A double-wall structure with an insulating air cavity between two thermal mass walls provides superior cold weather survival compared to single-wall construction because the cavity interrupts heat transfer in stages, with the outer wall absorbing and slowly releasing cold while the still air gap further blunts temperature transfer before it reaches the inner wall, maintaining a survivable interior temperature even when exterior temperatures drop to -20°F.
Deep Dive
Prerequisite Knowledge
- No data available.
Where to go next
- No data available.
Deep Dive
Homeless Boy Stacked Free Cinder Blocks Into Double Walls So Thick They Beat Minus 20
Added:You already know what minus 20 does to a body sleeping outside. You felt a cold that came through a wall and sat on your chest while you slept. You know what it's like to wake up and not be able to feel two of your fingers. You know that specific dread. The moment in October when the wind shifts and you realize what's coming and you have nothing.
Decker Hail knew all of it. He was 22, living in a city that didn't have any shortage of abandoned lots and old industrial buildings falling apart at the seams. Eerie, Pennsylvania, right on the lake. The kind of cold that rolls off Lake Erie in January is not like regular cold. It carries weight. It pushes through walls that weren't built right and under doors that don't seal and through gaps you never noticed in October because October hadn't tried to kill you yet. January and Erie tries to kill you. Decker had been outside for nine months when he figured out the blocks. Not nine months in a row. He'd had a room at a transitional shelter for 6 weeks around March, then lost it over a rule about curfew. He'd slept in a friend's car for almost 3 weeks in early summer until the friend's situation changed. The rest was his. A patch of semi-sheltered space behind a defunct machine shop on the east side of the city where the building's rear wall still stood solid and the city hadn't gotten around to tearing the whole thing down yet. He'd built something there by the time October came. Something that almost counted as walls. Almost wasn't good enough for January in Eerie. This story is about what he built instead.
Decker was the kind of person people didn't look at twice. tall, narrow through the shoulders, with the kind of lean that comes from months of not eating enough and moving a lot. He kept his hair buzzed close on the sides, and left the top a little longer, a dark auburn that had started to rust out at the edges from sun and wind. There was a stripe of white hair over his left ear, where a scar ran through the scalp, an old thing from before any of this, something that had healed badly and left that permanent pale streak. It was distinctive enough that some of the regulars at the drop-in center on 18th Street knew him by it before they knew his name. They called him Stripe for a while until he introduced himself properly. He wore a canvas jacket the color of old pine needles. Someone had left it at the shelter in March and it sat unclaimed for 11 days and then it was his. Both elbows had worn through by June. He'd wrapped them in duct tape, then wrapped the tape again when it started peeling. His boots didn't match.
One was brown, one was a dark charcoal that read as black from more than 3 ft away. Both were laced with orange paracord he'd gotten from a hardware store's remnant bin for free. He walked in them everyday, and they held together, and that was what mattered. He wasn't from Eerie originally. He'd grown up in Meville, about an hour south, which was smaller and quieter, and had its own problems, but not quite the same ones. He'd come to Erie at 19 for a job at a distribution warehouse that turned out to pay significantly less than the listing it said, and require mandatory overtime that still didn't cover first and last months, rent anywhere near the facility. He'd stayed at the warehouse for 7 months, sleeping first in a room he shared with two other guys from the warehouse, then one room when one of them left, then no room at all when the landlord raised the rent by $280 and gave 2 weeks notice, and neither Decker nor his remaining roommate could cover it. His roommate went back to his family in Cleveland. Decker didn't have that option in any practical sense, so he stayed in Erie. He stayed because leaving costs money and he didn't have it. And because he'd been there long enough to know where the food was and which shelters had beds and which drop-in places stayed open late and that knowledge felt like something even when everything else didn't. Knowledge of a place is its own kind of weight. You can't just abandon it. He worked when he could. Day labor mostly, sometimes loading, sometimes clean up on job sites. The day labor office on State Street was a 40-minute walk from the east side where he camped. And he made it most mornings by 5:45 a.m. because the dispatch opened at 6:00 and the early guys got the good spots. Not every morning had work. Some weeks he'd get two days out of 5. Some weeks three.
