What is the actual procedure through which this happens? You buy the land and then are granted permission on a discretionary basis? It seems to me that if you were a small business this becomes much harder to participate in because you need to acquire and hold the unproductive asset.
This would mean that land use tends towards that which large firms (which can sustain the costs easily by self-financing) find useful.
thewebguyd 1 days ago [-]
My employer went through a similar process, not for a data center but for a large recycling yard/center. We had to buy the land first, and it was basically unproductive for 2.5 years of mostly waiting for permits, and it was already zoned industrial so no zoning changes were needed.
The whole project was several million in expenses before even making a dollar. We aren't huge either, the permitting was not supposed to take that long it but a real strain on the business.
So yeah, you're correct. The current process favors large firms, at least those large enough to absorb the cost for multiple years or however long permitting takes, which in some municipalities can be a very, very long time.
nixgeek 1 days ago [-]
It’s unusual to buy the land and take a gamble on its utility, at least whenever datacenter construction is involved. Purchasing parties are risk averse to this exact scenario and work hard to craft contracts that reduce risk.
Often the choices are —
1. Buy land at $/acre that reflects very little premium, based on a short feasibility study, but without any ultimate contingency that permitting will occur. This is your example. But problematically all permitting applications are typically public record, so when you fail, the land can’t be sold on to someone else as if that didn’t happen, any sophisticated buyer will know the exact issues the city/county had with your usage. Land often transacts onward at firesale prices under these circumstances.
2. $/acre for land is bid upon at a substantial premium reflecting the future value as a datacenter, it remains under contract for potentially years pending outcome of approvals, then it transacts. Permitting being denied usually results in either no money changing hands or a small termination fee reflecting the carrying cost of the land during that period. If permitting works out the seller of land walks away very happy as the $/acre was extremely lucrative.
edmundsauto 1 days ago [-]
The second option also incentivizes the seller, who is often a local real estate magnate, to pressure local officials to issue the permits.
dv_dt 21 hours ago [-]
both options incentivizes the buyer to also lobby local officials
dnnddidiej 24 hours ago [-]
2 seems good as the seller can still make money from the land in the meantime. It does make it unsellable to anyone else in that time though.
voakbasda 3 hours ago [-]
I would not be willing to let the seller use the land under contract to avoid the risk of something happening that devalues or disqualifies it for its future intended purpose.
LtWorf 22 hours ago [-]
If you are microsoft, you know you will get the permit.
redanddead 22 hours ago [-]
No you don’t. The overall risk is baked into the strategy, but any individual plan is always a toss up, and they have insurance policies for if one or all fail
nixgeek 18 hours ago [-]
That’s not true, most hyperscalers and their datacenter building partners use Option 2.
cbdevidal 1 days ago [-]
Fun fact: Large businesses are often tapped to write the laws intended to target large businesses. The process is called “model legislation.” Fox and henhouse.
hylaride 1 days ago [-]
It's also called "regulatory capture" and it's been around in some form (implicitly or explicitly) since there's been laws in existence.
toss1 1 days ago [-]
Well, businesses — and all parties — who will be affected by legislation should be able to provide input. Otherwise we too often get clueless legislation that is massively mocked and rightly bemoaned on this site — because the legislators have no real clue of the technical issues involved.
Of course, the businesses should be only one part of the expertise that goes into writing the laws; other experts MUST be involved, or it will indeed be a fox and henhouse situation where the fox designs the legal locks so they can always be opened by foxes...
parineum 1 days ago [-]
That's not what model legislation is.
That can be an example of model legislation but, broadly, model legislation is created by an organization for use as an example for multiple different legislatures (usually states). Everyone from think tanks, busineses, the EFF, the ACLU and PETA draft model legislation.
Aurornis 23 hours ago [-]
I’ve had multiple friends try to start businesses that got stuck on this: You can’t really get permits until you have the property and start drawing up plans.
For simple businesses like a retail store in a location that has other retail it’s not too risky to bet that you’ll be approved, too.
For businesses with unique needs or that happen to be in the public crosshairs, you’re putting a lot at risk in the process.
The process favors big companies and developers who have established relationships and “connections” with the planning boards.
The situation is even wilder in some other countries, both more and less corrupt than the average US municipality. In some places you’re not getting a permit at all without a sizable bribe, or having an in with the planning board.
In my city one of the aspiring developers tried to run an expensive political campaign to get a family member into an office that could have helped with their approvals. People caught on and didn’t like it one iota.
nixgeek 18 hours ago [-]
It’s true it costs money to then hopefully make it. :)
Architecture, civil engineering, and other design and permitting fees can easily be 7–10% of the overall cost of smaller projects. Even in large projects they’re often 4-5% or more, and the number of billable hours for complex impervious surface and stormwater management adds up fast, as do engineering stamped plans for structural and other factors.
Most cost here is also incurred fairly early before you have any vertical construction done — Phase 2 in a program after land acquisition. So you feel like you’re spending a ton on paperwork and you can’t see anything yet.
You CAN spend a bit of money even before land acquisition on quick feasibility studies but in U.S. terms for something like a residential, small commercial or light industrial project every parcel you go “I like that, can it work?” you are dropping $15-50k during a 45-120 feasibility period. Should it 100% not work out you are NOT getting reimbursed that by the selling party. You’re out the money. Even within 90 days you may find some uncertainties like SEPA approval won’t close before you have to say deal or no deal on the parcel acquisition. This is quite unlike massive companies doing business where the land may not change hands until essentially every approval is locked in (but should it not work out the buyer may be out millions in engineering fees paid to try and make it work).
Borrowing to buy land and then borrowing more to build something is also treated very differently by most lenders. It carries tremendous uncertainty versus you buying a preexisting lot and structures which they know how to value. That’s fundamentally something that causes unwillingness to lend, or changes the rates and down payment or security terms. In contrast with a conforming mortgage for a SFR (single family residence) at 6%, borrowing to build (a construction loan) can be 10-14% APR, often secured via personal guarantee and other assets you possess, and then you have to convert to a personal or commercial mortgage after you complete building what you wanted.
Borrowing to buy the land is even more complicated again — you very often must be able to pay cash for land, and then just borrow to do the construction. Borrowing for both especially with limited assets to secure against will always be a polite “Sorry; we can’t help.”
bearjaws 23 hours ago [-]
Here in Orlando two large builders bought about 90% of downtowns empty lots (and several older buildings) promising new high rises in 2022, then threw in the towel after evicting all the tenants and now most of Church street station is completely empty. Of course local government had put nothing in the agreement about time frame of completion, what happens if the builder gives up, etc.
So now, all of it sits abandoned, no construction started, and now its not even worth building as they claim "down town is dead".
This is likely the fate of that land, a write off until its so valuable they can sell it to someone else.
toast0 1 days ago [-]
Depending on things, you might enter a land purchase (or lease) contract that's contingent on issuance of a building permit.
But a seller would probably prefer to sell without contingency, so what terms are available depends on market conditions.
Title insurance for residential real estate may sometimes cover properties that are unbuildable due to unsatisfiable permit requirements.
All told, it's easier as a buyer if you purchase an existing structure that was built under permits and is currently in use under appropriate occupancy permits.
beau_g 22 hours ago [-]
A comment on HN a few years ago in a thread about car washes brought this to light a little bit for me, the proliferation of subscription car washes everywhere in the US (self storage too) seems like it can't possibly make sense financially, but when you think about developers that also want to speculate on the land, car washes and self storage are about as easy as it gets to develop and maintain some cashflow, then you can sell or redevelop later. Now I am expecting a spinoff to Mister Inference and Quick Flops as these sprawling networks of carwashes turn into data centers in 2030
vel0city 22 hours ago [-]
I would imagine car washes to be one of the more difficult kinds of businesses to operate just to hold a lot. You often need a lot of work on drainage and permeability to ensure you're not contaminating local water systems with your car wash.
Storage though, yeah, that makes sense.
1 days ago [-]
teeray 19 hours ago [-]
It helps if the land is currently an eyesore and you’re the first to do something with it.
delusional 1 days ago [-]
From what I can tell Microsoft hasn't purchased the land yet. It's apparently owned by WE energies as part of the power plant next door.
burnte 21 hours ago [-]
SMaller business would find land already zoned for what they want. That makes it much less of a fight. These large companies find the land they want, then fight governments to allow the code and zoning changes to allow them to do what they want. It's a backwards process, but they think it's fine because they can afford it, and don't stop to realize that if they stopped letting execs make the decisions of where it goes, then a facility professional can find better land that they could get up and operational much faster.
I've seen it a decent number of times in my life. The exec gets their hart set on a specific building or parcel, but literally no one else in the entire project cares because they know it doesn't matter. Then the site desired won't work, and half the time the project fails.
I was at one company years ago, the execs were bound and determined the company was going to move the HQ from one building to another a handful of miles away. They saw the floors we were to rent, were completely set on that, and then proceeded to act like the deal was done. They had the credentialing department put in a change of address to MEDICARE before we had even signed the lease! We were 6 weeks from the move day when FINALLY there was a blow up in a meeting where people told the CEO and COO they were delusional and that we had to cancel the plan because we STILL DID NOT HAVE A SIGNED LEASE. There had been major negotiating hurdles between facilities and the building owner, but the C suite acted like none of it was happening for months. They spent 6 figures prepping for a move that in the end never happened. Building owner went bankrupt, we didn't move and instead just rented another suite in our own building.
thereisnospork 20 hours ago [-]
Pretty much. The discretionary nature of permitting, and other add-ons like CEQA, pose an enormous, and worse, unpredictable, burden on attempting to start a business in many places.
It'd be one thing if the requirements were merely onerous, but the discretionary nature adds corruption greatly favoring incumbents, the deep pocketed, and those willing to disregard the rules (start-ups with low capital requirements).
Never mind the timeline of these processes. Permitting can take 18-24 months, as can items like basic utility upgrades (adding 480V service, for instance, to an existing building can be an 18 month ~quarter million dollar endeavor.)
cucumber3732842 17 hours ago [-]
The permitting authority gets to pretend like it's not discretionary because the process has some "if we exercise out discretion to say no at this point then you sue us" thing to it that only megacorps with a legal team can afford to use
Dig1t 21 hours ago [-]
Yes it's called zoning, it's why we have a housing shortage in the USA. It makes building new things nearly impossible. It's also why none of the homes in Los Angeles that burnt down in massive fires a few years ago are being rebuilt, and likely won't be rebuilt for decades (if at all).
delfinom 1 days ago [-]
Zoning laws. Many parts of the US but not all have land use zoning. The zoning for any property you buy is public record, so any business knows well in advance of what they are buying. If you want to deviate from the zoning you have to submit an application for that zoning variance which requires usually a community hearing.
Neither small or large businesses really have any big advantages here. Got to win over the community. If anything, the small business may be local and the operators more readily able to convince the community for a variance than some corporate lawyer.
thewebguyd 1 days ago [-]
Zoning is only part of it. If a plot is already zoned industrial, but is empty, you still need to get the permitting for building construction, utility hookup, waste water & stormwater, environmental inspections, etc.
It varies from state to state (and city specific laws), but to go from empty land to productive asset can take several years.
cucumber3732842 17 hours ago [-]
Or better yet, farmer Johnson lets his D rate hay plot grow over because he never really thought it was worth it and the son that was managing that plot left for college.
Someone buys the plot 30yr later. They can't clear it and farm it without spending a quarter mil on environmental permitting because the government sees it as a pre-existing forest and the drainage ditch farmer Johnson's dad dug back in 1988 is now a stream (i.e. protected wetland) so they want the new owner to get the same permits that someone bulldozing a swamp for a strip mall would.
You see comparable fact patterns on every axis of regulation.
bobthepanda 1 days ago [-]
Also for a large enough utility hookup you will need to coordinate with the utility and or government since you can’t just plop down a large consumer on any old power line or pipe.
cucumber3732842 17 hours ago [-]
The community hearing is the easiest and cheapest part if your use is mundane.
The government will still screw you out of hundreds of thousands (mostly in the form of "pay these other people four figures for study X and plan Y" type requirements) to even get to that point though
3eb7988a1663 1 days ago [-]
Notably, this location is not far from where the Foxconn facility was going to be installed (the "eighth wonder of the world", 10k+ jobs, yada yada). After that debacle, I can imagine local residents are deeply skeptical of new big development projects.
pathartl 1 days ago [-]
It's about 15 miles north. Microsoft is already building a data center on some of the land that Foxconn didn't use
1 days ago [-]
delecti 1 days ago [-]
My first reaction is that 244 acres for a data center sounds absolutely obscene. But I have to admit that I'm coming from a place of ignorance.
How big "should" a data center be? How big are some other data centers? How big is us-east-1, for an example of a large one? I'm finding this to be rather difficult information to google.
manarth 1 days ago [-]
That's the land allocation rather than the building-size / data-centre size.
The average data centre is 10,000 square metres (2.5 acres).
As well as compute and network facilities, DCs also need to accommodate parking, personnel areas, cooling, fire-suppression, power substations, power redundancy (generators), ground-security…
244 acres is absolutely at the upper end of any DC site.
nixgeek 1 days ago [-]
Utah’s 40,000 acre datacenter proves it’s not absolutely at the upper end.
