Why would the state, the county, and the military even consider letting an inexperienced developer with zero past successful datacenter projects undertake such a vast project?
I’ve been around the datacenter world for many cycles, and this story seems like a tale old as time.
Again, why would someone with no experience take such an enormous risk?
It may seem easy to build these things; but it’s actually quite difficult to get it all up and running safely and in sync at full capacity.
Even experienced people in these fields mess up from time to time.
For example, the largest electrical contractor in the dc area that was multigenerational family owned went under building the NSA data center in Utah, and they knew what they were doing!
This was an expert electrical contractor with over 50 years of experience, and they got in way over their head with cost over runs and bad engineering.
Somehow phases got crossed and things blew up if I remember correctly, and it lead to power surges that took them under.
It's simple. Whoever gives them the biggest kickbacks and favors gets the award as long as they have some track record. At the top, you don't need a track record, you just need to be friends.
>Why would the state, the county, and the military even consider letting an inexperienced developer with zero past successful datacenter projects undertake such a vast project?
Because the exact details are likely structured from their vantage point of all the decision makers it's structured as a heads I win tails you lose proposition.
This is common in state and municipal permitting and just about everything else that the government has discretion for. I would even go so far to say that this is normal for anything that isn't so cookie cutter it gets the rubber stamp.
If you're an upstart or your project is just so outrageous that can't get what you want via a "here's what I'm gonna do, here's why it complies with your bullshit, send me permits or a court date" frontal assault you make concessions. You (pay a PI to) figure out who the approvers are and stuff about them. You use the "right" civil engineering firm for all the surveying bullshit, the carbon offset BS projects in your proposal are in the "right" towns and places that benefit the "right" people and causes. If the head of some relevant department is a bird geek then you make one of your offsets a wetland restoration abutting and in partnership with some Audubon Society site. And on and on and on. Maybe if they've really got you over a barrel you front load it so that all these niceties get done up front and the stuff that you actually want to do gets done second.
Slimy as it is, none of this is illegal. It's just how the sausage is made in a high regulatory environment. You have to make all the right decisions on the proposal side so that the people on the government side making their decisions have all the reason to approve you. It's an overpriced appeal to emotion of sorts.
So anyway, you structure it like this and as far as they care worst case the endeavor goes tits up and leave a Parks and Rec style pit for them to use other people's money to clean up and it's still a win because it legitimizes the need for further extractive concessions paid up front from the next dreamer who thinks they're gonna pull off some project like yours. Or they know who they want to finish out your project if/when it fails.
I’m reminded (as I frequently am) of Vernor Vinge’s sci fi novels. In A Fire Upon the Deep, the main character comes to the realization that all civilizations collapse when the web of legal and supply chain dependencies collapses in on its own weight. As he is a sib-lightspeed traveler, he has to calculate the likelihood a civilization will still exist when he arrives.
AI data centers are a new, expensive, and (soon to be) highly integrated layer on top of our economy’s tirimasu of dependencies. Especially as new generations come and rely more fully on the machines than we are comfortable doing ourselves.
Right new if you destroy a data center you might destroy tax documents, calendars, sensitive user data, etc. 50 years from now destroying an AI center could mean no one can fly a plane or produce legal arguments anymore. So much of our productivity will be tied into these datacenters and our new skill sets will be entangled in their operation.
> MIDA executive director Paul Morris told county commissioners that the facility “will not take one electron” from the existing grid
Cool. Will it drive up the price of the existing grid's existing electrons though? It'll increase demand for the natural gas that keeps those electrons moving.
> MIDA executive director Paul Morris told county commissioners that the facility “will not take one electron” from the existing grid and could eventually feed surplus power back into it.
And when they do and the residents' electricity costs go up because of it (which is baffling to me; the companies that use it should pay a premium for it that offsets the residential use), I guess there's just nothing that could be done. Not. One. Thing.
in a California port city, there were newspaper headlines when a City Council swore to everyone that tearing down hundreds of good houses to put in a freeway to a far away transportation center "will benefit this great city" .. It is a historical laughing stock. Many neighborhoods are thick with foul diesel smoke for sixty years now, and the ordinary gardens and yards that were paved over, are long gone. The benefit was to the regional shippers, who grew truck traffic 4x then 40x.
Elon Musk built his whole identity around electric cars to save the world from fossil fuels, and now he's powering data centers with natural gas like everyone else. Lots of tech bros are like this out there.
10 years ago: We need to save the environment from CO2 global warming!
now: Broooo, you need to use AI or you're getting left behind!
This deal is very strange and smells funky.
Why would the state, the county, and the military even consider letting an inexperienced developer with zero past successful datacenter projects undertake such a vast project?
I’ve been around the datacenter world for many cycles, and this story seems like a tale old as time.
Again, why would someone with no experience take such an enormous risk?
It may seem easy to build these things; but it’s actually quite difficult to get it all up and running safely and in sync at full capacity.
Even experienced people in these fields mess up from time to time.
For example, the largest electrical contractor in the dc area that was multigenerational family owned went under building the NSA data center in Utah, and they knew what they were doing!
This was an expert electrical contractor with over 50 years of experience, and they got in way over their head with cost over runs and bad engineering.
Somehow phases got crossed and things blew up if I remember correctly, and it lead to power surges that took them under.
https://www.constructiondive.com/news/10th-largest-us-electr...
https://www.bizjournals.com/washington/breaking_ground/2014/...
