Daniel sent us this one — he's been watching Jerusalem's high-rise boom and it got him thinking. You know that feeling when you look at a lit-up skyscraper at night and realize it contains thousands of completely separate lives, businesses, little worlds stacked on top of each other? His question is: what if a high-rise wasn't just one building, but more like a structural platform — a mothership — where different developers could actually build their own independent mini-buildings inside it? Not leasing space, not renting floors. Actually constructing something modular and self-contained within the frame. He's asking if this approach exists, whether it has a name, and what the real-world limits are. And he draws a comparison to shipping containers — how standardization unlocked global trade. Could a building work the same way?
I love this question. And the shipping container analogy is exactly where you have to start, because it gets at both the promise and the central problem in one shot. A shipping container works because it's standardized — same dimensions, same corner castings, same load points. The ship, the crane, the truck, the port — they all know exactly what they're dealing with. The container doesn't surprise anyone. A developer-built pod inside a high-rise would be the opposite of that. It would be bespoke, unpredictable, and probably heavy in ways the original structural engineer never planned for. So the analogy is perfect, but it might also be why the idea hasn't taken off.
The shipping container as the thing that makes the case and then immediately breaks it.
But let's define terms first, because "building within a building" gets thrown around loosely. What Daniel is describing is not just a multi-tenant building or mixed-use. A hotel on floors ten through twenty with offices above it — that's still one structural system, one developer, one set of drawings. What he's asking about is a high-rise where the structural skeleton is permanent and shared, but the interior volumes are independently developed by separate entities who own their slice outright and build whatever they want inside it, within limits. It's the difference between renting an apartment in someone else's building and buying a lot in a subdivision and building your own house. Except the subdivision is vertical.
It's a condominium for commercial developers, basically. You buy a slot between floors forty and forty-five and you build your thing.
That's the idea. And it does have a name — or at least pieces of it do. In Singapore and Hong Kong they use something called strata-title vertical development. In legal terms, it's a three-dimensional cadastre — you own a volume of space, not just a floor plate. The Singapore Land Authority ran a pilot in twenty nineteen for the Duo Tower, which is a mixed-use high-rise where different volumes are legally distinct parcels. You can buy a cube of air sixty meters up and it's yours, registered, with a deed.
A cube of air with a deed. There's something deeply strange about that.
It gets stranger. In New York, air rights have been bought and sold for decades — but those are typically about the empty space above a building, the right to build upward. What Singapore is doing is parceling the interior volume of an existing structure. And this is where the legal framework is actually ahead of the engineering. The law can draw a box around a volume of space and say "this is parcel seven-B." The problem is that a legal box doesn't resist lateral loads during an earthquake.
Let's get into that. Can you actually build something structurally significant inside an existing high-rise without ruining it for everyone else?
The short answer is that we've been doing limited versions of this since the nineteen sixties, but full developer independence — where someone shows up with their own architect and builds a concrete-and-steel pod inside your building — that's basically never been done. The closest precedent is the John Hancock Center in Chicago, finished in nineteen sixty-nine. A hundred stories. It's what structural engineers call a megastructure — the exterior is a massive braced tube, those big X-braces you can see on the facade. Inside, different zones are structurally distinct. Residential occupies one zone, commercial another, parking another. Each has different floor-to-floor heights, different column grids, different mechanical systems. They're almost separate buildings that happen to share an elevator core and a foundation.
It was still one developer, one project.
One developer, one set of drawings, one structural engineer who knew what every zone was doing from day one. That's the key distinction. The Hancock Center was designed as an integrated whole. What Daniel's describing would be more like building the frame and then inviting developers to fill it in over time, each with their own design team, their own materials, their own construction schedule. And that introduces what the Council on Tall Buildings and Urban Habitat calls structural indeterminacy. The CTBUH published a major study on this in twenty twenty-one, looking at mixed-ownership towers. The core problem is that you don't know what loads are coming. If Developer A builds a heavy concrete data center on floor thirty and Developer B builds a lightweight steel-frame office on floor forty, your center of mass is now lopsided in ways the original wind-bracing didn't anticipate.
Wind matters more the higher you go.