Most hyperscalers now prefer to build larger sites as “campuses” which may consist of many buildings each consuming 40-100MW, and then yes each building needs most of what you mentioned, so it adds up.
A few sites are now also contemplating BTM or ‘behind the meter’ power generation which takes additional space.
Then some sites like Microsoft’s Fairwater design are optimized for a very large number of Accelerator cabinets — think GPU, TPU, etc. Those cabinets are each consuming 140kW today and with a path to 700-1000kW cabinets soon, so that’s one super dense building instead of a campus of less dense buildings filled with Compute.
dgellow 24 hours ago [-]
> Utah’s 40,000 acre datacenter proves it’s not absolutely at the upper end
So far it seems to be more of a concept of a plan. I wouldn’t be surprised if they build smaller scale data centers first, then cancel the 40000 acres expansion. That sorts of feel like a marketing tactic. If not and they are serious, are we close to peak bubble?
cyberax 24 hours ago [-]
The density of modern racks makes me wonder why they would want so much space. There's just no way to power all of that.
Storage? Even that is now ultra-compact.
nixgeek 18 hours ago [-]
Amazon is probably deploying in the range of 250,000 racks per year into AWS. That’s millions of square feet before you get into all the infrastructure around them so they’re powered, cooled and operated the way they need.
Figure on ~10 million square feet of conditioned DC space per year, approximately 5-10GW of additional power consumption to power those 250k cabinets (depending on the exact mixture of what’s in the racks — Compute, Storage, Network, Accelerators), and that’s just for one hyperscaler.
There are at least 5-7 companies in the hyperscaler weight class although likely none individually meaningfully larger than AWS, they’re the 800lb bear and everyone else is in the 500-750lb range.
It’s a lot. Datacenters also take long enough to build that a hyperscaler is pouring concrete today for shells they expect to serve real workloads in 2029 - 2031. What you’re seeing come online today in response to customer demand really started being built in 2021 - 2023.
gottorf 21 hours ago [-]
They're called hyperscalers for a reason.
manarth 1 days ago [-]
40,000 acres, aka 77 × Monaco's!
TIL.
dnnddidiej 24 hours ago [-]
10 micro-Russias
400 vaticans
anvuong 17 hours ago [-]
10,000 square meters sound suspiciously small for a datacenter, even more so if you have to account for supporting facility? Maybe a small one? it's just 100m by 100m, which is smaller than most Walmart Supercenter.
ninalanyon 20 hours ago [-]
How much parking does a data centre need, and why is it not placed underneath the building, underground? Why do cars have to take up so much land?
nixgeek 18 hours ago [-]
Building on grade is much cheaper. There is in general plenty of surface area on Planet Earth.
Datacenters aren’t built next to Nordstrom. Theres just no reason to spend on engineering and construction that increases density like underground parking.
dogcomplex 21 hours ago [-]
It also buffers for all the surrounding properties which would otherwise complain about noise.
jubilanti 1 days ago [-]
I assume you mean AWS us-east-1. It isn't a single data center. It is a cluster of data centers around Northern Virginia.
LeFantome 1 days ago [-]
us-east-1 is a region. That means that it is 3 to 6 “availability zones” within a 100 km or so. Each of these availability zones consists of a cluster of data centers. Each cluster is perhaps 3-5 that are a few km from each other. The data centers will have tens of thousands of servers each.
So that is the mental model you should have for “how big is us-east-1”. But also, the data centers are not going to be, individually, anything like 244 acres. Best guess is that individual data centers are between 200,000 and 400,000 square feet. That is 5 to 10 acres.
Do the math above and us-east-1 may be 300 acres of floor space spread over a very large area.
nixgeek 1 days ago [-]
AWS publicly stated I think in 2021 that the larger availability zones in US East 1 consisted of 17-18 datacenters each. It’s likely grown a lot since, and they recently announced AZ7 will be online in Maryland soon, so they must be running out of ability to grow the ones in NoVA.
I can’t find a link now but it was one of the re:Invent talks like Peter DeSantis briefly explaining AZs before he dug into how Amazon optimizes their concrete mixtures to be more environmentally friendly or something…
All things point to that being the biggest region any hyperscaler has in the world, and several gigawatts of power consumption.
James Hamilton also gave a talk in 2021 about AWS having crossed 20 million Nitro cards deployed and 12GW power consumed —
If so, us-east-1 may be over 1000 acres of floor space. For comparison, Disney World is 600 acres and Central Park is about 850.
tptacek 1 days ago [-]
Caledonia is an exurb of Milwaukee, so it's pretty sparse and spread out. There isn't that much demand for land on the outskirts of Milwaukee and most of the demand out there is industrial. Compared to the other industrial uses you'd get, data centers are almost certainly preferable.
CPLX 23 hours ago [-]
> Compared to the other industrial uses you'd get, data centers are almost certainly preferable.
Why? The ratio of jobs and taxes to local resources tied up has to be one of the worst possible trade-offs of any industrial use that you can envision. That is precisely why data centers are proving to be profoundly unpopular.
Compare and contrast to something like a Boeing assembly plant with thousands of high-paid skilled jobs, and knock-on effects with local service providers and OEM vendors.
gottorf 21 hours ago [-]
> The ratio of jobs and taxes to local resources tied up has to be one of the worst possible trade-offs of any industrial use that you can envision
Why? There aren't much "local resources tied up" since it's land that isn't that valuable for the most part. DCs, once built, don't emit fumes or loud noises, or make traffic worse like other industrial land use might. They just sit there quietly and generate tax revenue for the locality.
Sure, there aren't many ongoing jobs at an operating DC. But I can't see why it's one of the worst possible industrial uses, from the point of view of the residents around one.
> data centers are proving to be profoundly unpopular
From what I can tell, this is just from a combination of AI being unpopular plus opportunistic fearmongering on the part of politicians.
tptacek 21 hours ago [-]
The bigger thing here is just the idea that the alternatives are "Boeing assembly plant" or "data center". Like: if you have the option of getting the Boeing plant, get the Boeing plant! Racine County does not have that option. Meanwhile, the happy sounding businesses there --- like the duck farm --- are actually major polluters, and more problematic for the region than a data center.
The people opposing data centers oppose them everywhere, including in the middle of the desert in Utah, which really gives away the game.
dehrmann 18 hours ago [-]
> he bigger thing here is just the idea that the alternatives are "Boeing assembly plant" or "data center".
This is a false dichotomy, though. That region has enough land that you could do both. There just isn't much demand for manufacturing in the Upper Midwest right now.
CPLX 17 hours ago [-]
What you're saying is just factually incorrect. Data centers absolutely are loud and they're disruptive.
They're not good for the environment. They consume large quantities of water, typically, and huge, absolutely staggering amounts of energy, which typically has the effect of raising rates for everybody else by driving up demand and causing capital infrastructure needs that are financed by everybody, not just the data center.
There is absolutely widespread dislike for data center construction in basically every region spanning almost every political axis in the United States right now.
There's polling on it. It's abysmal. The assumption that all these people are stupid sheep probably won't be a productive way to approach this public policy question.
gottorf 16 hours ago [-]
> The assumption that all these people are stupid sheep probably won't be a productive way to approach this public policy question
I agree, and I do not assume that these people are stupid sheep. However, having said that:
> Data centers absolutely are loud
Compared to undeveloped land, sure. Compared to other uses of industrially-zoned land? Most likely not. I've been to a lot of datacenters. They're not very loud from the outside.
> They consume large quantities of water
I honestly do not understand why they don't just build closed-loop systems, or geothermal where they have a closed loop with the ground acting as the heat sink. Open-loop systems where they consume water just to evaporate sounds stupid to me. But I don't know how common that is.
Politicians showing off in front of the camera with a dirty jug of water, of course, is just grandstanding and has no real relation to datacenters other than they made construction happen.
> huge, absolutely staggering amounts of energy, which typically has the effect of raising rates for everybody else
This isn't just true of datacenters. They're building an aluminum plant near me that will consume on the order of 100+ MW, which is comparable to a fairly large DC. Any other large industrial plant would consume a lot of electricity, but nobody's blaming them for utility capital investment needs.
tptacek 17 hours ago [-]
They do not in fact consume large quantities of water, and "they consume lots of water" is a pretty good shibboleth for "this isn't going to be a productive discussion". Golf courses in the US consume more than 8x the water of all data centers combined.
Obviously, that's my response to your polling claims as well: people getting polled thing data centers are consuming all the water.
juliusdavies 15 hours ago [-]
It would make more sense to complain about plants consuming sunlight than it does to complain about data centers consuming water.
CPLX 7 hours ago [-]
You guys really don't get it, do you?
You've got these people going on TV and doing interviews saying that this new technology will be capable of replacing huge amounts of jobs.
The people that say that most jobs will be eliminated by AI are making direct threats of violence against the families of millions of people. Telling someone that they will no longer have any way of feeding their children is telling them that you're going to kill their family.
I'm not saying this to convince you that that's true. You could make some counter-arguments to that, and I'm sure people will come along and do that. The point is to understand it that that's how it's perceived by the people hearing it.
And then to add insult to injury, they want to come do it in your town.
They want to pitch a giant "development" project. But remember nobody ever wants “development” except for the fact that it might bring jobs and economic growth, to them, not you living somewhere else.
Meanwhile, you've already told them that the reason we need so many more data centers all of a sudden is specifically so that we can eliminate whatever jobs are left that we didn't get a chance to eliminate over the last several decades when we exported most of the jobs we could to different countries.
By the way we're the same assholes who said "trust us" last time, and pitched the idea that it would lead to a prosperous future for everyone. Instead you made money off sending the jobs to china, capturing all excess value with software, and addicting our children to opioids.
And your next exciting offer for us is a giant windowless building that will take up open land, nature, and yes, water and electricity, and employ almost nobody.
People don't feel like they control a whole lot, but they do have some ability to complain about this thing happening where they live.
The actions of people opposing these data centers are completely rational. If I'm trying to make one point, that's it.
The actually interesting part of this back and forth on HN is that Silicon Valley (aka "tech") culture has grown so fundamentally rotten at its core that not only do people not have values that place humans first, they can't even recognize when other people do.
phil21 4 hours ago [-]
> But remember nobody ever wants “development” except for the fact that it might bring jobs and economic growth, to them, not you living somewhere else.
This is really the only point that matters. It's held true my whole life, and I expect it will be true until I die. Doesn't matter what you build. I was watching a couple transmission line projects over the past decade that are still in the "lawsuits and community rage" phase - proposed to bring wind farm watts to load centers - and now they will likely be killed since the anti-development folks can pretend they were always due to Datacenters.
Datacenters are just an easy scapegoat for the anti-development crowd. It's been amazing to watch how quickly it's gone, and how folks have such a strong opinion on stuff that otherwise would have been built half a mile from them and they'd never have known any better until they were told to care.
The rest of what you wrote is largely social media driven ragebait in comment form. Kernels of truth, but largely immaterial.
Datacenter land use is the least interesting thing you could possibly discuss. Knocking out some corn fields and building some warehouses off the road no one can see or hear, with almost no traffic to/from them after construction is pretty much the lowest possible bar for local community impact for quite literally any project. It change nothing for anyone, other than the farmers who sold the land and that a few local trades companies have a couple decades of stable highly paid employment.
Therefore, the only way to get communities up in arms about these things is basically lie about it.
It's going to be a real head scratcher to folks when electricity rates continue to march upwards even if they get all AI datacenter construction banned. The green tech crowd who had the datacenter bogeyman land in their lap is playing an exceedingly dangerous game here. What we are largely seeing is the bill coming due for generational lack of investment into the grid.
If your local community can't figure out how to get the money raining down from the skies like it is now to subsidize the build out of your local infrastructure for something as minor as municipal water treatment in Wisconsin You likely will never be building anything at all.
The people who want these data centers are telling us that the reason that they want them is to keep momentum on their plans to destroy our entire way of life.
They just say it. They go on TV and do interviews and say it out loud.
So then everyone who has any ability to stop it in any possible way tries to do that.
What exactly the fuck do you think they're going to do?
And more importantly, what causes you to feel that it's obvious that they are only coming to these opinions through ignorance?
Lumping together anti-data-center sentiment with anti-development sentiment in general is bullshit. Yes, there are certainly impossible to negotiate with NIMBYs who don't like apartment buildings because they cast shadows on a sidewalk corner or something. That has nothing to do with this.
The politics of data centers are completely different. If this was a real, actual job-creating industrial project, people would react differently.
phil21 2 hours ago [-]
> Lumping together anti-data-center sentiment with anti-development sentiment in general is bullshit. Yes, there are certainly impossible to negotiate with NIMBYs who don't like apartment buildings because they cast shadows on a sidewalk corner or something. That has nothing to do with this.
I mean, you just did it yourself. You made a great point by doing so.
It's generally the same people. Your rants pretty much prove it. Plus I've been in meetings where it's literally the same people. They will use any and all reasons to stop local development and then stick with the one that gets the most popular traction.
I'm not talking about folks against residential development. I'm talking being around projects and in local meetings about industrial development - primarily electric generation and transmission. The arguments are pretty much the same.
