The Definition of Insanity

From Landscape Urbanism and its Discontents, courtesy Envision Baltimore:

Rather than having an urban fabric based on spatial definition by buildings, landscape would be the 'structuring medium.' The 'look and shape of the city' was to be a matter of 'open space within which buildings are set'.

It didn't work the first time; it's not going to work again.

Bouncing around my mind

Pertinent to this post:

Since the PPA has no inkling of what the actual supply of any of their permit zones actually is, they routinely issue far more permits than spaces that can possibly handle those permits. Anecdotal evidence (courtesy a certain foulmouthed blogger) suggests, for example, that PPA Zone 4 currently has ~2x the permits its physical constraints can possibly handle.

Their ignorance--abetted by the fact that one has ever found a way to assess onstreet supply (until now)--is the only way they can get away with this. The implicit guarantee of a permit is that you are entitled to a place to park in the area where it's issued; if you have more outstanding residential permits than residential spaces, you've reneged on that implicit guarantee. Hint hint.

That Eureka Moment

One of my long-term projects has been developing a mathematical model for scheduling in Jarrett Walker-type transit systems. Today, I made a breakthrough--

I had previously realized that a model of such a system would yield two intersecting planes in my defined 3-space, and that desired overall frequencies could be model by stacking instances of these planes atop each other. But I'd been at a loss as to how to model the routes within the planes (other than the fact that they were vectors).

What I realized today, however, is that the planes are actually a set of null space vectors; this means that a prescribed eigenvalue in the null space will seed an eigenvector that corresponds to that value. What that means is that the entire system can be modeled with nothing more than the equations of the two intersecting planes, and the set of eigenvalues that yield the eigenvectors corresponding to the known bus routes.

Fortunately, we remain in vector space so far, but it appears the apparatus I'm constructing will wind up generalizing into a differential equation, to accommodate "gridlike" systems which attempt to install a mass transit grid even over non-grid street networks.

Dominion

I've been playing a lot of Skyrim lately, and ruminating on Elder Scrolls lore, so if you pardon this nerding-out, I'd like to offer my opinions on what will happen in the next game.

1. It will be called Dominion, and be based in the heartland of the Aldmeri Dominion (Summerset Isle).

2. No matter how you play Skyrim, things turn out for the Dominion's benefit. Canonized lore will have it that the Dragonborn completed the Assassin's Guild questline (ends with the assassination of Emperor Titus Mede II), whose end results in the fragmentation and collapse of the Tamrielic Empire. The whole civil war questline ends up in the dustbin of history; in a certain sense, it does not matter whether or not the Empire or Ulfric wins, as the whole of Tamriel splits into a bevy of statelets which the Dominion then vacuums up.

3. Thalmor religious persecution has developed into full-blown genocide. And Mundus is worse off for it; in fact, it's buckling.

4. It will be the Hero's job to assume Talos' divinity and stabilize Nirn and the firmament. The second-most-important task is destabilization of the Dominion, by now roundly considered evil. 

Of course, this doesn't undermine the ability to make moral choices--an Elder Scrolls staple--any more than the Empire's overarching existence's does Daggerfall. It's likely that Dominion's principal moral hazard lies in how the Hero goes about unifying the resistance, and be informed by various groups having different degrees of moral rectitude and likelihood of carrying out their resistance.

5. Freedom of movement--to a significant degree--will be curtailed, especially in the early game. (Full freedom of movement occurs in the late game.) No matter how you go about it, the Hero will quickly be branded an enemy of the Thalmor state, and so will need to get quite good at avoiding Justiciars, at least until you're powerful enough to take them head-on. Much of the early game will thus take on elements of getting from Place A to Place B without attracting authorities' interest, and creating and inculcating friendly bases throughout Summerset. The Dragonborn you ain't.

Coll. Incidentally, the Shouts mechanic will be implemented here, but in a vastly different way. It's likely that, in the lore, under the Last Dragonborn's influence, Shouting was developed from a "special skill" to a school of magic built on the Greybeards' teachings.