Wind load on a tall building isn't just about how hard the wind blows. It's about vortex shedding — the way wind peels off the building's corners and creates oscillating pressure zones. Modern supertalls use tuned mass dampers, giant pendulums near the top that swing opposite the building's movement to cancel out sway. Taipei one-oh-one has a seven-hundred-ton steel sphere hanging near the top. Those dampers are calibrated to the building's specific mass distribution. If you've got independent developers adding and removing mass on different floors over the building's life, that calibration goes out the window. Literally — the building could develop sway patterns that make people seasick on upper floors.
The building equivalent of a washing machine with one heavy towel in it.
And it's not just comfort — it's fatigue. Steel and concrete don't like unpredictable cyclic loading. Over decades, unexpected sway patterns create micro-fractures at connection points. A building designed for a predictable load path suddenly has load paths nobody modeled. The Champlain Towers South collapse in Surfside, Florida in twenty twenty-one — ninety-eight people died — that was a single-structure failure in a conventionally designed building. Now imagine twenty independent structural entities inside one frame, each with their own construction quality, their own maintenance schedules, their own waterproofing. Who's inspecting the connection between Pod C's floor slab and the megaframe? Who's liable if it fails?
That gets to what I suspect is the real blocker here. Not the engineering — because engineers can solve almost anything if you give them enough money and clear parameters. It's the liability.
Insurance markets don't exist for this. If you're an insurer, you want to know exactly what you're covering. A conventional high-rise has a single structural engineer of record, a single general contractor, a single chain of liability. In a multi-developer platform building, if Pod B's contractor used substandard rebar and twenty years later the connection fails and causes a partial collapse, who pays? The pod developer? The building management organization that owns the frame? The original structural engineer who designed the connection points but never saw the pod's final design? No insurer will write that policy because the risk can't be priced.
You'd need some kind of building management organization, like a homeowners' association but for structural integrity.
That's exactly the model. And it's the only way this could theoretically work. The BMO would own the megaframe, the core, the lateral systems, the elevators, the plumbing risers. They'd sell or lease "structural slots" — pre-engineered volumes with fixed parameters. Maximum floor load, maximum vibration, approved connection types, utility tap locations. Each developer buys a slot and builds within those constraints. The BMO enforces a structural covenant — a deed restriction that says you can't exceed certain loads, you can't modify the connection points, you have to maintain your pod's fire envelope to a specified standard. It's a vertical industrial park with an extremely strict landlord.
A structural covenant. That's the phrase right there. You're buying air but you're also buying rules.
The rules would have to be draconian, because the stakes are catastrophic. This is where the Grenfell Tower fire in London becomes relevant. June twenty fourteen, seventy-two people died. The immediate cause was combustible cladding, but the deeper failure was compartmentalization — the building's fire strategy assumed that each apartment was a sealed box that would contain a fire for long enough for evacuation. The cladding created a path for fire to jump between compartments. Now imagine a building where each pod has a different developer, different materials, different fire-suppression systems. The compartmentalization is only as good as the weakest pod. If one developer cuts corners on fire-rated drywall, the whole building is compromised.
The fire code alone might kill this idea before it starts.
The fire code is the single biggest regulatory barrier. Each internal pod would need its own independently rated fire envelope, its own smoke control, and egress paths that don't compromise the host building's means of escape. If you're on floor fifty and the pod below you is on fire, your escape stair has to pass through or around that pod. The egress math becomes incredibly complicated. Most fire codes assume a single responsible party for the building's fire safety plan. With multiple developers, you'd need a fire safety governance structure that doesn't exist in current code anywhere.
Alright, but let's step back from the doom scenarios. What's actually been tried? You mentioned there are real-world partial implementations.
The Broad Group in China — this is the company that built a fifty-seven-story tower in Changsha in nineteen days in twenty fifteen, using prefabricated modules. That was a single-developer project, but it proved that you can stack large prefab units at high speed. The modules were built in a factory, shipped to the site, and craned into place. Each one was a finished apartment with plumbing and wiring already installed. The limitation is that they were all identical, or nearly so. The structural system was still a single integrated design.
That's the shipping-container model working — but only because everything was standardized.
And then there's SOM — Skidmore, Owings and Merrill — they developed something called the Brick system in twenty twenty-three. It's a flexible floor-plate concept for office towers where the structural bays are designed to be reconfigured. You can move walls, change mechanical zones, even shift floor openings without touching the primary structure. It's designed for adaptability over the building's life. But again — single developer, single structural system. The flexibility is planned from the start.
What about the Tokyo proposal you mentioned?
