For a few years in the early 1990s, a lot of New York streets sparkled. Picture a Manhattan avenue at dusk, wet from a passing shower, the streetlights and headlights catching something in the road surface itself. Not the usual dull black, but a faint, restless glitter, as if the asphalt had been dusted with something. That shimmer was crushed bottles. The city had figured out how to grind its glass waste into the road.

The material had a name that told you exactly what it was: glassphalt. It went from a curiosity nobody trusted, to something people demanded outside their own buildings, and then it quietly disappeared. And it’s a small case study in why good recycling ideas so often don’t stick.

The glittering street, and why the city made it

Asphalt is mostly sand and gravel held together by a black binder. Glassphalt is what you get when crushed glass replaces part of that sand and gravel.

The Federal Highway Administration, in its guidelines on waste glass in pavement, says road-surface mixes with roughly 10 to 15 percent crushed glass perform fine. From 1989, the city’s transportation department started putting the glass content of its mix at about 10 percent.

The FHWA called the New York program the “most aggressive” glass-recycling paving effort in the country. From 1990 through 1995 the city used roughly 225,000 metric tons, about 250,000 tons, of glass in resurfacing. By 1991, crews had already laid glassphalt on nearly 900 lane-miles of Manhattan and Brooklyn streets.

What made the idea clever, as far as we can tell, is the kind of glass it used. Bottle recyclers want clean glass sorted by color. They won’t buy the mixed-color material that piles up at the bottom of a recycling stream. Glassphalt didn’t care about color. It could swallow exactly the low-value waste glass that nobody else wanted and turn it into road. The city reported saving about $500,000 a year by using crushed glass in place of some sand and gravel, and keeping 38,000 tons of glass out of landfills in the process.

From public suspicion to actual demand

People did not warm to it right away. A road that shimmers looks, at first glance, like something is wrong with it. But the turnaround, if the accounts from the time are right, was fairly complete.

Harold Watson, the transportation department’s assistant commissioner for asphalt operations, had become something of an in-house champion for the material. Watson said in 1991: “Now, when we do a road, if we don’t have glass on it, people come in and start complaining.”

That’s a nice story about public mood, not a measured fact, and it comes from a fan of the material, so we’d hold it loosely. Watson was clearly frustrated by a quieter resistance too. In his telling, “There’s a lot of people who can’t find anything wrong with glassphalt but won’t say anything good about it.” Read that as one supporter’s view rather than a neutral account, but it captures the odd spot the material was in: hard to fault, hard to fully endorse.

The Plaza, and a request for more sparkle

The clearest sign that the public had come around involved one very visible stretch of Fifth Avenue. After the city resurfaced the section in front of the Plaza Hotel with glassphalt, the hotel’s owner, Donald Trump, wanted the surrounding streets given the same treatment. A recycled-glass road surface had become something worth lobbying for outside a luxury hotel.

There’s something funny in that. The exact same crushed bottles a bottle plant would reject as worthless were, when spread across Fifth Avenue and lit up after rain, a selling point. The sparkle did real marketing work that the engineering never asked for.

Why the biggest glass-paving program in the country faded

None of the appeal saved it. The program’s central weakness was that its raw material came from curbside recycling collection, and collection is expensive.

During a budget crisis, Mayor David Dinkins floated suspending the recycling operation to save about $65 million a year, which would have cut off the glass supply. Commissioner Lucius Riccio said at the time: “That would be a shame. We’re doing a record amount of paving and we could use all the glass they could give us.” That tension is the whole story in miniature. The paving side wanted all the glass it could get, and the collection side that fed it was the first thing on the chopping block when money got tight.

What we take from a road paved with bottles

New York didn’t invent glassphalt. Baltimore had paved at least 17 streets with it from the mid-1970s through the mid-1980s, chasing the same sparkle.

What New York proved was that the idea could run at real scale, absorb hundreds of thousands of tons of the exact glass nobody else wanted, and win over a skeptical public along the way.

The sparkle was arguably the least interesting thing about it. It got the headlines, it got Trump lobbying, it made a wet avenue look faintly enchanted. But the real achievement was much plainer: a workable use for low-value mixed glass that would otherwise cost money to bury.

The reason it faded had nothing to do with whether the roads glittered, and everything to do with the fragile system behind it, the collection network and the raw cost of glass against dirt-cheap gravel. A recycling idea can be sound as engineering and still lose to logistics. That, more than the shine, is what a road made of bottles has to teach.

The post Beginning in 1990, New York City repaved streets with asphalt containing crushed recycled glass. By 1995, crews had used 250,000 tons of glass in resurfacing applications across the city, including a section of Fifth Avenue in front of the Plaza Hotel. appeared first on Space Daily.

One family in Port Washington, Wisconsin sold 173 acres for $10.2 million. On the county’s books, that same ground was valued at $590,000. That comes to roughly $59,000 an acre, about 17 times the assessed figure, according to a Wisconsin Watch analysis of Ozaukee County property records. 