Coll 2. It's likely that Summerset's traditional joinable factions will be closed to non-Altmer (particularly if they keep this Nazi thing going). I'd bet that the "real" joinable factions will involve some combination of hidden factions and under-the-table dealings with traditional ones closely associated with the Thalmor, and relative freedom for the ones that aren't. So, for example, the Assassins' and Thieves' Guilds will be significantly easier to join than the Fighters' Guild or local Mages' Guild variant (unless that turns out to be the Psijic Order, which has zero known relations with the Thalmor overlords).

I don't really know what's on Bethesda's storyboards, but these ideas seem like they'd make the best possible Elder Scrolls game to me. Certainly, gameplay mechanics in Dominion will have a very different, unique, and interesting twist to them, particularly since what we know of the Thalmor implies that the next game will have to encourage stealth-based gameplay to a far greater extent than previous installments ever did.

A Sample Program

Some notes on what a sample elhi math program would look like...

Grades 1-3: Introducing Relations

• What are relations?
• What are operators?
• What are numbers?*
• Numbers in real life
• Telling Time, Counting Money
• Relations Between Numbers
• Operators Symbolize Relations
• Simple Arithmetical Operations
• Add and Subtract any Pair of Numbers
• Simple Multiplication and Division (to 10)

Grades 3-6: Greek Math, Intermediate Arithmetic

• Geometry and Ratios
• Plato's Meno
• Archimedes and the Method of Exhaustion
• Multiplying and Dividing Larger Numbers
• Multiplying and Dividing Fractions
• Decimals, Zeros, Negative Numbers
• How to Use an Abacus
• Pythagorean Theorem
• etc. 

Grade 7: Geometry. Elements. Euclid.
Grade 8: Algebra. The Compendious Book. al-Khwarizmi.
Grade 9: Analytic Geometry. La géométrie, Descartes.
Grade 10a: Trigonometry.
Grade 10b: Calculus.
Grade 11: Discrete Math.
Grade 12: Elective (if so chosen)

• Combinatorics
• Boolean algebras (logic, set theory, etc.)
• Linear algebra
• Multivariable calculus
• Graph theory
• etc.

___________
*Note order. The relations are more important than the numbers. For example, a first-grade project could be to make a family tree (create a statement of relations), and then a way to link any two members of the tree together without long text block or, well, the tree (that is, create operations). We then introduce the standard operators (perhaps in the context of the family tree project). Numbers also have to be introduced early, but teaching relations and operations needs to be separate than numbers for a while because (1) we want students to have a major aha! moment when we merge the two together, and (2) by teaching the relational/operational aspects separately from numbers, we can focus students' attention on the fact that mathematics is really relational in nature.

Pedagogy

I am very much of the opinion that the way we teach math here is fundamentally broken. Here are a few places where I think it can be fixed:

1. Use Greek geometry to help teach children how to experiment in math. Consider texts like Plato's Meno as an inspiration. Projects at the elementary school level could replicate Socrates' solution for finding a square with half the area as the original square (and its--easier--natural extension, a square with double the area). Projects can grow more complex* until e.g. a fifth or sixth grader can prove the Pythagorean Theorem. Using Greek math also helps introduce a deeper understanding of ratios (fractions), something teachers have confided in me American students have gotten progressively worse at.

2. Emphasize the relation, not the input and output. This is for arithmetic. Right now our arithmetic is rote, tabular, a strict interpretation of inputs into outputs. But the core of mathematical analysis is relational. Operators are nothing more than a statement of relation: We need to find a way to teach as such, instead of getting bogged down in numbers to an inordinate degree.