> The people who want these data centers are telling us that the reason that they want them is to keep momentum on their plans to destroy our entire way of life.
AI is now taking rural blue collar jobs? I find it very difficult to believe this is a real grass roots concern.
I find it very easy to believe white collar folks are using their relative positions of power to amp up concerns rural blue collar folks would actually care about. Often at the expense of said blue collar folks.
These facilities would often be a win for a local community with a little bit of foresight. It doesn't matter to your power bill if they are sited 5 miles down the county road from you, or 200 miles. Chances are they are using electricity from your regional interconnect and that's where your power bills come from.
> And more importantly, what causes you to feel that it's obvious that they are only coming to these opinions through ignorance?
Only? Of course not. Primarily? Obviously. For the simple fact that dozens of people in my orbit who never knew datacenters existed near them all of a sudden Care Very Much(tm) about the subject after watching a few very low information videos. These are folks who drove past local facilities most of their lives and never had a clue.
> If this was a real, actual job-creating industrial project, people would react differently.
Not in my experience. Pretty much every single industrial project is nearly impossible to build in the US. Heavy industry sounds fun until someone wants to build an aluminum smelter, copper mine, or wind farm down the way. There are tons of infrastructure projects that need to get done which will not because they are not point-source locations that employ lots of jobs for a single community.
My previous example are wind farms. Typically those are best sited a few hundred miles away from a major metropolitan/load center. Good luck getting anything of scale built these days now that we've more or less burned up every ounce of spare electric transmission capacity leftover from before we de-industrialized ourselves. Once you get out of re-using existing right-of-way you see nearly the same backlash as AI datacenters. The difference? The urban laptop classes don't take your side and the outrage tends to stay localized.
It's very interesting to me a certain class of folks have convinced people that the closest thing to "free money" for a local community is a bad thing. Construction phase might suck, but assuming it's simply a datacenter and not a power plant with a co-located datacenter off the to side, you really can't get any lighter-touch land use than this. It's probably less environmentally impactful (in a negative way) to the local community than the 100 acre corn field it replaces. And employs more people to boot.
I would 100% agree with you on any tax abatements/credits/etc. These facilities do not need them and would be quite happy to pay full-freight on taxes on top of contributing towards upgrading local infrastructure far beyond their expected impact on it. This is where I feel that local politicians have seriously shit the bed all over their communities.
condis 20 minutes ago [-]
> And employs more people to boot.
More Indians maybe.
CPLX 29 minutes ago [-]
The fact that opposition to one thing looks sort of vaguely like opposition to another thing isn't completely irrelevant information, but you're just missing the actual thing that's happening if you focus on it.
"Seven in 10 Americans oppose constructing data centers for artificial intelligence in their local area, including nearly half, 48%, who are strongly opposed. Barely a quarter favor these projects, with 7% strongly in favor."
This is another way of me countering your main point here.
It's not the same people you've seen in previous arguments.
It's those people plus almost all of the other people because overwhelming 70%+ majorities don't want these things.
So we know that. That's the given here, assuming you think there's at least some validity to Gallup and polling.
The question then becomes why? My rants are intended to illustrate why. They sound like "rants" because people are extremely fucking angry about how things are going, so accurately restating their opinions also sounds angry.
The culture has changed. People do not fucking trust tech companies and their leadership at all for extremely valid reasons. Talk about wind farms all you want. I can go ahead and talk about the Yankees. The conversation at hand is about data centers, and data centers are their own issue.
When people like you come along and say it's "free money" nobody fucking believes you because there is a rich and long history of that kind of bullshit already.
There's a point of view here that they oppose data centers out of ignorance. My countervailing point is that their distrust is almost certainly rational, considered, and correct.
Distrust is the only sane response to recent events.
tptacek 5 hours ago [-]
You could just say "I concede the point on water but stand by the rest of what I wrote". Instead, all this.
CPLX 4 hours ago [-]
My original post that started this discussion did not mention water:
> The ratio of jobs and taxes to local resources tied up has to be one of the worst possible trade-offs of any industrial use that you can envision. That is precisely why data centers are proving to be profoundly unpopular.
Nonetheless I can see that fixating on this instead of my obviously correct point which is that people absolutely hate data centers and the sociopathic tech billionaires behind them for rational reasons is a good dodge.
Hardly unexpected when I am posting on the promotional discussion board of a private equity firm responsible for launching many of these sociopaths into society.
My later comment about water usage is both unneeded for that point to stand and trivially proven true.
tptacek 36 minutes ago [-]
It can't just be that we disagree about this issue; no, it must be the "private equity firm's" fault.
CPLX 23 minutes ago [-]
This is a story about Microsoft building data centers and people opposing it.
So yes, one of the two key players here is the heavily financed tech sector.
Starting to think these disingenuous replies are in fact the point here.
Since none of of these conversations on HN make much difference anyways, my only hope is that one person, somewhere, who is a part of the "tech community" that usually gathers here (and of which I am a longtime member) becomes newly aware that people fucking hate us now for excellent and extremely well-supported reasons and thinks a little about how that happened and how they might change it.
tptacek 21 minutes ago [-]
I simply disagree with you. I'm not a fan of populism, and, more importantly, I think that people have had very good reason to "fucking hate us" for decades --- we are in the business of automating away people's jobs, and have been since the 1960s.
It's fine that you think the discussion is unproductive; I agree, our premises are too far apart to get anywhere. But you'd get further with, well, everybody that doesn't already agree with you if you'd stop accusing anybody who doesn't agree with you of being "disingenuous".
CPLX 20 minutes ago [-]
Well at least you're honest then. I hope your team loses because the outcomes here suck and are destroying our culture.
And on a personal level, I hope you reconsider your life decisions a little and realize the importance of a culture where people can feed their families and lead healthy productive lives of dignity, regardless of who their parents are or how gifted they happen to be intellectually.
mapt 1 days ago [-]
Based on a majority of games regions, US-East-1 is scattered properties in a <100 square mile area near Dulles Airport in Virginia, associated with an Internet backbone junction and former AOL campus in small town called Ashburn.
jeffbee 1 days ago [-]
Almost all of the site would have been open space, existing transmission corridors, an electric substation, and two flood control ponds they threw in to try to sweeten the deal by offsetting the new impermeable surfaces. The data halls are a small portion of the site.
dqv 19 hours ago [-]
Stormwater ponds are the bare minimum, not a deal sweetener. A deal sweetener would be an ecological management plan which demonstrates a true desire to limit the environmental impacts of this development as much as possible. And no, a storm pond company that just pours chemicals into the water to control algae and plants invasive privets around does not demonstrate that.
You're saying the data center has a small footprint, but it does not. You don't need that much stormwater pond if you don't have giant impervious surfaces.
Deal sweetener is footprint (buildings, parking lots, and substation) being 1/6 of the land with the other 5/6 dedicated to nature preservation.
Not all wide open space is created equal. If it's wide open space with Eurasian grasses that get cut every few weeks, it's useless and has 0 benefit to the local ecology. Even if they weren't cut, they're still do very little for the ecology in the area.
If those forests don't have keystone tree species, then it's the same.
The people running these companies are so incompetent, they can't even do greenwashng right.
jeffbee 15 hours ago [-]
I think you're a little miscalibrated on "giant impervious surface". The 244-acre site in question would have had three data halls, a substation, and one assumes parking lots and aprons. Here's a 300-acre Walmart distribution center in the same region. It has about 100 acres of concrete. I think that's a lot more than the data center would have had.
That's not how stormwater management works. Land elevation, adjacent land, the profile of the land (how much groundwater is there?) on which the structure is built (and probably even more that I'm not considering) all affect how big of a retention/detention pond is needed. And these requirements change over time when it is discovered that older sites' management techniques failed to adequately manage stormwater. The Walmart site was built at least two decades ago, no? That's enough time for stormwater management guidance and policy to change.
Even then, it is improper to assume anything about the stormwater management needs of one site based on another unrelated site. But even then, Walmart's site seems to be completely surrounded by stormwater management. Even the northwest corner of the site is a detention basin. [0]
The most obvious and logical conclusion to be made here is that an engineer told Microsoft they needed to have stormwater management of that size, so that's what they put in their plans. No sweetness, just lawsuit prevention.
If they want to be sweet, they should be building huge nature preserves (and they have enough money to afford it) into these plans instead of trying to be greedy by building the largest possible structures (they think) they can get away with.
Well my IP (regular plain residential Asian ISP) is blocked on this site. Zealous Cloudflare-blocking is breaking the web.
(also thanks for the useful message telling me to "contact the website owner... while blocking me from the website where the contact info should be)
dylan604 1 days ago [-]
I'm not defending Cloudflare, but what is a better solution? If small websites can just be DDOS'd out of existence because some group thinks it'll be funny, what protections do they have? It takes too much equipment and know-how to stop an attack for people to be able to survive online. The next thing you'll hear about is a monthly service fee to the hackers as a protection racket just like the mob.
somenameforme 1 days ago [-]
The site's also blocked for me, also on a normal residential IP. Making your site inaccessible for people out of fear that somebody might make it inaccessible for people feels reminiscent of blockading the strait because you don't want the strait blockaded.
dantillberg 1 days ago [-]
> feels reminiscent of blockading the strait because you don't want the strait blockaded
I think this is a poor analogy, unnecessarily politicizing the topic.
It might be a good analogy the other way around, if hackers DDOSed the website as revenge for partial IP-based blocking, in order to apply pressure to the website operator to remove IP-based blocking. But that wasn't the topic.
pocksuppet 23 hours ago [-]
It's accurate. The USA is doing "Iran can't blockade the strait if I blockade the strait!" and Cloudflare is doing "Hackers can't take your site offline if I take your site offline!"
Alive-in-2025 1 days ago [-]
occasionally a major site that I subscribe to blocks me with cloud flare. It was either nyt or a similar news site I subscribe to. I couldn't even get to any 'give me feedback' page because cloudflare was blocking. When cloudflare decides to block you, it should give you a contact page for that website so you can send them an email or tell them.
I work around this by using my phone connection with phone chrome.
expedition32 1 days ago [-]
The internet is killing itself.
And no I do not blame small website owners they just have to live with this mess same as everyone else.
Dylan16807 1 days ago [-]
The better solution to blocking entire continents is probably doing nothing.
For DDoS resistance... Well I can imagine a world where a tech in the same area as IPFS or freenet gives backup access to websites that are overloaded.
sethops1 1 days ago [-]
> tech in the same area as IPFS or freenet
Are we getting that before or after personal jet packs, flying cars, and my tacos delivered via tacocopters?
I'll protect my sites with Cloudflare until then, thanks.
Dylan16807 19 hours ago [-]
Go ahead, just don't set up enormous IP blacklists on cloudflare please
warkdarrior 1 days ago [-]
> For DDoS resistance... Well I can imagine a world where a tech in the same area as IPFS or freenet gives backup access to websites that are overloaded.
As a small website owner, I can use Cloudflare or I can wait for this imagined tech.
lstodd 23 hours ago [-]
You can use CF and lose relevance. Or even go beyond that and outright geoip-block .
Ofc it's your choice.
For some reason there are many small sites I have no problem visiting and then there are those CF users which may or may not work at any given moment, forcing me to ignore them.
Well, good luck. You are cutting yourself from the internet, not cutting me off.
Gander5739 21 hours ago [-]
You can use CF and lose, say, 10% of your audience (number pulled from thin air). Or you can not use CF, be taken down by scraping bots or a dos attack or whatever and lose 100% of your audience.
exe34 1 days ago [-]
Would IPFS need to be a part of the browser? Or is there an easy to use browser out there that runs on IPFS? If you need the average user to go find proxys, it won't work.
pocksuppet 21 hours ago [-]
Just host the site. Are you expecting to get DDoS'd? Why? And what will be the consequences if you are?
Someone said it's like protecting yourself from stabbings by banning kitchen knives, but it's more like protecting yourself from stabbings by wearing thick rubber armor at all times, that also happens to be filled with spy cameras somebody else owns. It's a bit of an overreaction to a very rare threat, don't you think?
sandeepkd 1 days ago [-]
Makes me wonder if the company protecting against the DDOS would have motivation to encourage or facilitate the DDOS efforts too, makes them the protection racket itself.
If DDOS is really the problem we want to solve then it would be awesome if one can do it without looking into the packet. SSL terminating at some centralized third party provider is way too much power.
deely3 1 days ago [-]
And don't forget about kids safety!
bshaughn 1 days ago [-]
Websites should have a lean markdown or .txt page for each human friendly webpage. A lot of the surge in bots is because of LLMs. Its insane that a technical documentation web page can use 200MB + of memory, when the core information I care about is << 1 MB of text. at the path of least resistance for many people is to have claude code hit 20 of such pages.
This is something that would be perfect for cloudflare to host and sell as a service - static web pages via their CDN network.
I do not work in web development, so im sure there are plenty of details im ignorant of, but the TLDR of "how to fight accidental DDOS because of AI tooling " is make it easier for them to get the content they want.
dylan604 21 hours ago [-]
Problem is there's no way to ensure that the bots will only suck in the bot friendly file instead of scraping the whole site. The bot owner would never be able to guarantee that the site doesn't have data not in the bot friendly file. Any time you have to maintain two sets of files is pretty much guaranteed that one will be out of sync at some point. Not all sites are driven by a database that generates files on the fly.