Enough to eat and enough to buy small things. Not enough to stack toward a deposit anywhere. He was not waiting to be rescued. That's something to understand about Decker before any of the rest of this. He wasn't sitting still hoping for a turn. He was solving problems with what he had the way you do when you don't have the option of buying solutions. The cinder blocks were a solution. It just took him a while to arrive at them. The space behind the machine shop was maybe 24 ft wide and 30 ft deep with the building's back wall forming one solid side. The rest was open. When Decker first started using it in early spring, he didn't think much about the winter. Spring is like that.
You solve the problem in front of you, and winter is abstract and far away, and there are things to figure out right now that are more urgent. He started with a tarp. He had one, a blue poly tarp with grommets, 10 by 12 ft, that he' pulled off a pile of stuff someone had left on the curb. He strung it diagonally across one corner of the space, creating a triangular pocket against the back wall.
In April, that was enough. In May, that was enough. In June, the nights were warm enough that it barely mattered.
July, he got a sleeping bag from the Saint Vincent Depal store on Parade Street for $8. A blue and gray one rated to 30° F, which would matter in 6 months, in ways he couldn't fully feel in the heat of July. He built the floor before he thought about the walls. That came from something a guy at the drop-in center told him. A man named Roland somewhere in his 50s, who'd been living rough in this city for a long time and knew things that didn't get written down anywhere. Roland said the ground will take your heat faster than the air ever will, said it like it was a fact he'd paid for personally. Decker believed him and started collecting wooden pallets.
Pallets are not hard to find in a city with light industrial uses. factories, warehouses, restaurants, loading docks.
They generate pallets constantly, and most of them end up stacked in a pile outside or broken down in a dumpster.
Decker found four behind a restaurant supply place on Sassifrass Street that were in reasonable shape and dragged them one at a time the six blocks back to the machine shop. He laid them flat over the ground in his corner, fitting them together like a rough puzzle. The floor sat about 4 in off the ground, enough air gap to interrupt the cold transfer Roland had warned him about. He slept better after the pallets, noticeably better. Roland had been right, but Roland had also told him in the same conversation that pallets and a tarp wouldn't be enough come January.
That what the lake sends isn't just cold air, it's wind, and wind finds every gap, and a tarp has nothing but gaps.
Roland said, "You want mass. You want something that takes a long time for the cold to push through. You want something that stores warmth when there is warmth and gives it back slowly when there isn't." He said, "Concrete does that.
Stone does that. Anything with real weight does that." Decker asked him where he was supposed to get concrete.
Roland said he'd figure it out. Said that was the difference between people who made it and people who didn't. Not luck, not resources, just the patience to look at what existed before giving up on it. Decker didn't find the cinder blocks right away. He spent most of August and into September thinking about what Roland had said and looking at what was around him with different eyes. The machine shop was crumbling on its front face. He could see through gaps in the brick on one side where the mortar had failed. But that brick was part of a standing structure, and he wasn't going to touch it. The lots to either side were mostly broken asphalt and gravel.
Nothing there. He started noticing other lots. Erie has a lot of lots. The city had been contracting since its manufacturing peak and there were abandoned properties in various stages of collapse all over the east side. He walked through them on his way to the day labor office on his way back on afternoons when there was no work. He was looking for what Roland had described without being sure he'd recognize it. He found the first pile on a Tuesday in late September on a street three blocks north of his space. A former residential lot, the house long gone, the foundation still partly there and stacked against what remained of the foundation wall, may be 30 or 40 concrete blocks. standard 8 by 8x6 cinder blocks, the hollow kind with two chambers running through them. Not clean, not new. Some were cracked, some had old mortar stuck to their faces, but most were intact. They were just sitting there. He stood looking at them for a while. He went back the next day with a city property records print out from the library. The lot was listed as owned by the city of Erie. Tax delinquent, no active building permit, no active demolition permit. No one was claiming it. No one was coming for those blocks.