3. Reclaim the ancients' works. Of particular note, there is no reason whatsoever why Euclid's Elements, the standard textbook on geometry for more than two millennia, is not used in high school classrooms today. Likewise, analytic geometry should be based on Descartes' La géométrie (with van Schooten commentary and an augment of Fermat etc. papers and more illustrations); algebra, al-Khwarizmi's Compendious Book. Euler's textbooks can also be considered, but his treatment of trigonometry with complex numbers is perhaps a bit too Baroque for a modern audience. The flow algebra -> analytic geometry -> calculus also needs to be emphasized, as it is actually quite a bit more natural than most people realize. There is a case to be made that the middle school curriculum can run Euclid -> al-Khwarizmi -> Descartes, with standard calculus instruction occurring no later than the sophomore year of high school (and is a freshman subject for most students).

4. Rebuild the calculus pedagogy. One of the reason why calculus is perceived to be "hard" is because it begins with limits, but limits are discussed with insufficient tools for their analysis; by the time differentiation and integration are gotten around to, too many minds have been closed to those operations' ease and intuitive arising. Instead the pedagogy should begin with the easier operations, and once a good handle has been gotten on differentiation and integration proceed to the discussion of limits, why they matter, and how calculus operators both depend on limits and make calculating limits bearable. In addition, calculus needs to be used to cement the relational nature of math, as the elementary system is just an augment of unary function operators on analytic geometry (the only unary operation most students are exposed to prior is arithmetical negation--the difference between 2 and -2). Making use of Newton's, Leibniz's, the Bernoulli's' etc. papers is also useful, although at this point the textbooks begin to massively improve in quality.

5. Make discrete mandatory, and prerequisite to linear algebra. There are several reasons for this: (1) discrete math effectively functions as Intro to Higher Math; (2) a lot of linear algebra is essentially set theory and function theory in vector spaces; (3) with programming-language knowledge now needed even in the fine arts (according to UArts students) understanding its underpinnings is more important than ever; and (4) being able to perform in other types of discrete math, like combinatorics, is becoming increasingly necessary in our modern-day world. Understanding iterated operations and floor and ceiling functions, for example, should be SAT-level expectations.

6. Embrace models. Current mathematical pedagogy ill prepares us to actually put math to real-world use--the construction of models. This is universally covered in the various applied disciplines, but being able to construct simple models such as speed v. time (yielding velocity, acceleration, and jerk) or simple binary-string-based programs should be a skill shared by all high-school graduates. Grasping the principles needed in such modeling allows students to develop significantly more complex models to fit and extrapolate available data.

7. Embrace tradition. My final major critique is that most (not all) math teachers I've had have had little interest in exploring the tradition of math. While there are always students that complain about classroom trivia, it does a serve an important pedagogical purpose: It links learners to the discipline's tradition, helps open avenues for exploration, is often entertaining and a break from the hard work of the day, and generally makes the discipline more human, and therefore more interesting. This is true throughout most disciplines--it is hard to imagine a physics class, for example, without mention of Galileo or Newton or Einstein, or a philosophy class without Descartes, Spinoza, Leibniz, or Nietzsche; why, then, does the mathematical pedagogy persist on undervaluing its human component?

My experience with math is that when I graduated high school, I had a deep, fundamental, abiding hatred of it; it took nearly a decade to learn what I needed to to overcome this hatred. Mathematics is not valued in our society (but nor is literature); the problems turn on deep-seated pedagogical issues that tend to idealize math as a sterile, mechanical thing, instead of embracing its true nature as a very human endeavor, and a very human logical construct. To teach math right--teachers must humanize it, no matter how great the temptation otherwise.
____________
*For example, utilizing Archimedes' method of exhaustion to find a range for π (for fifth graders, for example).

Rebuilding the Diesel Network

In Gunn's Great Mistake, I outlined what I believe was SEPTA's greatest failure during the Center City tunnel construction era--namely that, in the name of tunnel electrification, SEPTA eliminated all of its diesel commuter services. While it is fair to note that the tunnel construction itself consumed all of the agency's resources, and thus that staff wouldn't have wanted to touch diesel-network issues with a ten-foot pole, I am also certain that had there been a will, there would have been a way.

Now I want to concentrate, going forward, on the moves necessary to recreating this network.