PunchyHamster 1 days ago [-]
nah they will be selling their service to both hackers and their targets
sph 1 days ago [-]
How many small websites served by Cloudflare risk being DDOS'd? How many small website owners would incur serious loss of livelihood if they are DDOS'd for a few days?
Is DDOS risk so important that the web needs a protection racket?
> If small websites can just be DDOS'd out of existence
DDOS doesn't destroy websites. It just makes them unreachable until the disgruntled person decides it's been running long enough.
Please stop exaggerating a very real problem only a few entities on the web have; what you are perpetuating is FUD, which enables companies like Cloudflare to kill the web.
> The next thing you'll hear about is a monthly service fee to the hackers as a protection racket
How do you not even see the irony of this?
ivanmontillam 1 days ago [-]
> DDOS doesn't destroy websites. It just makes them unreachable until the disgruntled person decides it's been running long enough.
You can be absolutely destroyed if your hosting provider later hits you as a Website Owner with an excess traffic bill.
pocksuppet 23 hours ago [-]
Fire your provider then. They're probably not paying for inbound traffic (most orgs are billed on the dominant traffic direction, so inbound for eyeballs and outbound for hosting), so it's pure extortion on their part.
lstodd 23 hours ago [-]
While it is entirely possible to enter such a contract with a provider, it is your fault in the end. Don't maybe enter them?
dylan604 1 days ago [-]
> Please stop exaggerating a very real problem only a few entities on the web have; what you are perpetuating is FUD, which enables companies like Cloudflare to kill the web.
I'm not exaggerating, I'm just playing what if. That's a game where you think of random things that could go wrong, and then deciding if it is worth the expense. Just because maybe you can't think of things of varying plausibility does not make me exaggerating. We already see ransomware working from the hacker's perspective. There's no reason to think that greed will not come into play. If I can think of it, there's no reason to think that hackers are not also considering various ways to expand on ransomware as a service
>> The next thing you'll hear about is a monthly service fee to the hackers as a protection racket
> How do you not even see the irony of this?
How do you not? If every hacking group can come along and extort any site they choose to pay them a protection fee, there's no way websites will accept any of this. Compare that to paying a single legit service protecting against all of those hacking groups. Can't imagine why people would be willing to do that.
therein 1 days ago [-]
This is the same mindset that wants to make it illegal to sell kitchen knives with sharp tips.
dylan604 1 days ago [-]
How do you come to that conclusion? It's so far off of what I said that you've got some splainin' to do
theflyinghorse 1 days ago [-]
I don't understand why cloudflare is loved by tech people. They ARE breaking the web.
throwaway2037 21 hours ago [-]
Real question: What is the alternative to prevent DDoS? The killer app is that their DDoS prevention is free for small sites.
toast0 20 hours ago [-]
Most sites can probably make do with ddos mitigation from their hosting provider.
Larger hosting providers may have enough bandwidth to accept and drop most inbound volumetric DDoS. Smaller providers often contract with a DDoS mitigation service; during an attack the mitigation service BGP announces the attacked range, filters out problematic traffic and forwards the rest.
It's been at least 7 years since I ran servers that routinely saw DDoS... at that time, we didn't tend to attract attacks anywhere near 10Gbps, so having 10G connections for severs likely to be targetted did most of the work. I'm sure it's worse now, so 10G might not cover easy attacks but 100G servers aren't hard to find. Attacks do get larger than 100G, but most sites won't be hit with a large attack.
For a small site, taking an outage during a DDoS is often an acceptable alternative too.
isodev 1 days ago [-]
It's amazing how after all these years, tech world still believes it's a good idea to sell the entrance to their websites and services to the same bouncer. Cloudflare not in the mood for your IP/ISP/Country? Tough cookies.
llm_nerd 1 days ago [-]
In this case a site is geoblocking. It's an affiliate station is Wisconsin, serving up local news. Sites have been using things like MaxMind and blanket blocking entire regions long before Cloudflare.
If you've run a site like that, pretty soon you realize 100% of the traffic that hits you from Asia, Russia, the Middle East and even Eastern Europe / the Baltics is exploit detection scripts and is just noise in your logs. Okay, 99.999999% as once every decade something ends up on HN and gets a broader audience.
isodev 1 days ago [-]
This is such a narrow PoV and I don't see how it relates to Cloudflare currently being de factor gatekeeper for every web properly of consequence.
What's the point of publishing news or content on the web if you don't want it to be accessible? What about locals who travel? What about locals who share links with others?
llm_nerd 1 days ago [-]
You brought up Cloudflare, when in reality this site went into their caching services configuration and purposefully enabled geofencing restrictions. They could have done this a million ways, and Cloudflare is basically irrelevant to this conversation, and is utterly fungible.
>What's the point of publishing news or content on the web if you don't want it to be accessible
Because you don't want the burden of far away users who will never represent a penny of income for your content? This is a weirdly entitled comment.
Quite aside from certain countries disproportionately account for malicious traffic, often there are legal issues that come into play as well. This is why many regional sites block EU locations because they don't want the compliance costs for users that aren't their base.
Mickelby 1 days ago [-]
If you run a website then you quickly learn that 99% of traffic is bots/spam/scam/ai, and will overwhelm your resources. You can block it with the click of a (Cloudflare) button. There is nothing to consider here.
not_a_bot_4sho 1 days ago [-]
> (also thanks for the useful message telling me to "contact the website owner... while blocking me from the website where the contact info should be)
I hit this a lot with Firefox VPN and it's ridiculous
sammy2255 1 days ago [-]
You could use a search engine to find their contact information
kogasa240p 1 days ago [-]
Microsoft has decided not to move forward with its proposed site for a data center in the Village of Caledonia after facing significant community pushback from residents.
Posted
and last updated
VILLAGE OF CALEDONIA, Wis. — Microsoft has decided not to move forward with its proposed site for a data center in the Village of Caledonia after facing significant community pushback from residents.
PREVIOUS COVERAGE | Microsoft data center proposal continues to divide Caledonia residents as rezoning plans move forward
“Based on the community feedback we heard, we have chosen not to move forward with this site,” a Microsoft spokesperson said in a statement Wednesday.
The tech giant’s decision comes after hundreds of residents voiced opposition to the project over recent weeks. More than 2,000 people signed a petition opposing a rezoning proposal that would have allowed the data center to be built on 244 acres of land.
Watch: Microsoft pulls plug on plans for 244-acre data center in Caledonia after community pushback
The proposed site was situated on County Line Road and State Highway 32, southwest of the WE Energies Oak Creek Power Plant, and was surrounded by farmland and residential properties.
47032805-Concept Site Plan - Project Nova by TMJ4 News
Despite abandoning this particular location, Microsoft indicated it remains interested in investing in Southeast Wisconsin.
The spokesperson said the company looks forward to “working with the Village of Caledonia and Racine County leaders to identify a site that aligns with community priorities and our long-term development goals.”
TMJ4’s Jenna Rae, who has been following this story, reached out to Todd Willis, the village administrator, who provided the following statement:
“Nothing official has been submitted to the Village regarding their pending application, and have no comment until such time.”
- Todd Willis, Village Administrator
Resident Prescott Balch told TMJ4 that his phone did not stop ringing on Wednesday morning, as people delivered the news.
PRESCOTT BALCH
TMJ4
Prescott Balch lives in Caledonia. Balch welcomed the news that Microsoft is changing plans to bring a data center in the area.
"We're ecstatic that those arguments held water and ultimately convinced a large corporation to back off, so great day here in Caledonia," Balch said.
Village trustee Nancy Pierce says she learned about the change from a news article.
"I have a lot of respect for Microsoft, making the decision when they say they listened to the constituents. They also listened to board questions both at the planning commission at the board level. I believe that they took a lot of different pieces of information into play," Pierce stated.
Nancy Pierce
TMJ4 News
Nancy Pierce is a village trustee in Caledonia.
Both Pierce and Balch made it clear that they are not opposed to working with Microsoft in Caledonia.
As the tech giant looks for a new site, there is hope that there are improvements to the overall process.
"I would’ve liked to been able to engage directly with Microsoft much earlier in the process. We were not allowed to do that. I think that became an obstacle for a lot of different points and reasons," Pierce explained. "I feel like now they would come forward much quicker and engage directly with the community, really get to understand the community."
"There are people that have an opinion about what they want to do with their village, and that was absent in this to me. That's the real message of this thing," Balch explained. "Let's help Microsoft find the right spot in southeast Wisconsin."
PunchyHamster 1 days ago [-]
block was enabled by site owner, not cloudflare. If they used something else they'd just blanked block ASN IP blocks
zinekeller 23 hours ago [-]
And as stated by others, it's about not complying with GDPR et al. Heck, this is not even Cloudflare, it's AWS Cloudfront!
Symbiote 22 hours ago [-]
The site works fine from the EU.
1 days ago [-]
LogicFailsMe 4 hours ago [-]
At what point was this datacenter going to benefit the residents beyond a trifling number of jobs in exchange for draining local resources for pennies on the dollar Microsoft should have paid? If the tech companies can't make these things into genuine win/win scenarios, then don't let them build anything, they're currently as useless to local communities as stadiums except you don't even get to watch games in them.
Absolutely 100% force them to make a genuine case for them and to sign contracts demanding they live up to their obligations. Otherwise, it's more evidence why billionaires need to write and sign some reality checks on their dragon piles.
dmix 1 days ago [-]
I'd say they should come build them in Canada instead since our energy is cheaper, but anti-industrial development policy and NIMBYism is even more embedded than America, which is probably why no one is bothering.
SecretDreams 23 hours ago [-]
I think it's less this and more they're encouraged to build in America over Canada.
Danox 1 days ago [-]
Probably a wise decision on their part Microsoft already is all in on Copilot AI if it fails, the CEO probably is gone.
simonebrunozzi 1 days ago [-]
Are you sure? AFAIK, Satya did a great job from the point of view of a shareholder - +1,200% from 2014 to today [0]
That was then. Today the stock is trading at a lower price than in May 2024. Google is up over 100% in that time.
shimman 22 hours ago [-]
Not that impressive when MSFT has a monopoly in several verticals that allows them to subsidize their many failures.
thewebguyd 1 days ago [-]
They're already backtracking on, at least, the consumer Copilot being shoved everywhere in Windows. I think Nadella is already on his way out though (voluntarily). Judson Athoff is already taking over as "CEO of Commercial Businesses" so Nadella can be closer to engineering, which he's said he misses.
ralph84 23 hours ago [-]
One of the emails that came out in the Musk v. OpenAI trial was Nadella in 2022 saying he wanted to own the entire AI stack. Four years later Microsoft is no closer to owning any part of the AI stack. Complete fumble of the bag.
rbanffy 1 days ago [-]
For a moment I thought they were referring to the Scottish Highlands, but I guess the name fell in disuse when the Roman Empire fell...
simianwords 1 days ago [-]
I wonder if USA would have futuristic Datacenter Towns much like coal mining towns - they might have their own lore, vibe and aesthetic about them.
Certain new emerging towns would get the moniker of "DC towns". New economies might flourish - perhaps not jobs from DC but certainly the tax money should help.
This could happen if the NIMBY movement weren't so extreme.
saulpw 1 days ago [-]
If it actually helped the economies flourish, then that would be interesting. NIMBY is in some part a reaction to the belief (likely true) that there would be neither jobs nor tax money from a data center in their community backyard--it's all downside with no upside.
toast0 20 hours ago [-]
Post construction, data centers don't have a lot of employment, so you don't get much income tax or in-community spending from employees.
But, companies put a lot of capital equipment in them, and capital equipment accrues property tax, so there should be taxes paid from that.
simianwords 1 days ago [-]
The Loudoun county has Datacenters that pay enough in taxes equalling the salary of 30k residents earning $40k dollars each (in a county of ~400k population).
This is a shockingly high upside relative to any other industry, like Car or Steel industry which pollute way more.
These taxes can be used to create better infra and have other things going on. Better schools and maybe even research facilities.
But I do think the opposition is largely ideological in nature so these arguments don't matter at the end.
AngryData 23 hours ago [-]
They could be used for infastructure, but when power and/or water bills go up in responce to local data centers being built it is hard to believe because power and water is infastructure and should have costs going down.
JuniperMesos 24 hours ago [-]
Taxes have relatively little to do with how good schools are, that is dominated by the quality of the students who live close enough to attend the schools.
y1n0 19 hours ago [-]
I can't believe you blame students for bad schools.
JuniperMesos 9 hours ago [-]
A school is fundamentally a social environment - children are legally compelled to be there, and the specific educational structure is centered around being forced to be in classrooms with large numbers of similarly-aged kids, for large portions of the day. Much of what one learns in school is social acculturation to the specific segment of society represented by the other kids there, rather than the contents of classroom learning.