He moved one, just one that first day to see how heavy it was and whether his back could manage it. A full cinder block weighs between 28 and 36 lb, depending on the mix. The ones on that lot were on the heavy side of that solid dents, the kind they don't make as often now. He carried it three blocks with both arms wrapped under it, set it down against the machine shop wall, and stood there breathing hard for about 2 minutes. Then he went back and got another one. He moved blocks for 11 days. Not all day, he still made the labor office in the mornings and worked when work came, but afternoons and evenings, sometimes in the dark, with a flashlight borrowed from a guy at the shelter, he moved blocks, two at a time when he could manage it, one at a time when he couldn't. He ended up with 94 blocks staged against the back wall of the machine shop before he stopped when the original pile was mostly gone, and picking through the remaining cracked ones didn't seem worth it. 94 blocks. He stood and looked at them and thought about what Roland had said about mass.
He didn't mortar them. This is the part that surprises people. The assumption is that you have to mortar blocks for them to do anything useful. That's what the building industry sells you. You need mortar. You need rebar. You need a footing. You need permits. You need professionals. You need to buy a solution. What Decker figured out through trial and error over two weeks of building is that dryst stacked cinder blocks laid carefully level interlocked at the corners are stable. Not infinitely stable. Not earthquake proof, but stable enough for walls that aren't being loaded from above. Stable enough for walls at human height. Stable enough for what he needed. He built two walls.
That's the part that matters. Not one wall, two. He built the outer wall first. It ran about 11 ft along one side of the corner space and about 8 ft along the other side. Closing off the open sides of the triangular space and extending the protection beyond the tarp he'd been using. He stacked them three blocks high, about 2 feet, then four, then five. Five courses high as 40 in.
It reached his chest. He staggered each course by half a block the way actual masons do. So, the vertical joints didn't line up and the structure locked into itself. It took him 4 days to get both runs of the outer wall to five courses and to get them plum enough that he trusted them. Then he started the inner wall. This is what Roland's advice actually pointed at. Even though Roland had never said the words double wall, he'd said mass. He'd said something that takes a long time for cold to push through. A single 8-in block wall has some thermal resistance, but not much.
Air moves through the hollow cores, especially when wind finds an angle. But if you build a second wall inside the first with a gap between them, the gap itself becomes insulation. Still, air doesn't conduct heat the way moving air does. Still, air trapped between two masses of concrete is a genuine thermal barrier. The gap Decker left was about 4 in. He had no engineering reason for that number specifically. He'd read something at the library. He went to the library often. It was warm and no one bothered him if he was quiet. About cavity wall construction in old British housing, and the gap in those walls was typically 2 to 4 in. He went with four because it left him enough room to stuff insulation into it if he ever got any.
He built the inner wall two courses lower than the outer wall. So from outside you saw a block wall five courses tall. From inside you were looking at a wall three courses tall 24 in with the taller outer wall above it and a cavity between them running the full height of the outer wall. The inner wall was enough to keep his body from radiating heat directly through the structure. The cavity and the outer wall handled what came from outside. The two walls together at their thickest, five courses of block plus 4 in cavity plus three courses of block measured about 28 in from outside face to inside face. 28 in of wall. When he finished and stood inside the completed structure for the first time, it was an afternoon in mid-occtober and the temperature was 44° outside with a stiff wind off the lake.
inside his three-walled corner, back wall of the machine shop, two runs of double block wall on the open sides, the blue tarp overhead. It was still, noticeably still, not warm, but the wind was completely absent, and the air didn't feel like it was moving, and that stillness had a quality to it, a quality of mass that he hadn't felt in any of the outdoor spaces he'd used before. He sat on his pallet floor and thought about January. He added a door, not a real door, but he'd found a section of solid core interior door, doorork knob still attached, in a pile of demolition debris behind a property on East 26th Street. It was heavy, a real solid core door, not a hollow one, which are common in the debris piles, and it was intact except for some water swelling at the bottom corner. He dragged it back in pieces of a trip, leaning it against the machine shop wall, and then spent an afternoon figuring out how to use it. He cut or had someone at the day laborard cut in exchange for 20 minutes of help loading a truck, two short sections of 2x4 from a scrap pile. He wedged these sections between blocks in the opening of his outer wall at an angle that held the door in place when he pushed it closed. It didn't seal perfectly.
Nothing does. But a solid core door standing in a block opening with friction fits on two sides is substantially better than a tarp hanging over a gap. It has mass. It blocks wind.