1. Build the Swampoodle Connection. A Swampoodle Connection, in its strongest form, has multiple uses: Not only does it shunt the Chestnut Hill West Line down the Reading trunk, it also offers a grade-separated junction between the Northeast Corridor and the SEPTA mainline. This is because (incremental) construction of the Connection is, at each step, an order of magnitude less expensive than the expense needed to rehabilitate Pencoyd back to the standard necessary for active rail service.

Because of this, Swampoodle would naturally include another connection beyond the ones I previously noted, as it would need to provide the following paths:

• NEC to SEPTA Mainline
• NEC to R6 Line
• Chestnut Hill West Branch to SEPTA Mainline

Of these, the first was never considered, but is not difficult to extend the design to add it in.

2. Work with NJ Transit for Newark Service. NJT's West Trenton proposal: After all, the combined SEPTA and NJT route reflects the Reading's Crusader alignment; a direct train would offer a secondary Philadelphia-New York commuter service to complement the very busy Keystone Service - Northeast Regional trains. With a high-quality Swampoodle Connection (and Kearny Connection at the other end), this would allow for regular Suburban Station-Newark Penn trains.

3. Integrate with Reading commuter rail. 'Nuff said.

4. Rebuild the Bethelehem Branch north of Shelly. Not only that, but extend it into Allentown. This would complement NJT's own Lehigh Valley designs. It would also help springboard work on the key (missing) regional connection from Philadelphia to Scranton and beyond.

5. Rebuild the Newtown Branch north of Fox Chase. 'Nuff said.

These projects--Swampoodle in the core supporting reconstructions and upgrading (return to service) of SEPTA's 1980 commuter rail network--would do more than just return our system (nearly) all the way to its Bicentennial extent, it would also offer a framework to hang more rail expansions from. For example:

6. Consider the Octoraro Branch. Kennett Square is a strong natural terminus, and the Brandywine Valley, while semi-rural, has centralized population centers. Reactivation of the Octoraro Branch, while requiring ROW reconstruction from Chadds Ford to Wawa, would also offer improved service to the Concordville/Painters Crossing/Brinton Lake area.

7. Consider the Perkiomen Branch. While the line up the Perkiomen has been abandoned for so long that the ROW has been severed in places, the region also holds several good conurbations crying out for better service: Collegeville, Schwenksville, and the Green Lane/Pennsburg/Red Hill complex, to name a few. Reclamation of this line would improve access to these towns.

8. Consider the New Hope Branch. While the line out of Ivyland runs through the at turns exurban and rural Buckingham Valley, New Hope, despite its size, offers strong destination value.

This list can be augmented until one runs out of historic easements. Pennsylvania has been blessed with a wealth of good, high-quality small towns--not all of them even near historic rail easements--and while their continued development is by no means tied to rail access, the access itself functions as an augment, offering improved transportation choices to these places. And the key is, a reborn diesel network functions as a framework for providing service to these awesome places. This is, more than anything, our ancestors' endowment to us: Isn't it time to make it be all it can be?

Gunn's Great Mistake

One of the major moves David Gunn made, while he was head of SEPTA back in the late '70s and early '80s, was the curtailment of its regional rail lines back into electrified territory, a move rail advocates have regretted ever since. The stated reason was that the new Center City Commuter Connection wouldn't allow scheduled diesel service, implicitly that the Reading Terminal was going to be closed down. But the problem is that this was a red herring: SEPTA's planners could have, with some clever easement tricks, retained diesel services into 30th Street or Suburban Stations. Below is one way I penciled out how.

That isn't to say there aren't structural weaknesses. Without a proper Swampoodle Connection, trains to Newtown and Newark (at minimum) are forced to use the freight alignment on the west side of the Schuylkill, for example; to avoid overloading this line, I use the Stony Creek Branch to bring trains out of Bethlehem into Philly via the R6 alignment. (A minor advantage of this setup is that it would have forced SEPTA to repair the Pencoyd Viaduct in 1986 instead of truncating the Ivy Ridge line back to Cynwyd.)