How good a school is - which we can expand into more detailed definitions like "how much do children learn there?", "how much does any individual kid like or hate it?", "is this a good use of a child's time?" - is dominated by what the other kids they are in a shared social setting are like, and not by how expensive the facilities are or how well the teachers are paid.
simianwords 12 hours ago [-]
They are somewhat right but it’s irrelevant. Good students can be brought in because there’s enough tax money.
dgellow 24 hours ago [-]
But a datacenter doesn’t employ that many people, compared to mines. To develop a town you need to give a bunch of people a job, build small commerces, invest in the local infrastructure, so a real local economy can sustain itself. There is none of that with datacenters
1shooner 1 days ago [-]
Do you see coal mining towns as a model to pursue?
simianwords 1 days ago [-]
Yes they contributed enormous growth without which USA could not have been a dominant power. There are better alternatives now however.
23 hours ago [-]
ChrisArchitect 1 days ago [-]
(2025) OP.
1 days ago [-]
bigdick1 1 days ago [-]
[dead]
sterlind 1 days ago [-]
it's not the size of the chip, it's the motion of the digitalocean.
jeffbee 1 days ago [-]
Wealthy white exclave succeeds in using environmental justice language to keep cheap coal-fired power to themselves. Very American outcome.
Although I obviously don't care about Microsoft's outcome here, this was clearly a great site at the intersection of two transmission lines and with essentially infinite water resources.
Some of us would like to keep our “infinite” water resources which actually aren’t infinite.
I live beneath two transmission lines (overlapping, I guess, but not intersecting) and would prefer no data centre built here. Why? Because it will provide me no benefit whatsoever, reduce my property value, and worsen my quality of life due to things like light pollution and noise.
If data centre operators would fix these things perhaps people would feel differently. For example - provide multi gigabit fibre Internet to everyone nearby.
thewebguyd 1 days ago [-]
> provide multi gigabit fibre Internet to everyone nearby.
Kind of a cool idea, actually. These data centers could turn the towns where they build into startup incubators. Offer free high speed internet and heavily subsidized compute to residents in exchange for building there. At least gives back economically somewhat, as a data center itself doesn't provide much in return.
culi 1 days ago [-]
Gigabit internet doesn't power startups. It powers consumer streaming. As long as you can run Zoom, you don't need high internet speeds for startups
trollbridge 1 days ago [-]
You pretty much need low latency 1G+ to work from home these days.
nixgeek 18 hours ago [-]
I work from Starlink regularly and it’s just fine for meetings and everything else I do.
Meetings are usually under 10Mbps of downstream and 2Mbps upstream.
It’s simply false you need 1Gbps of “low latency” internet to work from home effectively unless your job is something like content production and working with huge files.
Producing for OnlyFans? Benefit from 1G to send raw to your editor.
Doing most other forms of work? Doesn’t require remotely close to this amount of bandwidth.
weakfish 23 hours ago [-]
... do you? I have 500mpbs and both my partner and I WFH with no problems.
culi 21 hours ago [-]
For group calls you just need 1.0 Mbps. If you want 1080p group calls you need 3.8 Mbps. Zoom itself says for a "reliable WFH setup" you should have "at least 25 Mbps download and 5 Mbps upload"
I've gotten by on McDonald's wifi
r_lee 1 days ago [-]
how exactly is high speed fiber + subsidized compute a recipe for making startups?
RajT88 22 hours ago [-]
I'm seeing a lot of pushback against data centers in my town, building where there is currently cornfields.
I would disagree that it provides no benefit whatsoever. I ran the numbers on the proposed campuses, and at half price for industrial property taxes, it's at full build-out going to bring a billion dollars a year to the city. The property taxes I pay, which 73% of go to the local school district is about 108 million per year.
Unless the city royally screws this up, my property taxes are going to go down. Only a tiny number of residents will live near the data centers. But, the opposition is massive locally here, and I've been trying to understand why. Talking to people in various groups, there seems to be more going on with the opposition than simple NIMBYism (although that's part of it).
Some of the cited reasons are understandable - concerns about electricity. The state passed a bill to provide some relief, and there should be no water impacts locally here.
> Some of us would like to keep our “infinite” water resources which actually aren’t infinite.
We're talking Lake Michigan, which is where the local data centers by me will be sourcing their water from (our city and surrounding cities are switching from deep aquifer wells to lake water brought in via very long pipe). Forget the data centers (that's a drop in the bucket compared to the cities' usage), such pipes usually have a capacity of 50-100 million gallons per day (MGD). That'd be able to drain Lake Michigan in about 35000 years, assuming no rainfall. Forget draining, how many such pipes do you have to build for it to start draining water faster than it's flowing into the lake? At 100MGD, 300 or so.
So yeah, it's not infinite. Enough of these big pipes built out across multiple states could have an effect. IL is apparently limited to ~2000 MGD, and any other state bordering a lake (any great lake) has limits/usage far, far lower than that (Michigan is apparently 5MGD?). It doesn't appear to be too much of a concern. None of what's allowed would add up to taking out more lake water than is flowing into the lake. As long as the current limits are kept as-is.
jeffbee 22 hours ago [-]
Given the predicted rise of Lake Michigan, and the predicted future volatility of its level, I would suggest that the cities and states of the region be building as many megascale waterworks as they can, and the energy systems to operate them. If they can get that money from data center builders, all the better.
jeffbee 1 days ago [-]
I support in principle the rights of towns to set their own land use rules, but on the larger societal picture I don't support people benefiting from things like intermodal shipping, goods distribution, and information services that they refuse to host. So I perceive a certain hypocrisy in this story.
3eb7988a1663 1 days ago [-]
Surely I benefit from a host of things for which I want nowhere near me. Strip mining, petroleum refining, chemical processing, coal fired electricity, etc.
Am I allowed any autonomy or must we all accept that if a rich group wants to plop down a leather tanning factory across the street, I should have no recourse?
jeffbee 1 days ago [-]
That's a mix of different issues. The site of a natural resource isn't one of the things that political systems control, whereas the site of a petrochemical refinery or a power station is chosen by those systems. So yes, it is obviously hypocrisy to consume petrochemical products while insisting that the refinery can't be in your "rural character" exclave with the arbitrary line drawn around it, but allowing the same facility to be built over the county line in the poorer, browner community that you consider sacrificial. Anyway, the impacts of a data center are not in the same ballpark as the other things you mentioned.
Scroll_Swe 1 days ago [-]
I kind of agree but I live in a city in Sweden.
Should I not be able to use youtube or order online because we don't have a DC right next door?
KptMarchewa 10 hours ago [-]
Sweden has a bunch of data centers. Doubt any of them consume coal based electricity though.
Scroll_Swe 5 hours ago [-]
Yup exactly :)
It is all about how you plan, build etc how good you are.
mindslight 1 days ago [-]
Maybe everyone in this village already has their own local AI rig. From a technical perspective, data centers aren't providing public goods - rather they're more like attractive nuisances that foster centralized control.
thewebguyd 1 days ago [-]
Mentioned earlier in the thread, but maybe these data centers should start providing public goods to the towns/counties in which they build? Free high speed internet, heavily subsidized compute, maybe partner with local colleges to offer labs & internships, etc.
There's no reason they can't be economic accelerators for the towns they are in.
mindslight 1 days ago [-]
I agree it would help their image, but this all sounds like more corporate centralization to me. I've got a functioning electric coop that provides gigabit internet for a flat fee every month. I think they're even doing trial runs of 10Gb, but I haven't looked into it because I simply don't need it. Internet access can be a solved problem these days, where there is the political will. And if anything local colleges should be offering compute resources to the larger community, not themselves relying on scraps of generosity from commercial buildouts.
trollbridge 1 days ago [-]
With a new data centre, your electric coop rates can go up with no benefit to you whatsoever.
jeffbee 1 days ago [-]
In America electric rates are increasing due to 20 years of stagnation and failure to invest in transmission, and new data centers are on average associated with lower, not higher retail electric rates. This is because large consumers drive down marginal prices.
In your country things may be proceeding differently, but that's the story here.
mindslight 22 hours ago [-]
This doesn't really hold when we're talking about discrete pieces of infrastructure that may need to be built. If the current transmission infrastructure is fine for the current users, and then a new large user comes along that necessitates new infrastructure, that is a huge new cost that simply that simply didn't exist without that large user.
I believe in many areas this is the kind of infrastructure investment that regulated utilities can easily pass onto existing rate payers, which is where the problem/narrative comes from in the first place. So your theory doesn't match the actual results people are complaining about.
Same thing when datacenters take power generation into their own hands and build new gas turbines rather than solar/batteries or paying to expedite grid construction. And I'm willing to believe these problems happen in a minority of cases, but the problem is that they seemingly do happen and existing residents are left without recourse.
jeffbee 21 hours ago [-]
Your model of the infrastructure as a perpetually sustaining asset is not correct. These assets must be continually refreshed and improved or their value disappears. The emergence of the data center as a point load has merely coincided with the existing grid infrastructure falling to pieces. The places with the highest marginal energy prices such as California are the places with the fewest data centers. Large consumers invest in transmission assets that benefit everybody.
mindslight 18 hours ago [-]
You would seem to be saying that grid infrastructure needs continual refreshing and improvement to continue carrying the same amount of power? That strikes me as a tall claim, given what I already know about electricity (electronic design and house wiring). I am open to entertaining it, but you've got to provide a reference or at least just describe the mechanism.
This is talking about the capital assets themselves, which are of course separate from regular maintenance like tree trimming.
About the only thing I can think of is something like a bulk of the grid infrastructure was built out 30/40/50/whatever years ago, and is now all reaching its expected lifetime all at once, and perhaps some types of equipment are strictly retired at the end of their lifespan lest they fail catastrophically.
jeffbee 17 hours ago [-]
Yes, exactly. It's a bunch of metal that stands out in the weather all the time.
mindslight 14 hours ago [-]
To be convinced, I'd need to see actual data showing that the costs for replacing existing assets are what's actually spiking.
22 hours ago [-]
parineum 1 days ago [-]
This water usage meme needs to die. Although, it is nice to have an indicator for people who believe whatever they told without trivial verification.
energy123 1 days ago [-]
One of the rare cases where nimbys can't do damage because the hyper scalers will (and are) building their data centers across MENA, South Asia and SEA where they're welcomed with generous tax breaks and incentives.
Sending kilobytes of text over thousands of miles is a lot easier than piping energy or housing across distance!
expedition32 1 days ago [-]
Welcomed by corrupt politicians perhaps the locals not so much.
Data centers do not provide jobs and they are run by sociopath Americans who couldn't give a shit about human rights or the environment.
866-RON-0-FEZ 1 days ago [-]
> The data center would have been built in this scene
You mean the carefully cropped photo of pristine rolling farmland in the article is in reality next door to a coal-fired power plant? Say it ain't so.
insane_dreamer 1 days ago [-]
"a great site" -- you frame it like Microsoft was working for the public good
nixgeek 1 days ago [-]
Building a datacenter typically employs thousands of people in the trades, often for hundreds or a thousand plus hours, per person, from the start of a big build to the campus being complete. It’s quite literally millions of billable hours in trade labor.
Modern datacenters also require very high standards of construction and are complex, so these projects create jobs and also represent a real training, upskilling and work experience opportunity for labor. There are many examples of electricians, plumbers and groundwork teams who did Microsoft’s site getting future work from Meta, Google or Amazon in the same part of the state because the experience has value.
It’s easy to dismissively say datacenter is bad, or that it consumes too much water (despite many datacenters accused of this being a closed-loop cooling system), and ignore the billions of dollars spent during the project on labor which supports that local economy, or the improvements negotiated for the local area and paid for the hyperscaler, bundled in by the city/county planning as part of the permits and approvals.
It’s also rare the tax for a campus is fully rebated, although it’s normal for the improvements to be partially rebated for some period (this is an investment incentive). Viewed over 20-40 years these sites are often tremendously lucrative in tax for the county/city as well.
SoftTalker 1 days ago [-]
There are a lot of jobs during construction.
There are very few jobs during operation. Mostly site security and a few tech support staff. There will be some steady work for maintenance contractors, but that's much less than the initial construction.
nixgeek 1 days ago [-]
I mean, you’re right, but when the alternative is all that investment and those construction jobs AND the post-commissioning operational jobs go to a different community and a different economy, …
What would you prefer? To me, local communities tend to benefit in multiple different ways during and after these projects, poorer communities become richer, communities with little opportunity now have more opportunity. I’m always a bit baffled by someone saying “Please don’t invest $5B and create 100s of jobs and taxable improvements in my back yard”.
This is a common argument: wanting 1000s of jobs during construction and 1000s of jobs after construction, but this isn’t a car manufacturing plant. That’s a “we want our cake and we want to eat it too” argument — not saying it’s your argument just that this comes up frequently.
SoftTalker 1 days ago [-]
But what often happens is that the company is granted a ten year tax abatement in exchange for the jobs, the jobs end up being fewer than promised, and then in ten years the company closes the site and now the community has an empty industrial brownfield that was built for one thing and can't be easily repurposed.
nixgeek 18 hours ago [-]
Datacenters rarely get mothballed after 10 years, they’re usually 15-year depreciation schedules in terms of the building and the plant. Equipment within them like servers typically depreciates over 72 months but is rarely removed by hyperscalers promptly after 6 years. More typically it’s around Year 8.