The Toranomon Podium City proposal from twenty twenty-four. This one is interesting because it's adjacent to what Daniel's asking about. The concept is a single sixty-story podium — a massive structural base — that hosts five independently designed towers on top. The towers are separate buildings, each with their own architect, but they share the podium's foundations, loading docks, and mechanical plant. It's not "buildings within a building" — it's "buildings on top of a building." But the legal structure is similar: each tower developer owns their volume, and a shared management entity handles the podium. It's the closest thing to a working model.
Foster and Partners had that Stack concept.
Twenty twenty-two. A residential tower where each floor is a prefabricated home that can theoretically be swapped out. If you want to move, you don't sell your apartment — you crane your entire floor unit out and take it with you. Or more realistically, you sell it to someone else who slots it in. It's a provocative idea but still single-developer. The building frame is designed for specific connection points and known weights. It's modular but not multi-owner in the way Daniel's describing.
We've got the engineering precedents — megastructures, modular prefab, adaptable frames. We've got the legal precedents — strata title, three-dimensional cadastre, air rights. And we've got something that looks like a management model in the BMO concept. None of it has been combined into a single building where independent developers build internal structures.
Because the combination creates a problem that none of those precedents had to solve individually. Let me give you a concrete example. Say you've got a sixty-story megaframe. The BMO sells a slot from floors thirty to thirty-five to Developer A, who wants to build a boutique hotel. The slot has pre-approved load limits: say, five kilonewtons per square meter live load, which is standard for hotel use. Developer A hires an architect who designs a beautiful interior with a two-story atrium, a mezzanine lounge, and a rooftop bar on the top floor of their slot. To make the atrium work, they need to remove a section of the megaframe's secondary beams. They submit a structural analysis showing it's fine — they're adding reinforcement elsewhere. The BMO's engineer reviews it and approves it. Five years later, Developer B buys the slot directly above and builds a swimming pool. The pool's weight plus the modified structure below creates a deflection that nobody modeled. Water starts seeping through micro-cracks into the hotel's ceiling. Who's at fault? Developer A for modifying the frame? Developer B for the pool load? The BMO for approving both? The original structural engineer who designed connection points that turned out to be too permissive?
Everyone sues everyone and the lawyers get a new wing on their own building.
And the insurance market looks at this chain of interdependent liability and says "no thank you." This is the fatal flaw. Not the engineering — we can engineer almost anything. Not the legal framework — Singapore's three-dimensional cadastre proves we can parcel airspace. It's the intersection of the two: the moment something fails, the liability web is so tangled that no one can untangle it.
Which brings us to Champlain Towers South again. That was one owner, one building, one structural system — and it still took years of litigation to sort out liability. Multiply that by twenty independent structural entities and the legal system just seizes up.
Champlain Towers had a single point of failure — deteriorating concrete in the pool deck area that compromised structural columns. In a multi-developer megaframe, you've got dozens of potential failure points, each owned by a different entity with different insurance, different contractors, different maintenance histories. The forensic engineering alone would take a decade.
Where does that leave us? The idea is tantalizing. It solves real problems — it lets development happen incrementally, it spreads risk across multiple developers, it could make high-rise space more accessible to smaller players who can't finance an entire tower. But the practical barriers are enormous.
I think the most promising path is a hybrid that sacrifices some of the developer freedom for predictability. Think of it as a vertical shipping container port. The BMO builds the megaframe with pre-engineered slots that have absolutely fixed parameters. You can't modify the frame. You can't change the connection points. You get a steel-and-concrete box with pre-installed utility taps, and everything you build inside has to fit within that box. Your loads are capped. Your vibration is capped. Your fire rating is specified. It's less "build whatever you want" and more "decorate this structural Tupperware.
The glockenspiel of architectural possibility.
It might be the only way to make it insurable. If every slot is identical from a structural standpoint — same dimensions, same load limits, same connection types — then the unpredictability problem largely goes away. The BMO's engineer knows exactly what loads the frame will see because the slots are pre-defined. The insurance model collapses from "twenty unknown entities" to "one known frame plus twenty identical slot policies.
That's the shipping container insight, really. The container didn't work because it was flexible. It worked because it was inflexible in exactly the right ways. Every container is the same size, same corner fittings, same twist-lock mechanism. The standardization is the innovation.