--The counter to this, though, is that the vehicles providing this service, Budd RDCs, were ending their design life at the time*, and their replacement, the SPV-2000, was so atrociously bad only the really clueless agencies ever ordered them.

--Countering that counter one would note that this was the era of the Comet I and the F40PH; again, had service retention been paramount, there would have been a way. But of course, all the political will of the era was focused on the tunnel, which leaves to us the task of picking up the pieces.

Floor and Ceiling

I just found about floor and ceiling functions the other day.

Briefly, a floor function,

is a function that returns the greatest integer value of a real without exceeding it, and a ceiling function,

is a function that returns the least integer value of a real once its value is exceeded.

Hence the floor of π is 3, and its ceiling is 4.

These types of functions are very useful when you've got a model whose natural output is in the real of e.g. the rationals, but whose output you need to be in the realm of the integers for the next step...This tends to happen in models built on combinatoric ideas (even if you didn't know they were combinatoric when you built them).

Slumlords

(Please note that this post is the intellectual chassis for some other work.)

Philadelinquency recently ran an excellent piece on how Philadelphia’s very poor school performance holds it back. Setting aside the chicken-and-egg problem of schools and class, let us focus on the final element in this piece, an element that ties back into the blog’s long-standing focus:

Now, about your suburban slumlord who smells the gentrification coming towards his rental property he was renting out for $600/mo and collecting a string of code violations on for a decade who might decide to sell his house to a rehabber and cash out, leaving that rental at the sake of increasing valuations?  Nobody has come up with a solution for that yet.

While this is a tie-in to pieces such as this and this, there is a more fundamental problem it touches on that needs addressing: Our zoning policy has been an abysmal failure at regulating landlords. Worse still, in its zeal to separate out homeowner and renter communities, it has resulted in a nasty unintended consequence: Slumlords are the result of the system.

Consider it for a moment. Time and again, sociological studies have shown that a landlord’s investment in his rental properties is directly tied to his geographical proximity to them. A landlord who lives in the same city is more inclined to invest in his properties than one who does not; in the same neighborhood, even more so; on the same block, ditto; and by far the most likely on premises. Since a slumlord is a landlord who fails to invest in their property, we can extrapolate that they are inversely correlated with distance: that is, the closer to their properties landlords live, the less likely they are to be slumlords. We can thus extrapolate that landlords of city property who live in the exurbs are likely to be slumlords; those who live in a different metro area entirely even more so. And guess what--they are!

It is not by accident that Philadelinquency spends most of its time chasing paper trails on slumlords who live far from the city. And in many cities, “institutional investors” are quite clearly slumlords-in-waiting.

But our claim, that institutional slumlords are an unintended consequence of our land-use policy, goes quite a bit further. To make this argument, let us recall how modern zoning came to be (see here, here, and here); they were implemented precisely because the homeowners of an affluent Cleveland suburb sought to keep renters out. And so it is unsurprising that modern zoning policy disenfranchises renters; what is a bit more surprising is that the jurisprudence required to get around earlier rulings also disenfranchise small landlords. And much as other side effects of “sorting” by use disenfranchised small businesses--to the benefit of larger malls, hypermarkets, and big boxes--so too has it benefited property management firms, and institutional investors.

Property management firms--companies of the type that run garden apartments--have full-time maintenance staff associated with each property. (In the absence of a landlord, a caretaker is the next best thing.) But institutional investors need not; all they need to maintain is the portfolio. Part of this is the--not unreasonable--justification that since they handle smaller properties (i.e. houses) than property managers, a caretaker per property would be excessive. But another part is that these organizations usually have a strong financial focus, often to the detriment to the properties they’re supposed to be managing. And of course, you also have bona fide slumlords who hide behind “institutional investor” masks.

Indeed, the whole system of institutional investing seems set up to encourage financialization and transactions at the expense of property maintenance. Is it any surprise, then, that to many people, “rentals” has become a dogwhistle for “slums”? Or that small rental properties are reflexively opposed, for the same reason development is in general?