So you won’t see the building cease being useful for about 20 years. You’re getting usually two full cycles of “the servers inside” before a renovation program (or potentially asset/building disposal to another party, or demolition, depending what changes in those 20 years really).
What’s often happening around these tax agreements is they are a mixture of incentive and an offset of prepaid improvement costs to the city/county for developing mains water, sewer, roads, schools, fire and police, and other infrastructure, sufficient to support the 100s of families who may move here to take the post-commissioning jobs, etc.
Often money out the door for the hyperscaler is about the same over 20 years, it’s just some number of $M’s is paid upfront as an “Contribution to Improvements”. That’s actually good for the municipality involved, too.
Municipalities typically know or are told these aren’t tax rebate for 10 years and then it’s getting bulldozed. They’re sophisticated enough and well advised to understand this is a 20+ year investment in their town.
1 days ago [-]
jeffbee 1 days ago [-]
It makes more sense when you realize that the same people also don't want anyone new living in their town, so they do not perceive any value to new jobs, only costs.
insane_dreamer 12 hours ago [-]
temporary jobs during construction that don't benefit the town in the long run
the tax rebates can extend for decades
> these sites are often tremendously lucrative in tax for the county/city as well
these datacenters haven't been around long enough to know that; these are _not_ your grandma's datacenters -- they are much more resource-hungry since they do substantially more processing
jeffbee 4 hours ago [-]
In a "tax break" (misnomer) situation the developer is paying either a lump sum up front, or a defined annual tax that commences immediately. You don't have to wait to find out. An example of this is the Google Kansas City deal.
Danox 1 days ago [-]
This is the last stand for Satya Narayana Nadella Copilot has to work…
Scroll_Swe 1 days ago [-]
[flagged]
HDThoreaun 1 days ago [-]
Water is a renewable resource. It actually is infinite if youre just using it to cool a datacenter, it does not disappear when youre done with it.
AngryData 22 hours ago [-]
It does if that water is being pumped from underground resevoirs that take 100s of years or more to refill then heated up and evaporated away to rain down upon the ocean or run down the Mississippi to the ocean.
I don't think people would care so much if they were using lake or stream water that got sent back, but often they don't because it isn't clean water which means more money spent on filtration and/or cooling system maintence.
HDThoreaun 22 hours ago [-]
This is a good point and I agree it is important to place systems that use a lot of water close to renewable water resources. That being said, the vast majority of non renewable water usage goes to wasteful desert agriculture and there's comparatively very little backlash toward that. It just seems really dishonest to me to complain about datacenter water usage while ignoring the much larger source of the problem.
pessimizer 1 days ago [-]
[flagged]
Rendered at 18:17:08 GMT+0000 (Coordinated Universal Time) with Vercel.
This would mean that land use tends towards that which large firms (which can sustain the costs easily by self-financing) find useful.
The whole project was several million in expenses before even making a dollar. We aren't huge either, the permitting was not supposed to take that long it but a real strain on the business.
So yeah, you're correct. The current process favors large firms, at least those large enough to absorb the cost for multiple years or however long permitting takes, which in some municipalities can be a very, very long time.
Often the choices are —
1. Buy land at $/acre that reflects very little premium, based on a short feasibility study, but without any ultimate contingency that permitting will occur. This is your example. But problematically all permitting applications are typically public record, so when you fail, the land can’t be sold on to someone else as if that didn’t happen, any sophisticated buyer will know the exact issues the city/county had with your usage. Land often transacts onward at firesale prices under these circumstances.
2. $/acre for land is bid upon at a substantial premium reflecting the future value as a datacenter, it remains under contract for potentially years pending outcome of approvals, then it transacts. Permitting being denied usually results in either no money changing hands or a small termination fee reflecting the carrying cost of the land during that period. If permitting works out the seller of land walks away very happy as the $/acre was extremely lucrative.
Of course, the businesses should be only one part of the expertise that goes into writing the laws; other experts MUST be involved, or it will indeed be a fox and henhouse situation where the fox designs the legal locks so they can always be opened by foxes...
That can be an example of model legislation but, broadly, model legislation is created by an organization for use as an example for multiple different legislatures (usually states). Everyone from think tanks, busineses, the EFF, the ACLU and PETA draft model legislation.
For simple businesses like a retail store in a location that has other retail it’s not too risky to bet that you’ll be approved, too.
For businesses with unique needs or that happen to be in the public crosshairs, you’re putting a lot at risk in the process.
The process favors big companies and developers who have established relationships and “connections” with the planning boards.
The situation is even wilder in some other countries, both more and less corrupt than the average US municipality. In some places you’re not getting a permit at all without a sizable bribe, or having an in with the planning board.
In my city one of the aspiring developers tried to run an expensive political campaign to get a family member into an office that could have helped with their approvals. People caught on and didn’t like it one iota.
Architecture, civil engineering, and other design and permitting fees can easily be 7–10% of the overall cost of smaller projects. Even in large projects they’re often 4-5% or more, and the number of billable hours for complex impervious surface and stormwater management adds up fast, as do engineering stamped plans for structural and other factors.
Most cost here is also incurred fairly early before you have any vertical construction done — Phase 2 in a program after land acquisition. So you feel like you’re spending a ton on paperwork and you can’t see anything yet.
You CAN spend a bit of money even before land acquisition on quick feasibility studies but in U.S. terms for something like a residential, small commercial or light industrial project every parcel you go “I like that, can it work?” you are dropping $15-50k during a 45-120 feasibility period. Should it 100% not work out you are NOT getting reimbursed that by the selling party. You’re out the money. Even within 90 days you may find some uncertainties like SEPA approval won’t close before you have to say deal or no deal on the parcel acquisition. This is quite unlike massive companies doing business where the land may not change hands until essentially every approval is locked in (but should it not work out the buyer may be out millions in engineering fees paid to try and make it work).
Borrowing to buy land and then borrowing more to build something is also treated very differently by most lenders. It carries tremendous uncertainty versus you buying a preexisting lot and structures which they know how to value. That’s fundamentally something that causes unwillingness to lend, or changes the rates and down payment or security terms. In contrast with a conforming mortgage for a SFR (single family residence) at 6%, borrowing to build (a construction loan) can be 10-14% APR, often secured via personal guarantee and other assets you possess, and then you have to convert to a personal or commercial mortgage after you complete building what you wanted.
Borrowing to buy the land is even more complicated again — you very often must be able to pay cash for land, and then just borrow to do the construction. Borrowing for both especially with limited assets to secure against will always be a polite “Sorry; we can’t help.”
So now, all of it sits abandoned, no construction started, and now its not even worth building as they claim "down town is dead".
This is likely the fate of that land, a write off until its so valuable they can sell it to someone else.
But a seller would probably prefer to sell without contingency, so what terms are available depends on market conditions.
Title insurance for residential real estate may sometimes cover properties that are unbuildable due to unsatisfiable permit requirements.
All told, it's easier as a buyer if you purchase an existing structure that was built under permits and is currently in use under appropriate occupancy permits.
Storage though, yeah, that makes sense.
I've seen it a decent number of times in my life. The exec gets their hart set on a specific building or parcel, but literally no one else in the entire project cares because they know it doesn't matter. Then the site desired won't work, and half the time the project fails.
I was at one company years ago, the execs were bound and determined the company was going to move the HQ from one building to another a handful of miles away. They saw the floors we were to rent, were completely set on that, and then proceeded to act like the deal was done. They had the credentialing department put in a change of address to MEDICARE before we had even signed the lease! We were 6 weeks from the move day when FINALLY there was a blow up in a meeting where people told the CEO and COO they were delusional and that we had to cancel the plan because we STILL DID NOT HAVE A SIGNED LEASE. There had been major negotiating hurdles between facilities and the building owner, but the C suite acted like none of it was happening for months. They spent 6 figures prepping for a move that in the end never happened. Building owner went bankrupt, we didn't move and instead just rented another suite in our own building.
It'd be one thing if the requirements were merely onerous, but the discretionary nature adds corruption greatly favoring incumbents, the deep pocketed, and those willing to disregard the rules (start-ups with low capital requirements).
Never mind the timeline of these processes. Permitting can take 18-24 months, as can items like basic utility upgrades (adding 480V service, for instance, to an existing building can be an 18 month ~quarter million dollar endeavor.)
Neither small or large businesses really have any big advantages here. Got to win over the community. If anything, the small business may be local and the operators more readily able to convince the community for a variance than some corporate lawyer.
It varies from state to state (and city specific laws), but to go from empty land to productive asset can take several years.
Someone buys the plot 30yr later. They can't clear it and farm it without spending a quarter mil on environmental permitting because the government sees it as a pre-existing forest and the drainage ditch farmer Johnson's dad dug back in 1988 is now a stream (i.e. protected wetland) so they want the new owner to get the same permits that someone bulldozing a swamp for a strip mall would.
You see comparable fact patterns on every axis of regulation.
The government will still screw you out of hundreds of thousands (mostly in the form of "pay these other people four figures for study X and plan Y" type requirements) to even get to that point though
How big "should" a data center be? How big are some other data centers? How big is us-east-1, for an example of a large one? I'm finding this to be rather difficult information to google.
The average data centre is 10,000 square metres (2.5 acres).
As well as compute and network facilities, DCs also need to accommodate parking, personnel areas, cooling, fire-suppression, power substations, power redundancy (generators), ground-security…
244 acres is absolutely at the upper end of any DC site.
https://www.theverge.com/ai-artificial-intelligence/933687/u...
Most hyperscalers now prefer to build larger sites as “campuses” which may consist of many buildings each consuming 40-100MW, and then yes each building needs most of what you mentioned, so it adds up.
A few sites are now also contemplating BTM or ‘behind the meter’ power generation which takes additional space.
Then some sites like Microsoft’s Fairwater design are optimized for a very large number of Accelerator cabinets — think GPU, TPU, etc. Those cabinets are each consuming 140kW today and with a path to 700-1000kW cabinets soon, so that’s one super dense building instead of a campus of less dense buildings filled with Compute.
So far it seems to be more of a concept of a plan. I wouldn’t be surprised if they build smaller scale data centers first, then cancel the 40000 acres expansion. That sorts of feel like a marketing tactic. If not and they are serious, are we close to peak bubble?
Storage? Even that is now ultra-compact.
Figure on ~10 million square feet of conditioned DC space per year, approximately 5-10GW of additional power consumption to power those 250k cabinets (depending on the exact mixture of what’s in the racks — Compute, Storage, Network, Accelerators), and that’s just for one hyperscaler.
There are at least 5-7 companies in the hyperscaler weight class although likely none individually meaningfully larger than AWS, they’re the 800lb bear and everyone else is in the 500-750lb range.
It’s a lot. Datacenters also take long enough to build that a hyperscaler is pouring concrete today for shells they expect to serve real workloads in 2029 - 2031. What you’re seeing come online today in response to customer demand really started being built in 2021 - 2023.
TIL.
400 vaticans
Datacenters aren’t built next to Nordstrom. Theres just no reason to spend on engineering and construction that increases density like underground parking.
So that is the mental model you should have for “how big is us-east-1”. But also, the data centers are not going to be, individually, anything like 244 acres. Best guess is that individual data centers are between 200,000 and 400,000 square feet. That is 5 to 10 acres.
Do the math above and us-east-1 may be 300 acres of floor space spread over a very large area.
I can’t find a link now but it was one of the re:Invent talks like Peter DeSantis briefly explaining AZs before he dug into how Amazon optimizes their concrete mixtures to be more environmentally friendly or something…
All things point to that being the biggest region any hyperscaler has in the world, and several gigawatts of power consumption.
James Hamilton also gave a talk in 2021 about AWS having crossed 20 million Nitro cards deployed and 12GW power consumed —
https://mvdirona.com/jrh/talksandpapers/JamesHamilton2022101...
Why? The ratio of jobs and taxes to local resources tied up has to be one of the worst possible trade-offs of any industrial use that you can envision. That is precisely why data centers are proving to be profoundly unpopular.
Compare and contrast to something like a Boeing assembly plant with thousands of high-paid skilled jobs, and knock-on effects with local service providers and OEM vendors.
Why? There aren't much "local resources tied up" since it's land that isn't that valuable for the most part. DCs, once built, don't emit fumes or loud noises, or make traffic worse like other industrial land use might. They just sit there quietly and generate tax revenue for the locality.
Sure, there aren't many ongoing jobs at an operating DC. But I can't see why it's one of the worst possible industrial uses, from the point of view of the residents around one.
> data centers are proving to be profoundly unpopular
From what I can tell, this is just from a combination of AI being unpopular plus opportunistic fearmongering on the part of politicians.
The people opposing data centers oppose them everywhere, including in the middle of the desert in Utah, which really gives away the game.
This is a false dichotomy, though. That region has enough land that you could do both. There just isn't much demand for manufacturing in the Upper Midwest right now.
They're not good for the environment. They consume large quantities of water, typically, and huge, absolutely staggering amounts of energy, which typically has the effect of raising rates for everybody else by driving up demand and causing capital infrastructure needs that are financed by everybody, not just the data center.