He found a section of pipe insulation at a second building supply dumpster. The gray foam tube kind that goes around water pipes slid up the side. He cut it into strips and pressed it into the seam where the door met the block frame on three sides. The bottom he left alone, sealing a threshold completely creates a different set of problems in damp weather. The door made the inside noticeably quieter. Sound is mass dependent the same way thermal transfer is. When he closed it and stood inside, the sounds from the street a block and a half away went from present to background. November came. The first hard freeze of the season came on November 12th. Eerie dropped to 19° Fahrenheit overnight with wind chills pushing it to single digits. Decker had been tracking weather forecasts at the library, printing them out when the library let him use the printer for free, and he knew this one was coming.
He'd spent the week before it laying in as much insulation as he could find.
Insulation is where the commercial version of this story diverges from the free version. A hardware store sells you rolls of fiberglass bat insulation or rigid foam board and charges you accordingly and the product works exactly as described. What the hardware store doesn't tell you because there's no profit in it is that the same thermal function can be approximated with materials that get thrown away constantly in every American city.
Decker used three things. He used moving blankets. two of them thick furniture pads that moving companies use and that he'd found being thrown out behind a storage facility on Depot Road when they were too worn to rerent. Moving blankets are dense fibers tightly stitched into a quilted pad. They don't insulate as well as proper fiberglass bat by R value, but they absorb radiant heat and they slow conductive transfer and they're real and heavy and they can be hung or stuffed or draped or folded as needed. He stuffed the moving blankets into the cavity between his outer and inner block walls.
The 4-in gap was just wide enough to compress them in from the top. He had two blankets, and they covered most of the cavity on both sides of the corner.
The sections they didn't reach. He filled with crumpled newspaper, a poor insulator, but not nothing, and it was free and available, and the library kept a pile of discards by the door. He used cardboard for the ceiling. The tarp stayed, but above the tarp he layered cardboard from boxes he'd collected, broken down flat, stacked three deep, held in place by lines run between the outer block wall and the machine shop wall. Cardboard is not waterproof, and it's not fires safe, and it compresses and loses effectiveness when wet. But dry cardboard layered over a waterproof tarp is a real radiant barrier. It slows the loss of heat upward. It holds something in. The third thing was the floor. He added a second layer to the pallet platform, flattened cardboard boxes between the pallets, and a section of outdoor carpet remnant he'd found rolled up in an alley behind a flooring store. Carpet over cardboard over pallets, three layers between his sleeping bag and the ground. On the night of November 12th, he went inside at 9:00 p.m. when it was 22° and falling. He noted the temperature inside his structure when he went in. He had a small digital thermometer clipped to a pallet, the kind sold for wine fridges, which he'd found at a Goodwill for $2.
Inside when he entered, 38° cold, but not the same cold as outside.
He woke up at 4:00 a.m., a habit from months of early mornings for the labor office. He noted the temperature again, 34° inside. outside at that hour, according to the forecast he'd memorized, was 19°, 15° warmer inside, with no heat source, but his own body. He lay there in his sleeping bag for a while thinking about that. His 30° bag had kept him comfortable at 34° ambient. His bag would technically fail at 30° ambient, meaning he'd be cold and possibly at risk if it went lower. 34° was enough margin. He went back to sleep. December was when he started understanding what the double wall actually did that a single wall wouldn't have. A single wall of block five courses 8 in thick would have brought outside temperatures inside faster. The blocks themselves conduct heat. Given enough time at minus temperatures on the outside face, the cold works through the material and into your space. How fast depends on the material's thermal mass and its R value.
Concrete block doesn't have a high R value. What it has is thermal mass, the ability to store energy and release it slowly. But in a single wall, that thermal mass only works in your favor for so long before the cold winds. The cavity changes that math. The cavity, those 4 in of still air between the two walls, has an R value of roughly one per inch of still air. 4 in of still air is R4 roughly. Not impressive on paper, but in combination with the thermal mass of two separate block walls, it creates a system that buffers the cold in stages.
