United States Power Outage Map | Hacker News Reader

United States Power Outage Map

poweroutage.us

320 pointsjonbaer151 comments

This is a very interesting business. Simple and appealing to a general user but apparently worth $1k/m for business users who want access to consolidated data via API.

I'm sure there's a host of businesses like this, but I don't know who the customer base is. $1k/m seems to be a lot, but I don't have a need!

Grocery chains where lack of power = no freezers = spoilage come to mind.

Certainly grocery stores don't need to ask someone else if their freezers are getting power, no? They would have monitors, people and computers, that are doing this job.

Additionally, even as a residential customer, my utility reaches out to me when they suspect there's an outage...

Is there not a possibility that if a neighbouring area has gone down, then your area might be at risk of going down if they need to do maintenace on all surrounding areas to fix the issue? I would imagine its useful for store owners to know the power is out a mile down the road, and might enable them to start making preparations in case they are next.

Think more about centralized ops. Any individual store will of course notice their power has gone out.

But what do they do then? Call around to every nearby store asking if they still have power, and have space in cold storage? While those stores may well be calling yet other stores?

How does that work when there are 6 affected stores and 12 nearby ones that still have power? Maybe 3 of those have lots of cold storage, and another 4 have some but not lots free.

A centralized view could easily be worth $1k/month.

(and you are very fortunate if your utility reaches out to you in mass outages (>1000 customers), and especially if they do it via API your systems can act on)

I mean, your stores certainly must be pinging back HQ. If they stop pinging, something is up. Gotta determine if it is power or connectivity.

But the ISP case definitely is a valid case for 1k/month.

> Think more about centralized ops. Any individual store will of course notice their power has gone out.

And those individual stores report to centralized ops. This is a terrible usecase

How do those stores report to HQ when they don't have power? Even cell towers could be down if the situation is bad enough.

The lack of a signal/data gets investigated, upon which you find out the store has no power.

And that's _much_ more expensive than 1k/month. At least if you keep the ability to investigate effectively around all the time.

No it’s not. This isn’t a developing country.

A grocery chain will see maybe one outage in 10 years that is bad enough to take out the store and internet+cell phone networks.

Dispatching someone is absolutely more cost effective for these rare occasions.

A backup battery. If you’re worried about cell phone outages, use Starlink.

I’ve never worked in grocery but from my experience in other pre-internet industries I can easily imagine a politically marginalized CTO who would rather spend $1k on a turnkey solution than herd 100 cats who can’t reliably format a spreadsheet of data.

That turnkey solution is powered by similar cats. But at least there’s a contract to club them over the head with.

Last year, we had a major wind event that caused significant loss of power in the area. My local store pulled in multiple trucks dropping off their freezer trailers. The logistics of getting those took some time and spoilage was definitely an issue. There was no saving the produce, but the frozen items were saved.

I could see a "smart" corporate operation that detects the outages and starts to enact plans to tentatively start moving equipment around depending on predicted outage duration. Then again, I could also see where that would be deemed too expensive to do, and this is just a tech fantasy.

I’ve noticed the smaller the store/chain the more likely they are to try to save the food (and if you lose power because of a freezing storm that can be relatively easy). The larger Walmarts just go to generator or immediately give up.

They don't have backup generators? Ever grocery store bigger than a radioshack ive ever been to has backup generators specifically because the potential for losses is so damn high, not to mention all the extra work that has to be paid out to move all those goods twice only to still be risking spoilage. Even the little off-brand grocer serving a town 900 nearby has backup generators, and they are right on a main power supply line so basically have zero outages because it is top priority.

Your experience is unusual. Generators are expensive and require maintenance, and many stores (even larger chains) don’t have them. And, as anyone who’s worked with or in a data enter knows, sometimes they don’t work when needed.

Many grocery stores rely on short notice rental of refrigerated semi trailers rather than the ongoing maintenance / testing of generators.

Very few places actually have generators with the capacity to run everything. The specific store had just enough power to run limited lighting and the registers.

People always think generators are an easy solution, but that tends to be the response from people that have not actually looked into what generator solutions look like

Supermarkets pretty much all have generators on site now. There's enough value in frozen and refrigerated inventory that it's worth it. Small grocery stores probably not.

Complex datasets are always in demand! I wish we'd gotten into something simpler, because in this specific case it's very high maintenance. The data sources constantly change and have no standard for reporting. We try to make it as simple as possible for people who need the information, so it's a lot of work on our side.

Are you folks working directly with utility companies or pulling the data yourselves?

Both!

Would love to hear more about the demand for data like this. I have a “complex data” startup that’s likely similar in complexity (totally different area though) and I’d love to learn about who I might be able to sell the data to.

You can email me at tech at poweroutage.us if you want to talk.

My local ISP shows power outages overlaid on their own service status map. I'm guessing it cuts down on support calls when they're down due to no power. (I suspect they get that data directly from PG&E.)

https://www.monkeybrains.net/map/

I can imagine that anything that involves logistics, moving people or things around. Lack of electricity might disable something that's important to you, but over which you don't have direct control or precise information about. Blackouts might affect only parts of a city, and I can see why having precise, real-time information about which parts of the city have power or not, is something really useful to re-route things around the problem.