One could say that the irony is that the system has come to disadvantage the small landlord, the homeowner who wants to add a granny flat above his garage, the community-minded owner who wants to fix that house up down the street and rent it out to a nice family, in favor of the institutional investor with Wall Street connections and falling-down flats. But that is just one irony buried in a whole system of deeper ironies. Perhaps it’s time to stand up and take notice.

Education and Slumlords

From this recent Philadelinquency piece:

Gentrification pressure in Philly affects less than 1/3 of the city’s total surface area and its rate of expansion is mostly kept in check by our horrible public schools above all other things.  And if anything, our public schools now suck more than ever; public opinion regarding the SDP has reached its lowest levels imaginable.

That basically means all of the gentrification we see in Philadelphia is missing a major factor that would cause it to explode here:  schools.   Everyone who has kids here either leaves Philadelphia or finds some other solution around the schools problem in order to stay here—or just doesn’t have kids.  Those with their kids in the School District of Philadelphia are either lucky their children are in one of the few performing schools the District has, or their kids are at the mercy of a system they can’t avoid.  That puts definite limits on the homebuying populations for sure.

You only have to look no further than the Penn Alexander School to see what fixing schools does to house prices:  parents pack into the school catchment which caused the area around Penn Alexander to shoot up a $100K prevailing premium over areas of West Philly just outside the school catchment.

But also consider this:   There are 4.4 million people living in Philly’s collar county suburbs and only 1.5 million living in the city proper.  If schools were suddenly fixed tomorrow, it’s likely that the cheapest homes in Mantua would be $450K.   Chew on that for a minute.  Your cheap house is mostly riding on the back of shitty schools keeping the bulk of parents who have means away.

Also note my response to this quote:

Now, about your suburban slumlord who smells the gentrification coming towards his rental property he was renting out for $600/mo and collecting a string of code violations on for a decade who might decide to sell his house to a rehabber and cash out, leaving that rental at the sake of increasing valuations?  Nobody has come up with a solution for that yet.

The shitty slumlords are the problem. And a major irony of our current land-use policies is that they, perversely, empower shitty slumlords at the expense of conscientious local landlords.

Urbanizing 30th St.

Back in September 2012 I wrote a post called The Space Between, in which I detailed urban design treatments linking the Penn and Drexel campuses together by extending the very successful Drexel platform a block further. I also proposed, more daringly, that the High Line be treated as the spine of a boulevard running from Penn Park up to JFK Boulevard, framing and defining a neighborhood-defining space.

The Space Between: Red is Drexel, Yellow UPenn, blue commercial, purple residential. Green is greenspace, and those two blue lines show the build-to lines framing 31st St. Boulevard.

The gist of this post is that: (1) Penn and Drexel ought to cooperate to extend the 32nd St. greenway/cap south to Walnut, (2) Penn needs to redevelop the 3201 Walnut garage, which has some cool features but is horrendous at street level, (3) it also needs to redevelop Rittenhouse Labs' Walnut streetwall, and (4) both have more to gain using the (quite active) High Line to define a 31st Street boulevard than mooning over a park proposal that ain't never gonna happen*. The Space Between also defines the key infrastructural and public-sector moves needed to fully urbanize the 30th Street area.

However, this area has two wings. Let us focus first on the south wing, defined by Penn Park.

Around Penn Park. The park is in green; blue is 100 Schuylkill; purple, proposed residential on air rights; yellow, Penn's land bank. In black note the extension of Schuylkill Boulevard atop the Schuylkill Expressway.

Fairly large chunks of the park, most famously the parking lot along Walnut, but also self-evidently the parking lot and disused greenswards anti-framing South, are land bank. The park was created in general to (1) sponge up and (2) help focus development around Penn's vast Schuylkill banks land bank.