The outer wall absorbs the cold, holds it, releases it slowly. Some of that never reaches the inner wall. What does reach the inner wall encounters the still air cavity first and the cavity slows it again. By the time any temperature transfer reaches the inside face of the inner wall, it's been blunted twice. You can test this yourself with a thermometer and a clear cold night. Stand next to a single concrete wall after it's been cold for 6 hours. Then find a double wall cavity structure that's been cold for 6 hours.
The inside face of the double wall will be noticeably warmer than the inside face of the single wall. Not dramatically warmer, but noticeably.
Decker understood this not through the physics, but through the experience. By December, he had a data point on the December 3rd forecast. The overnight low was 11° F. He noted inside temperature when he went to sleep. 41° warmer than the November night because he'd been in the space for hours before sleeping and the walls had absorbed some of his body heat which they then held.
He woke at 3:00 a.m. Inside 32° outside 11° 21° warmer inside than out.
That's what the double wall did. It didn't heat anything. It didn't generate warmth. It resisted the movement of cold from outside to inside well enough that his body heat and the thermal mass of the structure together held the interior at a temperature his sleeping bag could manage. He was not warm. 32° is not warm, but 32° and a 30° bag is survivable. Survivable is not a small thing when outside is 11. He reinforced the wall in late December. He found more blocks, a smaller pile, maybe 30 of them, behind a demolished garage on Parade Street. He used them to extend the outer wall on one side by another 4 ft, which closed off a gap he'd left in October because he'd run out of blocks.
He also used eight of them to add a sixth course to the outer walls most exposed run, the side that faced northwest, which was where the lake wind came from. Six courses is 48 in. That's four feet of wall on the windward side.
He also filled the cavity gaps he hadn't been able to reach with the moving blankets. He shredded a bag of old sweaters, things he'd accumulated that were too far gone to wear, and stuffed the shredded fiber into the remaining cavity spaces by hand. It worked, not as well as proper insulation, but it worked. He built a secondary windbreak at the door. He had a second smaller tarp 6x 8 ft that he rigged as a vestibule of sorts, stretching from the outer wall out about 3 ft on two sides of the door opening and anchoring it to the machine shop wall. When he opened the door, the vestibule trapped the first wave of outside air before it entered the main space. When he closed the door, the vestibule reduced the wind pressure against it. It sounds small. in a northwest wind at 15 mph. It's not small. By the time January came, the structure looked nothing like what he'd built in October. It had grown and been layered and reinforced and adjusted at every point where experience had shown him a weakness. That's what winter does.
It shows you the gaps not as a lesson, but as a threat. And you close the gaps or you don't. And the stakes of not are real. January 14th, Erie woke up to minus4 Fahrenheit. Not wind chill, actual ambient temperature. Wind chill pushed the felt temperature to minus22.
The National Weather Service issued a wind chill advisory starting at 6:00 p.m. the previous night, and Decker had seen it at the library at 5:00 p.m. and been inside his structure by 700 p.m. He spent the evening adding every additional layer he had. another cardboard sheet over the ceiling tarp, his spare clothing stuffed against the base of the inner wall. The door braced with an extra wedge of wood he'd been saving. He went to sleep at around 1000 p.m. The inside temperature was 44°, warmer than usual because he'd spent 3 hours in the enclosed space before sleeping. He woke at 5:15 a.m. because that was when he always woke, his body trained by months of early starts. He checked the thermometer. 28° inside. He lay there calculating outside was -4.
Inside was 28. That was 32° warmer inside than the ambient outdoor temperature, not counting windchill, which outside would have made it feel like -22.
Inside there was no wind. The walls stopped wind completely. There was no gap in the double block structure large enough for wind to move through. inside.
It was still 28° in a 30° sleeping bag is at the edge. He was cold. His feet were cold. His face, which was outside the sleeping bag, was cold in a way that registered as a warning. He pulled his jacket over his head. He slept in his jacket always and tucked his face into the collar. He was not in danger. This is not a story about a close call. This is a story about a system that worked.