It is tempting to read it as pure luck. Somebody’s farmland turned out to be worth twenty times what the tax roll said, the buyers overpaid, and a few landowners got rich. Parts of that are true. But the 17x gap is telling you something more specific, and the distance between what the assessor saw and what a data center developer would pay is the real story.

This article is for general information only, not financial, tax or real estate advice, and it is not a property appraisal or valuation. The prices discussed reflect particular transactions and should not be used to estimate what your own land is worth. Consult a qualified local appraiser or adviser before making property decisions.

What a 17x number is really measuring

A fair market value on a tax roll estimates what a buyer would pay for a parcel used the way it has always been used. In Port Washington, that meant farmland priced as farmland. It was never meant to capture what the same dirt is worth to a company assembling a site for a giant computing campus a mile inland from Lake Michigan.

The developer is Denver-based Vantage Data Centers, which broke ground on a 672-acre campus called Lighthouse on the city’s north side. It is the Midwest site of OpenAI and Oracle’s Stargate project. Vantage describes it as 2.5 million square feet of building, with completion targeted for 2028. A project that size does not want scattered plots. It wants one large, continuous block, locked down before a rival gets there.

That is why the assessed value and the sale price split so far apart. They answer different questions. One asks what the land is worth as land. The other asks what it is worth as the last missing piece of a $15 billion build.

The claim worth pushing back on

The idea that a price this high is just free money is where we’d push back. Not because nobody profited, plainly some people did, but because it treats the gap as proof the buyers were foolish or the sellers simply blessed.

The mayor’s version of the lucky-landowner story is memorable. Ted Neitzke, the project’s most prominent supporter, told Wisconsin Watch that the sellers “woke up one day and they just happened to live in the right spot.”

Right spot, yes. But “happened to” undersells what was being bought. Wisconsin Watch notes that, nationally, data center developers are often willing to overpay for land near electric power, to beat rivals in what has become a land rush. The premium is not a mistake. It is the price of speed and certainty in a market where losing one parcel can strand an entire campus.

Adam Mauro, a valuation director at Altus Group, put this plainly in a study of 80 data center land deals. The high price for well-placed land, he wrote, “is compensation for immediacy and certainty, not an indicator of superior underlying real estate.” That is one firm’s view rather than settled fact, but it fits Port Washington closely. The buyers were not paying for better dirt. They were paying to not lose.

Why single parcels commanded outsized prices

That $10.2 million deal did not happen alone. Wisconsin Watch found developers spent at least $125 million assembling 1,500 or more acres around the project, buying from individual owners at many times the assessed value. Members of the Karrels family and their farming business earned at least $20 million across their deals.

When you have to stitch together that much neighboring ground, every last holdout gains power. As Scott Finfer, a Texas land developer writing for HousingWire, put it, “data centers typically require hundreds of contiguous acres to deploy capital at scale.” If your farm sits in the middle of the block a developer needs, its value is not what farmland trades for. Its value is what the whole project is worth without it, which is nothing.

Altus Group’s numbers show the same dynamic from the other side. Its analysis found a roughly 120-fold spread in per-acre prices depending on parcel size: a median near $3.3 million an acre for sites under 25 acres, versus about $28,000 an acre for tracts over 1,000 acres. Small, well-placed pieces cost far more per acre precisely because the buyer cannot walk away from them.

What the sale price doesn’t capture

The land is the cheap part. A campus this size runs on water and power, and those carry costs the purchase prices never show. Vantage says it is funding a minimum of $175 million in regional infrastructure: water and wastewater capacity, mains, sewer lines, a new water tower and power hookups. Against numbers like that, a $125 million land bill looks more like a down payment than the main event.

The residents’ side is more mixed than either the windfall story or the horror story allows. Ryan Nowak sold his 65-acre Lake Drive property for $1.75 million and now lives about 10 miles north. “On paper it looked OK,” he told Wisconsin Watch, “until you go to replace what you had.”

Curtiss Smith, a crane operator who sold a four-acre property assessed at $258,000 for $895,000, described the strange feeling afterward: “After the fact, you hear what everybody else got.” Neither man says he was cheated. They are describing what it feels like to put a price on something that had never been priced before.

Port Washington is not unusual, which is part of why the multiples look the way they do. Land purchases for future US data centers reached about $6 billion in the first half of 2026, up 79% from a year earlier, according to figures from Avison Young reported by CNBC. Data centers now make up 27% of US development sites, second only to apartment buildings. A local land rush and a national one are the same thing at different scales.

What a fair-market figure can and can’t tell you

An assessed value is a good measure of what land is worth for the use it already has, and a poor measure of what it is worth to a buyer who needs that exact acre right now. The 17x figure does not mean the land was secretly premium. It is the price of speed, of buying a continuous block, of being the one parcel a $15 billion project cannot route around.