That's the part of Daniel's analogy that actually holds up. The question is whether developers would accept that level of constraint. Architects like to design. Developers like to differentiate. If every slot is identical, you're basically selling blank boxes in the sky. Some developers would love that — it reduces their risk, speeds up construction, lowers cost. Others would find it stifling.
There's something else here too. Even if you solve the engineering and the liability, you've got a governance problem. The BMO has to enforce the structural covenant. That means inspections, fines, potentially legal action against owners who violate the rules. In a conventional condo building, the HOA can barely get people to stop throwing cigarette butts off balconies. Now imagine an HOA that has to enforce structural load limits, fire envelope integrity, and vibration standards across twenty commercial developers with their own legal teams.
The BMO would need real teeth. Not just a board of volunteer owners, but a professional engineering staff with enforcement authority. It would function more like a port authority than a homeowners' association. The Port of Long Beach doesn't ask nicely if you'd please not exceed the weight limit on your container — they weigh it, and if it's over, it doesn't ship. A vertical BMO would need the same kind of operational authority.
That raises the cost. A BMO with professional engineers, inspectors, legal counsel, insurance — that's a significant overhead that gets passed to the slot owners. One of the selling points of this model is supposed to be that it makes high-rise development more accessible to smaller players. But if the BMO fees are substantial, it might cancel out the savings.
That's the economic tension. The model works best at scale — a very large megaframe with many slots spreading the BMO overhead. But the larger the building, the more catastrophic a failure would be, which means more stringent engineering, more expensive insurance, more intensive inspection. The economics might only close for supertall buildings in very expensive markets — exactly the kind of luxury towers that are already economically dicey.
Which brings us back to Jerusalem, actually. The prompt came from watching the luxury tower boom here. Ghost towers, mostly — forty stories of investment properties that sit empty while housing costs keep climbing. The platform model is appealing precisely because it could theoretically let different kinds of development happen in the same vertical space — affordable housing next to offices next to a community center, each developed by different entities with different funding models. But if the economics only work for luxury, the model doesn't solve the problem it's meant to solve.
There's one more precedent worth mentioning that gets at this. There's a researcher at IUAV University in Venice, Dario Trabucco, who's been working on what he calls "long-life, loose-fit" skyscrapers. The idea is that a building's structure should last a hundred and fifty years, but its interior uses should change every twenty or thirty. The structure is over-engineered at the start to accommodate unknown future loads. It's not multi-developer, but it's philosophically aligned — the building as a durable platform for changing uses.
Long-life, loose-fit. That's a good phrase. It's almost the opposite of what we build now, which is short-life, tight-fit — designed for one use, one developer, one set of assumptions, and when those assumptions change, the building becomes obsolete.
That's the real argument for the platform model, even if it's technically messy. Our current approach to high-rises is incredibly brittle. A tower designed as luxury condos in twenty twenty-four might be in the wrong location, wrong configuration, wrong everything by twenty fifty-four. If the structure can't be adapted, it either sits empty or gets demolished — and demolishing a sixty-story tower is astronomically expensive and environmentally disastrous. The platform model says: let's build the structure once, really well, and let the interior evolve.
The question isn't really "can we do this today." The question is whether we'll have to figure it out eventually, because the alternative — single-use towers that become obsolete — is worse.
That's where I think the next decade gets interesting. Two things are happening simultaneously. First, sensor technology and structural monitoring are getting dramatically cheaper. You can embed fiber-optic strain gauges throughout a building frame for a fraction of what it cost ten years ago. These sensors can report real-time load data to a central model. If every pod in a megaframe had instrumented connection points, the BMO could actually see the building's mass distribution shifting and adjust accordingly.
The building becomes a live structural model of itself.
And second, active damping systems are getting more sophisticated. Instead of a single tuned mass damper at the top, you could have distributed dampers throughout the frame that adjust their behavior based on sensor data. If the building's center of mass shifts because someone added a heavy pod on the east side, the damping system compensates. This is already being done in earthquake engineering in Japan — buildings that actively adjust their stiffness during a seismic event.
The indeterminate load problem might be solvable with enough sensors and active systems.
Solvable in principle. Expensive in practice, but costs are coming down. The bigger question is whether the regulatory and insurance frameworks can evolve fast enough. Building codes change slowly for good reason — when they get it wrong, people die. Grenfell changed fire codes globally, but it took a tragedy. The question is whether we can develop the governance and insurance models for multi-developer towers without waiting for a Champlain Towers-style failure to force the issue.