But there is a second element, not as much considered: The park lies between the train tracks. While it is difficult to justify covering the western ones--at least until (unless) the Class of 1923 Ice Rink and Levy Tennis Pavilion plots are redeveloped--the Schuylkill's heavy traffic suggests that an extension of the surface avenue atop it to South is warranted. This is Schuylkill Boulevard.

And with the Boulevard, we create a platform for development. Now 30th St.'s south approach isn't just Penn Park's east edge, it's also very precious developable space sitting between it and Schuylkill Boulevard. This plan capitalizes on that by capping the NEC with five blocks--split by greenways--four of which would hold development, and one an extension of Penn Park to the river. A pedestrian bridge would also connect across the Schuylkill, onto the CSX crossing's west abutment, adding a valuable river link to these amenities. Finally, the 100 block of Schuylkill Ave. is highlighted (as if it couldn't be any more) as very valuable, riverfront, parkside developable space. Let's call this "Penn Park Place".

30th St. and Penn Park, with Drexel in red, Penn in yellow, commercial in blue, and residential in purple.

Between the space between and Penn Park Place, we have a relatively complete roadmap for redeveloping the 30th St. area: extend existing improvements south and east, and allow Penn (preferably in partnership) to redevelop the many banked parcels it owns; redevelop air rights on the riverside parcels.

Powelton air rights, from Drexelmasterplan

Finally, let us consider Powelton Yards, the vast coach yards north of 30th St. Station. Above is the site schematic produced for Drexel's master plan; below, I translate (and expand on) the general idea in a Google Maps map. Powelton Yards is a truly vast area: capping it would yield some 30 blocks of space. The height needed to clear 30th St.'s west approach, in the southwest corner, would create an artificial hillock, hence the name I've given this area as a neighborhood: Schuylkill Hill.

Schuylkill Hill. Blue are the Drexel Master Plan parcels; yellow, my extensions; red, the DMP streets; sky blue, my extensions; and green is parks and berms.

Between the core 30th St. area (including the Innovation Neighborhood), Penn Park Place, and Schuylkill Hill, a fairly extensive gull-winged western extension of the city core is taking shape. Let us hope that we fix the mistakes of our past, and don't make new ones we'll come to regret.
_____________
*Short of sinking billions of dollars into a bypass.

Capping Vine Redux

Remember this post? Remember this picture?

Well, with the new Mormon apartment building, I thought I'd bring it back to life, and put more meat on them bones.

Here we see, in purple, the major developments near Vine between Broad and Logan Square: the Mormon Temple, their Meetinghouse, the new apartment building at 1601 Vine, the Provence, and the apartment buildings buttressing Broad and Callowhill. Under it, we see the old cap plan. Intermingled with these, we also have some additions: several mews structures to take advantage of the increased desirability, and a small park over the exit ramp at 15th and Vine.

In the cloverleaf, we build a new "Sydenham Mall" (Sydenham being the name of the interstitial street between 15th and 16th); flanking it, large buildings fronting each of the four corners. They would share mechanicals and loading docks in the cloverleaf curve; inasmuch as possible, they would cover the interchange. Retail fronting Sydenham would lead up to the Provence*; retail fronting 16th would help urbanize that street further; flex space would front 15th.

Between all of these additions, the Vine Street Expressway would be completely hidden from 12th to 21st-ish. It would become the perfect example of the proper placement of heavy transportation infrastructure in the city--grade-separated, out of mind.**
___________
*If the Provence is built. As you may recall, I noted in A Provincial "Provence" that turning the cloverleaf green is, in fact, the worst possible for it.
**Remember, in Death and Life, Jane Jacobs noted that the most marginal uses--and hence the worst land values--were (in her time) usually located next to railroad tracks; this can be expanded to include Interstates today, for various reasons. Both are heavy infrastructure. The fact of the matter is that heavy transportation infrastructure, while indubitably necessary for the city to function properly (to get imports in and exports out à la her economies trilogy), also have a depressing effect on adjacent land. My extension of this point is--one of planning's major goals should thus be minimization of this effect by making heavy transpo infra "invisible" from an urban-dynamics point of view.