Urban Technology at University of Michigan week -06
A 'teardown' of the Gratiot Ghost Kitchen
If you’ve been reading this newsletter for a little while you may know that I am a little obsessed with the “ghost kitchen” that appeared in a parking lot downtown Detroit a couple months ago. As Allison Arieff recently described, there’s lots of depth at the intersection of food, cities, and technology, so this week we revisit the Gratiot Ghost Kitchen for a second look. We’ve borrowed the concept of the teardown, where a product is disassembled to understand how it works, to unpack the situation on Gratiot and try to make sense of what we’re seeing. Is it the kitchen that’s ghostly or the missing building?
Hello! I’m Bryan Boyer, Director of the Urban Technology degree at University of Michigan that will welcome its first students soon. If you’re new here, try this 90 second video introduction. While we launch the program, we’re using this venue to explore themes and ideas related to our studies. Thanks for reading. Have questions about any of this? Hit reply and let us know.
👻 Ghost Buildings is More Like it
The more I pass by the heap pictured below, the more confused I get about what exactly puts the “ghost” in “ghost kitchen.” The term was coined to refer to restaurants that appeared via online delivery apps but did not actually exist as a physical space. Instead, their orders were serviced by some other food business, like a restaurant with multiple personality disorder. Dear reader, this most certainly exists:
Not only does it take up physical space in the world, but with each passing week it looks more and more like a full fledged building minus the facade. What started as just a trailer now has a portable toilet, dumpster, separate grease dumpster, generator, external fuel tank for the generator, external cooking gas tanks, water supply tank, waste water collection tank, and signage galore.
This thing is so much like a building, it even keeps junk in the basement (aka on the ground underneath the trailer unit):
If this ghost kitchen has most of the parts of a building, and serves the function of a building, why didn’t the creators just make a building to house their not-so-ghostly kitchen? Do they hate architecture or undervalue the public realm? Maybe! Perhaps it’s avarice? Sure, possible, but these are all simplistic. If you really look at the images, they depict a scene that is quite unusual for a western city like Detroit. (Side note: while plenty of places on earth enjoy active cooking in the public realm, that usually happens in settings even more ad-hoc than this, without the trailer as a quasi-building container). Detroit has has cooking formalized inside of things called buildings for a very long time, but the scene captured here shows one small example of the tide reversing, with traditional buildings apparently coming to be seen as an inhibitor rather than a utility. Why?
As a thought experiment we can revisit a tool from a prior newsletter and look at tech vs. urbanist perspectives.
The first thing that stands out is the legal question. Is it the idea of having a kitchen inside a building that is obsolete, or is it the process of making a building that has become clunky and onerous? A basic question: Is the not-so-ghostly kitchen legal? What should be a simple question is harder to answer than it looks, because we need to know what kind of thing we’re analyzing. Do we interpret the not-so-ghostly kitchen as a vehicle/trailer (which it is) or as a building (which it acts like)?
It has gas and water services like a building, but those are not connected to the municipal supply lines that run under the ground. It produces sewage, but instead of connecting to the city sewer, someone has to come to site on a regular basis, drop off an empty portable toilet, and haul away the full one. Like a building, this thing provides basic shelter from the elements and is a workplace for one or more people, but unlike a building it does not comply with the Americans with Disabilities Act, because steps are the only way into this structure. Good luck, wheelchair users.
Was a building permit required before creating this building-like thing or not? based on what I’m seeing here on the street, it is unlikely that a permit was requested. Which makes sense, really, because building permits are for buildings not things that are kinda like buildings. That tells us whoever created this assemblage wanted the qualities of something like a building without the hassles of creating (and maintaining) a structure.
Playing in this gray area is a question of risk tolerance. Unless you have nearly infinite money, you probably do not like the prospect of getting sued, and messing around with buildings is a great way to get sued because, traditionally, the built environment has been very protected due to the safety implications of poorly constructed homes and workplaces. Reef Technology, who operate this kitchen, raised $700M last year from venture capital firms, meaning that they have plenty of money on hand to fight law suits if they arise. They can afford to make something that’s kinda building-like and wait to be sued. Bring it, universe! But not so fast, because risk tolerance is not merely a question of how costly things may be when they go wrong. Risk also involves a calculation of the likelihood that something bad might happen, which means we need to care also about risk mitigation.