There is absolutely widespread dislike for data center construction in basically every region spanning almost every political axis in the United States right now.
There's polling on it. It's abysmal. The assumption that all these people are stupid sheep probably won't be a productive way to approach this public policy question.
I agree, and I do not assume that these people are stupid sheep. However, having said that:
> Data centers absolutely are loud
Compared to undeveloped land, sure. Compared to other uses of industrially-zoned land? Most likely not. I've been to a lot of datacenters. They're not very loud from the outside.
> They consume large quantities of water
I honestly do not understand why they don't just build closed-loop systems, or geothermal where they have a closed loop with the ground acting as the heat sink. Open-loop systems where they consume water just to evaporate sounds stupid to me. But I don't know how common that is.
Politicians showing off in front of the camera with a dirty jug of water, of course, is just grandstanding and has no real relation to datacenters other than they made construction happen.
> huge, absolutely staggering amounts of energy, which typically has the effect of raising rates for everybody else
This isn't just true of datacenters. They're building an aluminum plant near me that will consume on the order of 100+ MW, which is comparable to a fairly large DC. Any other large industrial plant would consume a lot of electricity, but nobody's blaming them for utility capital investment needs.
Obviously, that's my response to your polling claims as well: people getting polled thing data centers are consuming all the water.
You've got these people going on TV and doing interviews saying that this new technology will be capable of replacing huge amounts of jobs.
The people that say that most jobs will be eliminated by AI are making direct threats of violence against the families of millions of people. Telling someone that they will no longer have any way of feeding their children is telling them that you're going to kill their family.
I'm not saying this to convince you that that's true. You could make some counter-arguments to that, and I'm sure people will come along and do that. The point is to understand it that that's how it's perceived by the people hearing it.
And then to add insult to injury, they want to come do it in your town.
They want to pitch a giant "development" project. But remember nobody ever wants “development” except for the fact that it might bring jobs and economic growth, to them, not you living somewhere else.
Meanwhile, you've already told them that the reason we need so many more data centers all of a sudden is specifically so that we can eliminate whatever jobs are left that we didn't get a chance to eliminate over the last several decades when we exported most of the jobs we could to different countries.
By the way we're the same assholes who said "trust us" last time, and pitched the idea that it would lead to a prosperous future for everyone. Instead you made money off sending the jobs to china, capturing all excess value with software, and addicting our children to opioids.
And your next exciting offer for us is a giant windowless building that will take up open land, nature, and yes, water and electricity, and employ almost nobody.
People don't feel like they control a whole lot, but they do have some ability to complain about this thing happening where they live.
The actions of people opposing these data centers are completely rational. If I'm trying to make one point, that's it.
The actually interesting part of this back and forth on HN is that Silicon Valley (aka "tech") culture has grown so fundamentally rotten at its core that not only do people not have values that place humans first, they can't even recognize when other people do.
This is really the only point that matters. It's held true my whole life, and I expect it will be true until I die. Doesn't matter what you build. I was watching a couple transmission line projects over the past decade that are still in the "lawsuits and community rage" phase - proposed to bring wind farm watts to load centers - and now they will likely be killed since the anti-development folks can pretend they were always due to Datacenters.
Datacenters are just an easy scapegoat for the anti-development crowd. It's been amazing to watch how quickly it's gone, and how folks have such a strong opinion on stuff that otherwise would have been built half a mile from them and they'd never have known any better until they were told to care.
The rest of what you wrote is largely social media driven ragebait in comment form. Kernels of truth, but largely immaterial.
Datacenter land use is the least interesting thing you could possibly discuss. Knocking out some corn fields and building some warehouses off the road no one can see or hear, with almost no traffic to/from them after construction is pretty much the lowest possible bar for local community impact for quite literally any project. It change nothing for anyone, other than the farmers who sold the land and that a few local trades companies have a couple decades of stable highly paid employment.
Therefore, the only way to get communities up in arms about these things is basically lie about it.
It's going to be a real head scratcher to folks when electricity rates continue to march upwards even if they get all AI datacenter construction banned. The green tech crowd who had the datacenter bogeyman land in their lap is playing an exceedingly dangerous game here. What we are largely seeing is the bill coming due for generational lack of investment into the grid.
If your local community can't figure out how to get the money raining down from the skies like it is now to subsidize the build out of your local infrastructure for something as minor as municipal water treatment in Wisconsin You likely will never be building anything at all.
https://www.axios.com/2025/05/28/ai-jobs-white-collar-unempl...
The people who want these data centers are telling us that the reason that they want them is to keep momentum on their plans to destroy our entire way of life.
They just say it. They go on TV and do interviews and say it out loud.
So then everyone who has any ability to stop it in any possible way tries to do that.
What exactly the fuck do you think they're going to do?
And more importantly, what causes you to feel that it's obvious that they are only coming to these opinions through ignorance?
Lumping together anti-data-center sentiment with anti-development sentiment in general is bullshit. Yes, there are certainly impossible to negotiate with NIMBYs who don't like apartment buildings because they cast shadows on a sidewalk corner or something. That has nothing to do with this.
The politics of data centers are completely different. If this was a real, actual job-creating industrial project, people would react differently.
I mean, you just did it yourself. You made a great point by doing so.
It's generally the same people. Your rants pretty much prove it. Plus I've been in meetings where it's literally the same people. They will use any and all reasons to stop local development and then stick with the one that gets the most popular traction.
I'm not talking about folks against residential development. I'm talking being around projects and in local meetings about industrial development - primarily electric generation and transmission. The arguments are pretty much the same.
> The people who want these data centers are telling us that the reason that they want them is to keep momentum on their plans to destroy our entire way of life.
AI is now taking rural blue collar jobs? I find it very difficult to believe this is a real grass roots concern.
I find it very easy to believe white collar folks are using their relative positions of power to amp up concerns rural blue collar folks would actually care about. Often at the expense of said blue collar folks.
These facilities would often be a win for a local community with a little bit of foresight. It doesn't matter to your power bill if they are sited 5 miles down the county road from you, or 200 miles. Chances are they are using electricity from your regional interconnect and that's where your power bills come from.
> And more importantly, what causes you to feel that it's obvious that they are only coming to these opinions through ignorance?
Only? Of course not. Primarily? Obviously. For the simple fact that dozens of people in my orbit who never knew datacenters existed near them all of a sudden Care Very Much(tm) about the subject after watching a few very low information videos. These are folks who drove past local facilities most of their lives and never had a clue.
> If this was a real, actual job-creating industrial project, people would react differently.
Not in my experience. Pretty much every single industrial project is nearly impossible to build in the US. Heavy industry sounds fun until someone wants to build an aluminum smelter, copper mine, or wind farm down the way. There are tons of infrastructure projects that need to get done which will not because they are not point-source locations that employ lots of jobs for a single community.
My previous example are wind farms. Typically those are best sited a few hundred miles away from a major metropolitan/load center. Good luck getting anything of scale built these days now that we've more or less burned up every ounce of spare electric transmission capacity leftover from before we de-industrialized ourselves. Once you get out of re-using existing right-of-way you see nearly the same backlash as AI datacenters. The difference? The urban laptop classes don't take your side and the outrage tends to stay localized.
It's very interesting to me a certain class of folks have convinced people that the closest thing to "free money" for a local community is a bad thing. Construction phase might suck, but assuming it's simply a datacenter and not a power plant with a co-located datacenter off the to side, you really can't get any lighter-touch land use than this. It's probably less environmentally impactful (in a negative way) to the local community than the 100 acre corn field it replaces. And employs more people to boot.
I would 100% agree with you on any tax abatements/credits/etc. These facilities do not need them and would be quite happy to pay full-freight on taxes on top of contributing towards upgrading local infrastructure far beyond their expected impact on it. This is where I feel that local politicians have seriously shit the bed all over their communities.
More Indians maybe.
https://news.gallup.com/poll/709772/americans-oppose-data-ce...
"Seven in 10 Americans oppose constructing data centers for artificial intelligence in their local area, including nearly half, 48%, who are strongly opposed. Barely a quarter favor these projects, with 7% strongly in favor."
This is another way of me countering your main point here.
It's not the same people you've seen in previous arguments.
It's those people plus almost all of the other people because overwhelming 70%+ majorities don't want these things.
So we know that. That's the given here, assuming you think there's at least some validity to Gallup and polling.
The question then becomes why? My rants are intended to illustrate why. They sound like "rants" because people are extremely fucking angry about how things are going, so accurately restating their opinions also sounds angry.
The culture has changed. People do not fucking trust tech companies and their leadership at all for extremely valid reasons. Talk about wind farms all you want. I can go ahead and talk about the Yankees. The conversation at hand is about data centers, and data centers are their own issue.
When people like you come along and say it's "free money" nobody fucking believes you because there is a rich and long history of that kind of bullshit already.
There's a point of view here that they oppose data centers out of ignorance. My countervailing point is that their distrust is almost certainly rational, considered, and correct.
Distrust is the only sane response to recent events.
> The ratio of jobs and taxes to local resources tied up has to be one of the worst possible trade-offs of any industrial use that you can envision. That is precisely why data centers are proving to be profoundly unpopular.
Nonetheless I can see that fixating on this instead of my obviously correct point which is that people absolutely hate data centers and the sociopathic tech billionaires behind them for rational reasons is a good dodge.
Hardly unexpected when I am posting on the promotional discussion board of a private equity firm responsible for launching many of these sociopaths into society.
My later comment about water usage is both unneeded for that point to stand and trivially proven true.
So yes, one of the two key players here is the heavily financed tech sector.
Starting to think these disingenuous replies are in fact the point here.
Since none of of these conversations on HN make much difference anyways, my only hope is that one person, somewhere, who is a part of the "tech community" that usually gathers here (and of which I am a longtime member) becomes newly aware that people fucking hate us now for excellent and extremely well-supported reasons and thinks a little about how that happened and how they might change it.
It's fine that you think the discussion is unproductive; I agree, our premises are too far apart to get anywhere. But you'd get further with, well, everybody that doesn't already agree with you if you'd stop accusing anybody who doesn't agree with you of being "disingenuous".
And on a personal level, I hope you reconsider your life decisions a little and realize the importance of a culture where people can feed their families and lead healthy productive lives of dignity, regardless of who their parents are or how gifted they happen to be intellectually.
You're saying the data center has a small footprint, but it does not. You don't need that much stormwater pond if you don't have giant impervious surfaces.
Deal sweetener is footprint (buildings, parking lots, and substation) being 1/6 of the land with the other 5/6 dedicated to nature preservation.
Not all wide open space is created equal. If it's wide open space with Eurasian grasses that get cut every few weeks, it's useless and has 0 benefit to the local ecology. Even if they weren't cut, they're still do very little for the ecology in the area.
If those forests don't have keystone tree species, then it's the same.
The people running these companies are so incompetent, they can't even do greenwashng right.
https://www.google.com/maps/search/Walmart+Distribution+Cent...
Even then, it is improper to assume anything about the stormwater management needs of one site based on another unrelated site. But even then, Walmart's site seems to be completely surrounded by stormwater management. Even the northwest corner of the site is a detention basin. [0]
The most obvious and logical conclusion to be made here is that an engineer told Microsoft they needed to have stormwater management of that size, so that's what they put in their plans. No sweetness, just lawsuit prevention.
If they want to be sweet, they should be building huge nature preserves (and they have enough money to afford it) into these plans instead of trying to be greedy by building the largest possible structures (they think) they can get away with.
[0]: https://beacon.schneidercorp.com/Application.aspx?App=DodgeC...
(also thanks for the useful message telling me to "contact the website owner... while blocking me from the website where the contact info should be)
I think this is a poor analogy, unnecessarily politicizing the topic.
It might be a good analogy the other way around, if hackers DDOSed the website as revenge for partial IP-based blocking, in order to apply pressure to the website operator to remove IP-based blocking. But that wasn't the topic.
I work around this by using my phone connection with phone chrome.
And no I do not blame small website owners they just have to live with this mess same as everyone else.
For DDoS resistance... Well I can imagine a world where a tech in the same area as IPFS or freenet gives backup access to websites that are overloaded.
Are we getting that before or after personal jet packs, flying cars, and my tacos delivered via tacocopters?
I'll protect my sites with Cloudflare until then, thanks.
As a small website owner, I can use Cloudflare or I can wait for this imagined tech.
Ofc it's your choice.
For some reason there are many small sites I have no problem visiting and then there are those CF users which may or may not work at any given moment, forcing me to ignore them.
Well, good luck. You are cutting yourself from the internet, not cutting me off.
Someone said it's like protecting yourself from stabbings by banning kitchen knives, but it's more like protecting yourself from stabbings by wearing thick rubber armor at all times, that also happens to be filled with spy cameras somebody else owns. It's a bit of an overreaction to a very rare threat, don't you think?
If DDOS is really the problem we want to solve then it would be awesome if one can do it without looking into the packet. SSL terminating at some centralized third party provider is way too much power.
This is something that would be perfect for cloudflare to host and sell as a service - static web pages via their CDN network.
I do not work in web development, so im sure there are plenty of details im ignorant of, but the TLDR of "how to fight accidental DDOS because of AI tooling " is make it easier for them to get the content they want.