At 5:15 a.m. on the coldest night Erie had had in 6 years with the actual temperature at minus4 and the windchill at minus22. Decker Hail was inside a structure he had built from discarded concrete blocks and a borrowed idea about mass. And his core temperature was fine and his extremities were cold but not dangerously cold. And he was calculating in his head how the morning was going to go and whether the labor office would be open. The labor office was open. It was never not open. The guys who ran it understood that the people who showed up at 6:00 a.m. and minus 4 were the people who would show up for anything. And those were the people you dispatched first. He got work that day, shoveling at a commercial property on 12th Street. 8 hours in the cold with a shovel in exchange for cash and a hot meal. He walked back to his structure that evening and sat on the pallet floor. And the inside temperature after a full day of being unoccupied in a day that never got above 3° was 24°.
Still above outside, still doing what it was built to do. The thing about what Decker built, and this is the part that doesn't make it into the conversation about homelessness, which is usually about systems and funding and policy, and rarely about what the person in the situation actually knows, is that it was replicable. Not in every situation. Not without access to the kinds of abandoned lots that Erie has in quantity. Not without the ability to move blocks over distance repeatedly, which requires a functional body and time and the specific stubbornness to keep going back. But the knowledge is transferable.
The principle is clear. Two walls are better than one wall because the gap between them interrupts the transfer of cold in a way that one wall cannot. Mass is better than lightweight materials for thermal stability because it stores and releases slowly. The ground will take your heat faster than the air if you let it. So, don't let it. Still, air insulates. Wind destroys the thermal advantage of every other measure you've taken. So, stop the wind first and worry about everything else second. None of that is proprietary. None of that is sold by a company, Roland told Decker some of it in a conversation at a drop-in center. Decker figured out the rest by moving blocks in the cold and paying attention to what happened. What the industry sells you is ease. Pre-cut insulation at exactly the right R value.
Spray foam that fills every gap in one application. A door that seals because it was manufactured to seal. These things work better than what Decker built. They are also not free. And for someone in Decker's situation, not free is the same as not available. The gap between what exists and what costs money is where Decker built his answer. That's the commercial interest in keeping the knowledge abstract. Not a conspiracy, not a deliberate suppression, just the ordinary drift of information toward what's being sold. The free version, blocks, moving blankets, cardboard, newspaper, doesn't have a marketing budget. Nobody's running an ad for drystacked cinder block construction. So the knowledge sits with people like Roland who tell people like Decker in conversations nobody records and it moves slowly and person by person and a lot of people never hear it at all.
February. By February Decker had been in the double wall structure through 12 nights below zero and one night at - 11 Fahrenheit ambient, the worst single night of the winter, and had woken up from all of them. That is the measure.
He woke up from all of them. The structure had changed his relationship to the winter, not eliminated the hardship, not made the cold comfortable.
He was still cold every day. His fingers were still cracked and bleeding from the dry air and the repeated exposure to outdoor work. His mornings were still the effort of convincing a cold body to move when every nerve ending argued against it. The winter was still the winter, but the structure held. Every morning that he woke up and checked the thermometer, and saw the interior temperature still above the outside temperature, the gap between them was the structure working, the blocks holding, the cavity doing what it was there to do, 28 in of wall that cost him nothing but 11 days of labor, and the specific willingness to pay attention to what Roland had said about mass. He started talking to other people about it. Not officially, not as any kind of program or initiative, just the way Roland had talked to him in conversations at the drop-in center, at the labor office waiting room, wherever people were dealing with the same problem he'd had in October and hadn't yet figured out their own version of November. He told them about the blocks first because that was the question people always had. Where do you get them? In Erie specifically, he could direct people to three lots he knew of that still had blocks. Different cities have different lots and different demolition patterns and different city property management, but the principle travels wherever there is aging infrastructure being demolished. Blocks exist. Wherever there is urban contraction, and a lot of American cities are in urban contraction, there are blocks. He told them about the cavity second because that was the part people didn't intuit it. One wall feels like a wall. People understand one wall.
The second wall, the gap that takes explaining. Why not just make the single wall thicker? Because thickness alone in block adds mass without adding the insulating value of the still air gap. A 16-in thick solid block wall is harder to build than two 8-in walls with a gap, and it doesn't perform as well thermally. The cavity is the point. The gap is the point. What's between the walls is as important as the walls themselves. He told them about the floor third, because Roland had told him that last, and it's the thing people are most likely to skip. The ground will take your heat. Stop it first before you worry about anything above your knees.
