nothing worse than a white elephant in the middle of the room.

Yet as dispiriting as these initial examples are, there’s a glimmer of hope that they’ll eventually be okay. With enough time, enough density, and enough owners, even these bland malls could evolve into something interesting. Hong Kong’s experience shows how the weird linkages that result from generations of ad-hoc decisions and relentless foot-traffic flows have created a hyper-dense end result that can be beautifully complex in a postmodern, emergent-urbanism way. This isn’t immediately apparent from the workaday commercial architecture, but can be mesmerizing when expressed in diagram form — as the recent book Cities Without Ground shows (high-res image slideshow). A two-dimensional plan, or even a figure-ground diagram, is useless when expressing vertical spaces.

Hong Kong architect Peter Cookson Smith described this structure in more essentialist terms in The Urban Design of Impermanence (pg. 84; excerpt):

A cityscape of streets, internalized routes and multi-level links, even without clear articulation, is open to casual exploration, and there is little need for city form to be overly organized or pronounced in order to be legible… This underlines an essential difference between the formal framework of Western public spaces and the more diffused and informal realm of social space associated with the Hong Kong street, where the relationship between public and private spaces is less tangible, and the routes between them work just as effectively in three dimensions as in two.

Perhaps China’s homogeneity and sheer density (bear in mind this is about 1-2 orders of magnitude higher than urban America’s) might increase social trust and thus break down the hierarchy of spaces — people there feel more comfortable wandering down dark alleys. Yet perhaps we could shortcut to that future: emerging spatial technologies, like smartphone-based mapping, are quickly obviating highly legible spatial hierarchies. Customers can now just as easily find a shop hidden in the back corner of a buildings as one that shouts its presence with highway-sized signs.

Or maybe not. Toronto’s PATH system is now 40-odd years old, serves more than 100,000 pedestrians a day (so many that its closure would send downtown into gridlock), and has apparently outcompeted street-level retail spaces. It’s also a navigational nightmare of corporate sameness, according to Spacing’s Kieran Delamont:

The PATH is a mall, first and foremost. Beyond access to over half a dozen food courts, at least two massage parlours, and more sushi restaurants than I cared to count, I put it to you that, with its existing wayfinding system still in place, the PATH offers nothing especially preferable over supra-terranean navigation… It is a seemingly endless maze of mall corridors, hallways, and atriums. Unless you are a person for whom the distinctions between Jamba Juice and Jugo Juice are particularly meaningful, everything in this place feels exactly the same; with each new tunnel you encounter in here, the space expands physically while being visually constricted. Because this space lacks the distinctive landmarks that you often find above ground, there is very little to distinguish it from every other mall you’ve been in. The more of it you explore, the less it feels like you could ever remember any of it…

The balance between mall and transit network is slanted heavily towards the PATH’s commercial interests; the dominant incentive of the landowners is to keep you in their slice of the PATH, not to move you through efficiently.

What we need are architects and developers who understand that, and aren’t afraid to create more interesting, if less-legible and less predictable, places — perhaps even fractal-like, medieval-esque street plans in three dimensions.

Sai Yeung Choi St

America’s lower urban densities mean that our over-commercialized transit districts may never quite achieve this level of spatial complexity, but by golly, why not try? 

[Previous transit-station walk shed coverage: walk sheds & excessively grand rail stations, walk sheds & water transit]

Friday photo: Incrementalist lessons from Seaside

Seaside lessons: plan for evolution, not revolution

The ULI office is moving in a few months, so a lot of old files are being tossed out. One that I saw poking out of a garbage can was a 1986 Project Reference File written about Seaside, Florida. The “lessons learned” section worth excerpting, if only because it doesn’t talk about the PoMo architecture, or even the planning — instead, it’s all about the incremental nature of the development.

Most resort development today is characterized by a highly refined design concept coupled with central ownership and tight control over design and building decisions. In contrast, Seaside’s approach is to encourage authentic diversity by delegating to others as many design decisions as possible, within the dictates of a sophisticated urban design plan….

A significant factor in Seaside’s development is that, by owning the land outright, Davis was able to 1) invest in a considerable amount of upfront planning, and 2) proceed cautiously with the development. By going slowly, he says, a developer can reduce risk and can correct small mistakes. At Seaside, the master plan was not recorded until after the developer had had several years of experience with building and marketing this unique product. This allowed for minor refinements in the development strategy, plan, and timing, while prices rose accordingly….

Instead of assuming that large upfront investments in amenities would produce marketing payoffs, the developer moved slowly and carefully, guided by the master plan.

In short, plan far ahead, regulate what matters, and phase to allow adaptation and evolution. (Evolution, not revolution.) Alas, the world still has a lot to learn from Seaside.

This point is echoed by Peter Cookson Smith in a book that I’m reading about Hong Kong, a seemingly very different place:

Overly engineered environments leave little flexibility to make incremental adjustments in response to the evolving economic circumstances that normally represent the lifeblood of towns and cities.

Cartographic pet peeve: Who skipped 94101?

94101

For most US cities, I can quickly get a map zoomed right to the central city by going to most any mapping application and typing in a five-digit ZIP code that ends in “01.” That’s because the first three digits of any ZIP code are actually a meta-ZIP, best known by the Census Bureau’s term “ZCTA3.”

Back when ZIPs were assigned, the large post offices that served entire cities or broad rural areas were assigned these three-digit codes, and subareas within them were numbered off in roughly concentric order from the sorting center. ZIP code XYZ01, then, was usually at the then-principal post office, either right downtown or by the railyards where mail was usually offloaded then.

The Post Office assigned these codes in an ascending order following a meandering east-west path across the Lower 48, beginning in New England and ending in the Northwest. Thus, it’s not terribly difficult to remember a city’s ZCTA3 — since they’re systematic, it’s easier than memorizing cities’ telephone area codes (a taxonomy that, while having its own charming history, is increasingly irrelevant).

  • Boston 021
  • New York City 100
  • Pittsburgh 152
  • Philadelphia 191
  • Washington 200
  • Raleigh 276
  • Durham 277
  • Atlanta 303
  • Miami 331
  • Minneapolis 554
  • Chicago 606
  • Denver 802
  • Los Angeles 900
  • Portland 972
  • Seattle 981

BUT there’s one (one!) exception that thwarts my little mnemonic: San Francisco. Its ZCTA3 is 941, but there is no 94101; the lowest-numbered ZIP in town, centered on the main post office by the Civic Center, is 94102. Who stole 94101?!

Postscript via Eric Fischer, proud owner of a pre-ZIP postal map:

//platform.twitter.com/widgets.js