The post A Wisconsin family sold 173 acres at Port Washington for 10.2 million dollars, ground with a fair market value of 590,000. Developers spent at least 125 million dollars buying more than 1,500 acres around the project, and a 672-acre data center is now going up about a mile inland from Lake Michigan. appeared first on Space Daily.

When a pipeline company condemns a strip of your land and the pipeline is already going in, what are you actually fighting for? Not the land.

By the time the Terry family walked into federal court in Roanoke, that had been decided somewhere else, and it was never going to be reopened in front of a jury. All that was left on the table was money.

That distinction is the whole story of what happened to a 560-acre tract on Bent Mountain, in Roanoke County, Virginia, land the family had held for seven generations. It helps explain how a four-day trial ended with a $523,327 verdict, and why that number then moved twice before it stuck.

We’re not lawyers, and this article is for general information only—not legal advice. It summarises a particular dispute; eminent-domain rules and compensation depend on the jurisdiction and the facts of each case. If a pipeline or other project affects your property, consult a qualified lawyer in your jurisdiction.

What actually gets taken, and why the pipeline stays

An easement is not the same as buying land outright. Mountain Valley Pipeline took permanent and temporary easements totaling about 8.37 acres across the 560-acre tract, plus a wide strip to lay and maintain the line. The family still owns the ground. What they lost was the right to control that corridor, and, in the family’s telling, some of the value of everything around it.

The power to do this comes from the Natural Gas Act, which lets an approved interstate pipeline use federal eminent domain. An explainer from the University of Maryland’s Agriculture Law Education Initiative lays out the two things a landowner can contest: whether the taking is really for a public use, and how much they are owed. The Terrys had lost the first fight before this trial began. The pipeline already had its public-use certificate, so the corridor was staying no matter what a jury decided. Only the second question, the amount, was live.

That is the constitutional floor the whole dispute sits on. The Fifth Amendment protection against taking private property without just compensation, the Supreme Court held in Armstrong v. United States, “was designed to bar the Government from forcing some people alone to bear public burdens which, in all fairness and justice, should be borne by the public as a whole.” The land goes to a public purpose. The public, through the taker, pays what it’s worth. Deciding that last figure was the entire job of the court.

Why it came down to two appraisers

When the only question is money, an eminent domain trial tends to collapse into a contest between valuation experts. In this case, these were not slightly apart. They were looking at the same land and describing two different properties.

The family’s appraiser valued the whole tract at $1.9 million, resting that figure in part on the land’s potential use as a commercial wind farm. From there the family asked the jury for $570,000, roughly a 30% loss in value.

Mountain Valley’s appraiser, Joseph Thompson, put the property at about $1.2 million before the taking and figured the damage at a much smaller reduction. At trial, the company’s attorneys argued the Terrys were owed about $153,000.

Back in 2018, Mountain Valley’s original offer for the roughly eight-acre easement was about $119,000. The family refused, the company condemned the corridor and began cutting trees, and four years later a jury was asked to sort out the gap between the company’s figure and the family’s.

How a jury lands on $523,000

The verdict, $523,327, sat much closer to the family’s number than the company’s. Juries in these cases don’t have to pick one appraiser’s total and adopt it whole. They weigh what they find believable from each side and can settle on a figure neither party explicitly proposed. That flexibility is exactly what turned into the next fight.

After the verdict came in, the judge decided the jury had gone too far. U.S. District Judge Elizabeth Dillon set the $523,327 aside, finding it ran against the clear weight of the evidence. In her 2023 opinion, she concluded the award was more than twice the highest loss in value anyone had presented, and she entered a reduced judgment of $261,033 instead.

Why the final figure moved twice

The number the jury spoke was not, for a while, the number that counted. The family appealed, and in May 2024 the U.S. Court of Appeals for the Fourth Circuit reversed the district court and ordered the full $523,327 verdict reinstated. The panel held the award fell within the range of testimony the jury was allowed to believe, and that the property’s residential values alone could support it.

Three numbers, then, for one strip of ground: the pipeline comapany’s $153,000, the judge’s $261,033, and the jury’s  $523,327. The verdict survived, but only after a second round of litigation over whether the jurors had been allowed to reach it.

What strikes us, reading the case from start to finish, is how narrow the argument always was. Four days of trial, dueling appraisers, a $1.9 million valuation against a $1.2 million one, a wind farm that will never be built, and two more years of appeals, all of it about one thing: the size of a check.

The corridor was cut, the trees were down, and the pipeline was going where the pipeline was going. The only thing anyone could still win was a number, and the number was never going to buy the land back.

The post In 2018, a gas pipeline company took an easement across a Virginia family’s 560-acre property on Bent Mountain. After years of fighting over compensation, the family won a $523,000 jury award—but the company kept its right to use the land. appeared first on Space Daily.