March came. The lake effect eased off.
Temperatures stopped dropping below 20° after mid-March. The urgency lifted the way it always lifts in March. And people who'd spent winter in a posture of raw survival started to breathe differently, to hold themselves differently, to look at the world with slightly more available attention. Decker kept the structure standing. He had no reason to dismantle it, and every reason to keep it ready. Winter comes back. He made one addition in March when the ground was soft enough to dig. He dug a shallow trench along the base of the outer wall, 6 in deep, 8 in wide, and lined it with smaller pieces of block and flat stones he'd collected. The trench didn't do anything structural. What it did was interrupt the path of rainwater flowing toward the base of the wall. In winter, water that collects at the base of a dryst stacked wall and then freezes, can shift blocks. He'd seen one section of the outer wall move slightly over February, just a small displacement, nothing that threatened collapse. But a tell the trench was the answer. He thought about Roland when he dug it. He thought about the specific quality of information Roland had given him. Not a blueprint, not a complete answer, but a direction. mass, thermal mass, heat that moves slowly. The rest Decker had built from that one idea, tested it in real cold, iterated when it failed, and ended up with something that worked. Roland was still at the drop-in center most Tuesdays. He was doing better himself that winter, had a room in a rooming house on French Street, had had it since December. He'd heard through the chain of conversations that Decker had built the block walls, and he nodded when Decker described it to him in more detail. He said, "That's it. That's exactly it." He said it like he'd been waiting for someone to figure it out.
The last thing to say about Decker Hail and the double walls on the east side of Erie is this. What he built did not solve homelessness. It didn't solve the absence of income that led to the absent deposit. It didn't solve the curfew rule that cost him the shelter bed in March.
It didn't solve the day labor economy where a bad week means 4 days of not eating enough. It didn't change the city's housing stock or the rental market or the gap between wages at the kind of jobs he could access and the cost of a room anywhere near those jobs.
He was still outside when April came, still making the 5:45 a.m. walk to the labor office most mornings, still eating at the drop-in center when there was nothing else. What the walls did was one thing, and only one thing. They kept the cold from reaching him on the nights when the cold was lethal. That's not everything, but it's not nothing. For 12 nights below zero, in a city that sits on the edge of a lake that generates its own weather, a 22-year-old with a white streak in his hair and duct taped elbows and mismatched boots woke up in the morning because he'd stacked blocks the way nobody had told him to stack them because an older man on a Tuesday told him to think about mass. Because he paid attention to what the cold showed him and closed the gaps it found. He woke up in the morning. That's what the double walls did. Go find a thermometer that goes below zero. Put it inside something thin. A tarp, a cardboard box, a tent on a night that hits - 10. Then put one inside the thickest, densest structure you can build from what exists around you. Read both temperatures at 5 a.m.
That's the test. That's the only test that matters. The walls don't do everything, but they do
Related Videos

Audi RS5 4.2 Tuning JDEngineering
JDEngineering
1K views•2013-11-14

DALI + KNX: 500 Lights Offline! BCU Code Lock & Short Address Fix!
EngineerIsmailTech
560 views•2026-04-13

Explaining Quality Control of Concrete
maherbader
4K views•2019-05-25

World Mining Production Peaks - Lead Antimony Arsenic Titanium & more
LucarioandDialga
1K views•2019-04-19

Tech Titans: LFP vs Sodium-Ion Battery | Which is the most effective energy storage solution?
Enfsolar
522 views•2025-11-06

Doing the Math: Analysis of Forces in a Truss Bridge
TeachEngineering
1K views•2025-06-06

Inside Midnight Fighter Jet Refuelling Secrets of Stealth Missions | WION Podcast
WION
3K views•2025-09-20

Flash Point, Fire Point & Auto Ignition Temperature
HSELessons
38K views•2019-08-28
Trending

One Must Imagine Sisyphus Happy
vlogbrothers
61K views•2026-07-21

Future of Taylor Farms
maighstirtarot5385
11K views•2026-07-21

The Downfall of OnePlus!
techwiser
65K views•2026-07-21

My Friend Locked Up The Engine On His K-Swapped Bug...
boostedboiz
128K views•2026-07-21