> If small websites can just be DDOS'd out of existence
DDOS doesn't destroy websites. It just makes them unreachable until the disgruntled person decides it's been running long enough.
Please stop exaggerating a very real problem only a few entities on the web have; what you are perpetuating is FUD, which enables companies like Cloudflare to kill the web.
> The next thing you'll hear about is a monthly service fee to the hackers as a protection racket
How do you not even see the irony of this?
You can be absolutely destroyed if your hosting provider later hits you as a Website Owner with an excess traffic bill.
I'm not exaggerating, I'm just playing what if. That's a game where you think of random things that could go wrong, and then deciding if it is worth the expense. Just because maybe you can't think of things of varying plausibility does not make me exaggerating. We already see ransomware working from the hacker's perspective. There's no reason to think that greed will not come into play. If I can think of it, there's no reason to think that hackers are not also considering various ways to expand on ransomware as a service
>> The next thing you'll hear about is a monthly service fee to the hackers as a protection racket
> How do you not even see the irony of this?
How do you not? If every hacking group can come along and extort any site they choose to pay them a protection fee, there's no way websites will accept any of this. Compare that to paying a single legit service protecting against all of those hacking groups. Can't imagine why people would be willing to do that.
Larger hosting providers may have enough bandwidth to accept and drop most inbound volumetric DDoS. Smaller providers often contract with a DDoS mitigation service; during an attack the mitigation service BGP announces the attacked range, filters out problematic traffic and forwards the rest.
It's been at least 7 years since I ran servers that routinely saw DDoS... at that time, we didn't tend to attract attacks anywhere near 10Gbps, so having 10G connections for severs likely to be targetted did most of the work. I'm sure it's worse now, so 10G might not cover easy attacks but 100G servers aren't hard to find. Attacks do get larger than 100G, but most sites won't be hit with a large attack.
For a small site, taking an outage during a DDoS is often an acceptable alternative too.
If you've run a site like that, pretty soon you realize 100% of the traffic that hits you from Asia, Russia, the Middle East and even Eastern Europe / the Baltics is exploit detection scripts and is just noise in your logs. Okay, 99.999999% as once every decade something ends up on HN and gets a broader audience.
What's the point of publishing news or content on the web if you don't want it to be accessible? What about locals who travel? What about locals who share links with others?
>What's the point of publishing news or content on the web if you don't want it to be accessible
Because you don't want the burden of far away users who will never represent a penny of income for your content? This is a weirdly entitled comment.
Quite aside from certain countries disproportionately account for malicious traffic, often there are legal issues that come into play as well. This is why many regional sites block EU locations because they don't want the compliance costs for users that aren't their base.
I hit this a lot with Firefox VPN and it's ridiculous
VILLAGE OF CALEDONIA, Wis. — Microsoft has decided not to move forward with its proposed site for a data center in the Village of Caledonia after facing significant community pushback from residents.
PREVIOUS COVERAGE | Microsoft data center proposal continues to divide Caledonia residents as rezoning plans move forward
“Based on the community feedback we heard, we have chosen not to move forward with this site,” a Microsoft spokesperson said in a statement Wednesday.
The tech giant’s decision comes after hundreds of residents voiced opposition to the project over recent weeks. More than 2,000 people signed a petition opposing a rezoning proposal that would have allowed the data center to be built on 244 acres of land.
Watch: Microsoft pulls plug on plans for 244-acre data center in Caledonia after community pushback
The proposed site was situated on County Line Road and State Highway 32, southwest of the WE Energies Oak Creek Power Plant, and was surrounded by farmland and residential properties.
47032805-Concept Site Plan - Project Nova by TMJ4 News
Despite abandoning this particular location, Microsoft indicated it remains interested in investing in Southeast Wisconsin.
The spokesperson said the company looks forward to “working with the Village of Caledonia and Racine County leaders to identify a site that aligns with community priorities and our long-term development goals.”
TMJ4’s Jenna Rae, who has been following this story, reached out to Todd Willis, the village administrator, who provided the following statement:
- Todd Willis, Village AdministratorResident Prescott Balch told TMJ4 that his phone did not stop ringing on Wednesday morning, as people delivered the news. PRESCOTT BALCH TMJ4 Prescott Balch lives in Caledonia. Balch welcomed the news that Microsoft is changing plans to bring a data center in the area.
"We're ecstatic that those arguments held water and ultimately convinced a large corporation to back off, so great day here in Caledonia," Balch said.
Village trustee Nancy Pierce says she learned about the change from a news article.
"I have a lot of respect for Microsoft, making the decision when they say they listened to the constituents. They also listened to board questions both at the planning commission at the board level. I believe that they took a lot of different pieces of information into play," Pierce stated. Nancy Pierce TMJ4 News Nancy Pierce is a village trustee in Caledonia.
Both Pierce and Balch made it clear that they are not opposed to working with Microsoft in Caledonia.
As the tech giant looks for a new site, there is hope that there are improvements to the overall process.
"I would’ve liked to been able to engage directly with Microsoft much earlier in the process. We were not allowed to do that. I think that became an obstacle for a lot of different points and reasons," Pierce explained. "I feel like now they would come forward much quicker and engage directly with the community, really get to understand the community."
"There are people that have an opinion about what they want to do with their village, and that was absent in this to me. That's the real message of this thing," Balch explained. "Let's help Microsoft find the right spot in southeast Wisconsin."
Absolutely 100% force them to make a genuine case for them and to sign contracts demanding they live up to their obligations. Otherwise, it's more evidence why billionaires need to write and sign some reality checks on their dragon piles.
[0]: https://www.macrotrends.net/stocks/charts/MSFT/microsoft/sto...
Certain new emerging towns would get the moniker of "DC towns". New economies might flourish - perhaps not jobs from DC but certainly the tax money should help.
This could happen if the NIMBY movement weren't so extreme.
But, companies put a lot of capital equipment in them, and capital equipment accrues property tax, so there should be taxes paid from that.
This is a shockingly high upside relative to any other industry, like Car or Steel industry which pollute way more.
These taxes can be used to create better infra and have other things going on. Better schools and maybe even research facilities.
But I do think the opposition is largely ideological in nature so these arguments don't matter at the end.
How good a school is - which we can expand into more detailed definitions like "how much do children learn there?", "how much does any individual kid like or hate it?", "is this a good use of a child's time?" - is dominated by what the other kids they are in a shared social setting are like, and not by how expensive the facilities are or how well the teachers are paid.
Although I obviously don't care about Microsoft's outcome here, this was clearly a great site at the intersection of two transmission lines and with essentially infinite water resources.
The data center would have been built in this scene. https://www.google.com/maps/@42.8440852,-87.8474228,2445m/da...
I live beneath two transmission lines (overlapping, I guess, but not intersecting) and would prefer no data centre built here. Why? Because it will provide me no benefit whatsoever, reduce my property value, and worsen my quality of life due to things like light pollution and noise.
If data centre operators would fix these things perhaps people would feel differently. For example - provide multi gigabit fibre Internet to everyone nearby.
Kind of a cool idea, actually. These data centers could turn the towns where they build into startup incubators. Offer free high speed internet and heavily subsidized compute to residents in exchange for building there. At least gives back economically somewhat, as a data center itself doesn't provide much in return.
Meetings are usually under 10Mbps of downstream and 2Mbps upstream.
It’s simply false you need 1Gbps of “low latency” internet to work from home effectively unless your job is something like content production and working with huge files.
Producing for OnlyFans? Benefit from 1G to send raw to your editor.
Doing most other forms of work? Doesn’t require remotely close to this amount of bandwidth.
I've gotten by on McDonald's wifi
I would disagree that it provides no benefit whatsoever. I ran the numbers on the proposed campuses, and at half price for industrial property taxes, it's at full build-out going to bring a billion dollars a year to the city. The property taxes I pay, which 73% of go to the local school district is about 108 million per year.
Unless the city royally screws this up, my property taxes are going to go down. Only a tiny number of residents will live near the data centers. But, the opposition is massive locally here, and I've been trying to understand why. Talking to people in various groups, there seems to be more going on with the opposition than simple NIMBYism (although that's part of it).
Some of the cited reasons are understandable - concerns about electricity. The state passed a bill to provide some relief, and there should be no water impacts locally here.
> Some of us would like to keep our “infinite” water resources which actually aren’t infinite.
We're talking Lake Michigan, which is where the local data centers by me will be sourcing their water from (our city and surrounding cities are switching from deep aquifer wells to lake water brought in via very long pipe). Forget the data centers (that's a drop in the bucket compared to the cities' usage), such pipes usually have a capacity of 50-100 million gallons per day (MGD). That'd be able to drain Lake Michigan in about 35000 years, assuming no rainfall. Forget draining, how many such pipes do you have to build for it to start draining water faster than it's flowing into the lake? At 100MGD, 300 or so.
So yeah, it's not infinite. Enough of these big pipes built out across multiple states could have an effect. IL is apparently limited to ~2000 MGD, and any other state bordering a lake (any great lake) has limits/usage far, far lower than that (Michigan is apparently 5MGD?). It doesn't appear to be too much of a concern. None of what's allowed would add up to taking out more lake water than is flowing into the lake. As long as the current limits are kept as-is.
Should I not be able to use youtube or order online because we don't have a DC right next door?
It is all about how you plan, build etc how good you are.
There's no reason they can't be economic accelerators for the towns they are in.
In your country things may be proceeding differently, but that's the story here.
I believe in many areas this is the kind of infrastructure investment that regulated utilities can easily pass onto existing rate payers, which is where the problem/narrative comes from in the first place. So your theory doesn't match the actual results people are complaining about.
Same thing when datacenters take power generation into their own hands and build new gas turbines rather than solar/batteries or paying to expedite grid construction. And I'm willing to believe these problems happen in a minority of cases, but the problem is that they seemingly do happen and existing residents are left without recourse.
This is talking about the capital assets themselves, which are of course separate from regular maintenance like tree trimming.
About the only thing I can think of is something like a bulk of the grid infrastructure was built out 30/40/50/whatever years ago, and is now all reaching its expected lifetime all at once, and perhaps some types of equipment are strictly retired at the end of their lifespan lest they fail catastrophically.
Sending kilobytes of text over thousands of miles is a lot easier than piping energy or housing across distance!
Data centers do not provide jobs and they are run by sociopath Americans who couldn't give a shit about human rights or the environment.
You mean the carefully cropped photo of pristine rolling farmland in the article is in reality next door to a coal-fired power plant? Say it ain't so.
Modern datacenters also require very high standards of construction and are complex, so these projects create jobs and also represent a real training, upskilling and work experience opportunity for labor. There are many examples of electricians, plumbers and groundwork teams who did Microsoft’s site getting future work from Meta, Google or Amazon in the same part of the state because the experience has value.
It’s easy to dismissively say datacenter is bad, or that it consumes too much water (despite many datacenters accused of this being a closed-loop cooling system), and ignore the billions of dollars spent during the project on labor which supports that local economy, or the improvements negotiated for the local area and paid for the hyperscaler, bundled in by the city/county planning as part of the permits and approvals.
It’s also rare the tax for a campus is fully rebated, although it’s normal for the improvements to be partially rebated for some period (this is an investment incentive). Viewed over 20-40 years these sites are often tremendously lucrative in tax for the county/city as well.
There are very few jobs during operation. Mostly site security and a few tech support staff. There will be some steady work for maintenance contractors, but that's much less than the initial construction.
What would you prefer? To me, local communities tend to benefit in multiple different ways during and after these projects, poorer communities become richer, communities with little opportunity now have more opportunity. I’m always a bit baffled by someone saying “Please don’t invest $5B and create 100s of jobs and taxable improvements in my back yard”.
This is a common argument: wanting 1000s of jobs during construction and 1000s of jobs after construction, but this isn’t a car manufacturing plant. That’s a “we want our cake and we want to eat it too” argument — not saying it’s your argument just that this comes up frequently.
So you won’t see the building cease being useful for about 20 years. You’re getting usually two full cycles of “the servers inside” before a renovation program (or potentially asset/building disposal to another party, or demolition, depending what changes in those 20 years really).
What’s often happening around these tax agreements is they are a mixture of incentive and an offset of prepaid improvement costs to the city/county for developing mains water, sewer, roads, schools, fire and police, and other infrastructure, sufficient to support the 100s of families who may move here to take the post-commissioning jobs, etc.
Often money out the door for the hyperscaler is about the same over 20 years, it’s just some number of $M’s is paid upfront as an “Contribution to Improvements”. That’s actually good for the municipality involved, too.
Municipalities typically know or are told these aren’t tax rebate for 10 years and then it’s getting bulldozed. They’re sophisticated enough and well advised to understand this is a 20+ year investment in their town.
the tax rebates can extend for decades
> these sites are often tremendously lucrative in tax for the county/city as well
these datacenters haven't been around long enough to know that; these are _not_ your grandma's datacenters -- they are much more resource-hungry since they do substantially more processing
I don't think people would care so much if they were using lake or stream water that got sent back, but often they don't because it isn't clean water which means more money spent on filtration and/or cooling system maintence.