If Reef were a traditional business they might hire an architect, develop plans, and then design a little building that meets all the proper codes and gets the right permits (if planning is done properly, risks will be eliminated). But since they’re a not a traditional business, instead Reef Technologies works in a more iterative manner, testing and prototyping various approaches to find out what works. From their perspective, the law inhibits flexibility and speed, so they need to find out how much they can stretch the law without causing some kind of major backlash. That means this heap that we’re examining in Detroit probably doesn’t look quite the same as Reef’s deployments in New York, Seattle, or Miami. Since the business operates as a network of ghost kitchen nodes around the country (in an exponential growth model), if one or two of them get closed down by local governments unhappy with the situation at a single node, the overall health of the business as a network is not detrimentally affected. Both their bank account and their business model allow them to experiment in ways that a mom-n-pop business cannot.
Another aspect of risk that we should consider here are risks (and benefits) to who? Are the humans that the business cares about focused narrowly on their own customers or users, or is there a broader concern for the public? In this case, the not-so-ghostly kitchen is focused on satisfying the desires of customers making orders, so if the appearance of the heap happens to degrade the dignity of the public realm, they don’t really seem to care. Or if the un-inspected (or lightly-inspected?) gas installation next to the sidewalk present risks to the public who may be passing by, so be it. That’s because the kitchen is treating the parking lot like a commons that is to be utilized for maximum private gain, rather than a commons that exist as a shared resource where norms help maintain a societal homeostasis.
If we analyze this story from the perspective of a technologist, we might observe that the formal processes for building spaces are “too slow” or “cumbersome,” and we might pride ourselves on combining a business model, financing, legal outlook, relationship to the customer, and operational skill to manifest a new approach to getting food to people in cities. Flip sides and look at it from the perspective of the urbanist, and this is a model that prioritizes private benefits at the expense of increased risk to the public, resulting in a degraded built environment, and for what reason? To get a new restaurant? Not worth it! Reconciling those perspectives is hard. Maybe impossible?
Instead, this little heap on the corner in downtown Detroit is more useful as a way to think about the challenges of how we see, shape, and inhabit cities in this digital century. It’s a symptom of an ailing system that needs the imagination of people who understand cities and technology both.
If we hold constant a deep ethical concern for the public and the public realm while also taking a hard look at the baroque supply chain of goods and services that currently control the built environment, it’s hard not to wonder if there are better ways to make everyone happy. What might an urban technologist do? Three of many, many possibilities:
See: Help city leaders have a more precise and up-to-the-moment picture of safety issues before they become too risky, perhaps by building a network of sensors that monitor temporary installations in an open, transparent, and realtime manner. Might this allow citizens to feel more secure that experimental approaches such as the not-so-ghostly kitchen can be tried without putting anyone at undue risk?
Shape: Streamline the building permit process by creating automated plan review systems, which in turn would require more information about the built environment to be kept in structured data formats such as “building information models”—something which is currently common for large projects but rare for small local efforts in the US. Why isn’t software radically better and radically cheaper?
Inhabit: Borrow the idea of parking lots as experimentation grounds and develop a more robust program to encourage ‘meanwhile usage in this parking lot and elsewhere. Alternatively, US cities could take a masterclass in night markets from Taipei.
Links
🍁 Front Porch Forum is like Nextdoor but a public benefit corporation and lots of people in Vermont love it.
📐 “Techniques of Use: Confronting Value Systems of Productivity, Progress, and Usefulness in Computing and Design” is a new paper from colleagues at the UM School of Information. Video here.
🐴 The mega-corporation Salesforce is considering building… a ranch? Huh.
🪚 Detroit’s Belle Isle park once had a saw mill! Pairs nicely with a past discussion of Michigan timber construction possibilities.
🧑🎨 A local in Detroit took it into their own hands to clarify parking regulations with a can of paint.
This week: Gail taught us how to build a machine. Friday included a call with civic design powerhouse Ariel Kennan. Kickoff for a research projects with <REDACTED>. Project update with Christopher on a drones project now underway. Dropping in at Harvard for MDes final reviews and Stanford for a guest lecture. Hey, the weather is nice out—hope you’re reading this with a glass of iced tea 🏃♂️