For someone who wants to follow this space — what should they actually watch?
First, the CTBUH — Council on Tall Buildings and Urban Habitat — they publish research on adaptable tall buildings and they're the main forum where these ideas get debated seriously. Second, Singapore's three-dimensional cadastre legislation. They're the global leader in parceling airspace legally, and any practical multi-developer tower would need a legal framework that looks a lot like what Singapore is building. Third, Dario Trabucco's work on long-life, loose-fit skyscrapers — he's publishing regularly and his ideas are filtering into practice through firms like SOM and Foster.
If you're in Jerusalem watching towers go up, the question to ask isn't just "who's going to live there" but "what happens to that building in forty years." A tower designed as a single-use luxury development is a bet that the luxury market will exist in that location for the full life of the structure. History suggests that's a bad bet.
The platform model is a hedge against uncertainty. You don't know what the building will need to be in forty years, so you design the structure to accommodate change. It's more expensive up front, but it might be cheaper over the building's life. The problem is that developers aren't incentivized to think about the building's life — they're incentivized to think about the exit. Build it, sell it, move on. The platform model would require a developer who plans to own the frame for decades, or a BMO that takes over after construction. That's a different business model entirely.
It's infrastructure thinking applied to buildings. We build bridges and ports and power grids as long-term platforms that multiple users connect to. We don't build them for a single user with a single use case. The platform model says a high-rise is more like a bridge than a house — it's infrastructure, and the units inside are the houses.
That might be the cleanest way to describe what Daniel's asking about. A vertical bridge that people build houses inside. The bridge doesn't care what the houses look like, as long as they don't exceed the load limit and they don't catch fire.
I like that. Though I suspect structural engineers would twitch at the analogy.
Let them twitch. It captures the idea.
To answer the prompt directly — yes, the concept exists in pieces. Megastructures, three-dimensional cadastres, air rights, modular prefab, active damping. No, nobody has combined them into a fully multi-developer internal building platform. The engineering is theoretically possible but currently uninsurable. The legal framework is emerging but incomplete. The governance model — the BMO with structural covenants — exists on paper but hasn't been tested at scale. And the economic case is unproven and probably only works in very expensive markets with very large buildings.
The limitations are exactly what you'd expect: fire safety and egress are the hardest regulatory problems, structural indeterminacy is the hardest engineering problem, and liability distribution is the hardest legal problem. Any one of these might be solvable alone. Together, they're a knot that nobody has untied yet.
Which is exactly why it's worth talking about. The ideas that seem impossible until suddenly they're not.
Now: Hilbert's daily fun fact.
Hilbert: In the nineteen fifties, a Byzantine-studies researcher named Egon Wellesz discovered that certain acclamations chanted for Byzantine emperors were tuned to resonate specifically within the Hagia Sophia's dome, effectively turning the building itself into a musical instrument — a property he verified by measuring reverberation times with equipment borrowed from a Tierra del Fuego seismological survey.
...right.
We've got Byzantine court ritual as architectural acoustics, measured with Patagonian earthquake gear. That's a sentence I didn't expect to hear today.
The cross-pollination of mid-century academic equipment is a whole genre unto itself, apparently.
That leaves us with one big open question. If sensor networks and active damping systems keep getting cheaper, and if Singapore-style three-dimensional cadastre law keeps spreading, does the "vertical bridge" model eventually cross from impossible to inevitable? The pressures pushing cities to densify aren't going away. At some point, the cost of not figuring this out might exceed the cost of figuring it out.
When that happens, the conversation shifts from "can we" to "how do we regulate it safely." The worst outcome would be a rushed implementation after a housing crisis creates political pressure to cut corners. Grenfell and Champlain Towers both happened because warning signs were ignored. A multi-developer megaframe with weak oversight would be those tragedies multiplied.
The boring stuff — the insurance, the inspections, the structural covenants — is actually where the future gets built. Or doesn't.
If you enjoyed this episode, please rate and review the show — it helps other weird-prompt enthusiasts find us. And send us your own weird prompt at myweirdprompts.We read every one.
This has been My Weird Prompts, with thanks to our producer Hilbert Flumingtop. I'm Herman Poppleberry.
I'm Corn. We'll catch you next time.