How exactly is the shark tank guys property management company going to pull this off?
> why would someone with no experience take such an enormous risk?
6% of Americans think they could defeat a grizzly bear unarmed[1]. There is no upper bound to human confidence[hubris].
1. https://yougov.com/en-us/articles/35852-lions-and-tigers-and...
> How exactly is the shark tank guys property management company going to pull this off?
By hiring expert subcontractors to actually do the work
how does an idiot selwct experts, is a interesting paradox.
It's simple. Whoever gives them the biggest kickbacks and favors gets the award as long as they have some track record. At the top, you don't need a track record, you just need to be friends.
>Why would the state, the county, and the military even consider letting an inexperienced developer with zero past successful datacenter projects undertake such a vast project?
Because the exact details are likely structured from their vantage point of all the decision makers it's structured as a heads I win tails you lose proposition.
This is common in state and municipal permitting and just about everything else that the government has discretion for. I would even go so far to say that this is normal for anything that isn't so cookie cutter it gets the rubber stamp.
If you're an upstart or your project is just so outrageous that can't get what you want via a "here's what I'm gonna do, here's why it complies with your bullshit, send me permits or a court date" frontal assault you make concessions. You (pay a PI to) figure out who the approvers are and stuff about them. You use the "right" civil engineering firm for all the surveying bullshit, the carbon offset BS projects in your proposal are in the "right" towns and places that benefit the "right" people and causes. If the head of some relevant department is a bird geek then you make one of your offsets a wetland restoration abutting and in partnership with some Audubon Society site. And on and on and on. Maybe if they've really got you over a barrel you front load it so that all these niceties get done up front and the stuff that you actually want to do gets done second.
Slimy as it is, none of this is illegal. It's just how the sausage is made in a high regulatory environment. You have to make all the right decisions on the proposal side so that the people on the government side making their decisions have all the reason to approve you. It's an overpriced appeal to emotion of sorts.
So anyway, you structure it like this and as far as they care worst case the endeavor goes tits up and leave a Parks and Rec style pit for them to use other people's money to clean up and it's still a win because it legitimizes the need for further extractive concessions paid up front from the next dreamer who thinks they're gonna pull off some project like yours. Or they know who they want to finish out your project if/when it fails.
I’m reminded (as I frequently am) of Vernor Vinge’s sci fi novels. In A Fire Upon the Deep, the main character comes to the realization that all civilizations collapse when the web of legal and supply chain dependencies collapses in on its own weight. As he is a sib-lightspeed traveler, he has to calculate the likelihood a civilization will still exist when he arrives.
AI data centers are a new, expensive, and (soon to be) highly integrated layer on top of our economy’s tirimasu of dependencies. Especially as new generations come and rely more fully on the machines than we are comfortable doing ourselves.
Right new if you destroy a data center you might destroy tax documents, calendars, sensitive user data, etc. 50 years from now destroying an AI center could mean no one can fly a plane or produce legal arguments anymore. So much of our productivity will be tied into these datacenters and our new skill sets will be entangled in their operation.
Not good, not bad, just different.
> MIDA executive director Paul Morris told county commissioners that the facility “will not take one electron” from the existing grid
Cool. Will it drive up the price of the existing grid's existing electrons though? It'll increase demand for the natural gas that keeps those electrons moving.
No, it will just burn GWs worth of natural gas.
Benn Jordan has captured quite an interesting evidence: https://youtu.be/_bP80DEAbuo
> will not take one electron
I'm torn between his intent of blatantly lying, and how technically correct that wording is.
Related 2d ago https://news.ycombinator.com/item?id=47910985
Heh this post right now is HN 103 for me while this[1] is 102
[1] https://news.ycombinator.com/item?id=47939160
> MIDA executive director Paul Morris told county commissioners that the facility “will not take one electron” from the existing grid and could eventually feed surplus power back into it.
And when they do and the residents' electricity costs go up because of it (which is baffling to me; the companies that use it should pay a premium for it that offsets the residential use), I guess there's just nothing that could be done. Not. One. Thing.
in a California port city, there were newspaper headlines when a City Council swore to everyone that tearing down hundreds of good houses to put in a freeway to a far away transportation center "will benefit this great city" .. It is a historical laughing stock. Many neighborhoods are thick with foul diesel smoke for sixty years now, and the ordinary gardens and yards that were paved over, are long gone. The benefit was to the regional shippers, who grew truck traffic 4x then 40x.
Setting a reminder on my phone to come check back on this in five years.
>powered by natural gas
I love how AI made all the tech bros flip on 40 years of calling CO2 "pollution."
Are you implying that people are wrong to call CO2 pollution?
Tech bro leadership never cared about the environment. You are thinking of two different groups of people.
Tech workers aren't any better, they have no qualms with "just following orders" as long as they get their high salaries and stock grants.
Every person in an industry is uniformly the same!
Elon Musk built his whole identity around electric cars to save the world from fossil fuels, and now he's powering data centers with natural gas like everyone else. Lots of tech bros are like this out there.
10 years ago: We need to save the environment from CO2 global warming!
now: Broooo, you need to use AI or you're getting left behind!
I'll admit, I was tricked by him too. Disillusionment started in 2016-2017 for me though.
Just because a someone goes and lies to the public as easy as he breaths, doesn't mean he ever had any convictions.