Know some folks who used to work at Genscape. The whole business was selling live power generation data. They mostly accomplished it "hands-off" and without the permission of utilities.

If you had property adjacent to a power plant they would happily lease space to setup sensors.

Apparently there are tons of businesses like this gathering data that is mostly live and selling to hedge funds etc.

Patio11 actually had a really interesting podcast (on his old podcast) with the founder of StormPulse about why businesses and governments care about these sorts of feeds [0]

[0] https://www.kalzumeus.com/2013/04/08/kalzumeus-podcast-4-sto...

I still get a kick out of this photo: https://x.com/patio11/status/332651272878575616

Ah, it's "the guy"!

You'll never get unicorn status with this, but it's probably pretty easy to get 20 customers to sign up. You'll probably cap out between 1k-10k customers if you start to really push it.

$1000 per month is what I used to charge to a corporate card for generic SaaS products without filing a purchase order. It's not a lot. Even if your business only makes $1M/yr in revenue, $1000 is 1/83rd of monthly revenue. If anything I think they're priced too low.

Honest question...

I've had several ideas sort of like this over the years in areas/industries where it hadn't been done yet, but I'm always somewhat scared of running into copyright law issues. In other words, the data isn't mine, but if I gather it, transform it, and repackage it, it would likely be useful to people and certainly more useful than in its current raw form. But again, the data came from elsewhere.

In this case, as someone else mentioned, they likely just scraped other states'/private power companies'/rural cooperatives' maps or websites to get the data. How is that not problematic?

I understand that larger companies do this all the time, but isn't the risk pretty large for smaller entities?

*Edit* - I see I didn't read far enough to see that you guys are saying that you are often working directly with the utility companies etc. But the question still stands. I understand that some things where data is pulled are just attempts to get paid for other people's data, but some are not - some, like this, legitimately add value via visualization, aggregation, and transformation. I'm talking about things like the latter case.

How much of an issue something like that is probably varies by jurisdiction and exactly what you're doing and where you're getting data https://libraries.emory.edu/research/copyright/copyright-dat...

We buy similar services for even more than 1k per month. Keep in mind that these aggregators also smooth over changes in the underlying apis. Dealing with 15-30 individual api vendors is a lot of hassle in itself.

US power providers send realtime data to the US government who provides it in a single point of access. See ODIN.

Very few do. Only 155 utilities out of over 3,000

I was referring to a aggregator service in a completely different business

Until Elon Musk fires them all and uses their servers to mine DOGE cryptocurrency.

The use case is any business with dozens/hundreds/thousands of physical locations spread throughout the country. They need to keep track of adverse weather events, road closures, power outages, etc. If you have a bunch of franchises $1k is a drop in the bucket. Let's say one of your stores opens at 7am but the system reports a power outage at 4am. Now you have actionable information to deal with it before your boots on the ground even get there.

when I worked in the IoT space this would have been a god send at that price, just so we could quickly answer if we had some large power outage occurring simultaneously with an outage.

Yea, we technically had the data to figure that out ourselves but the platform hadnt been built with that in mind and this would have probably taken 3-4 years at this price for an engineer to refactor everything to get the data and thats only if they succeeded in the task

Edit: would have probably taken 3-4 years at this price _to break even_

If you know anyone who might benefit from the data, let me know!

How does the US do this? In Canada, everything is so splintered. Open Data is available at various levels (municipal, provincial), but it's in different formats, many government bodies don't expose anything, and it's all very .. uncooridnated.

But this is not the first time I've seen data from the US which feels so well organized. Is the secret sauce the data/providers, or is the creator of this site just very good at organizing a big mess?

i think the creator is good. I think they are just scraping different power company outage maps. If you look in some places there's 0 data because the power company doesn't have a true outage map

Yeah, and for context, there is a Canada version of the map.

https://poweroutage.com/ca/

Euh, I don't think Manitoba Hydro has 600,000 customers in Québec. I wouldn't trust that map.

If you go on the page for Manitoba Hydro, it says there is 1 customer in Quebec: "South Cypress". Although South Cypress is a municipality in Manitoba.

Seems like a data issue where the municipality was double-counted and one in Quebec for some reason.

https://poweroutage.com/ca/utility/1408

An open data commitment across gov agencies was an explicit commitment and project started in 2013 during the Obama administration.

“On May 9, 2013, President Obama signed an executive order that made open and machine-readable data the new default for government information. Making information about government operations more readily available and useful is also core to the promise of a more efficient and transparent government.”

https://obamawhitehouse.archives.gov/open

I know about this because the Australian government then followed and worked on doing the same.

Data.gov became a central place to access and even request that departments open specific data sets and in more formats.

https://data.gov

https://data.gov.au

The annual GovHack hackathon was supported by many agencies, leading to a strong prize pool and people from various agencies on hand looking for teams who over the weekend had done cool things with their data (specially if it was public interest). Some projects were further funded coming out of the hackathons.

https://govhack.org/about

Year on year ongoing commitment to open data in government got us to here.

It was a great long term initiative furthering values of cizitzen engagement, open data and open government.

This is a third party site that just scrapes utility websites. So open data initiatives are not relevant.

Over a decade of open data culture got us here.

Privatised electricity companies often seek to maintain a veneer of social good. Opening data for which you plan to not make open tools yourself can be a good way to have tools built that you don’t pay for that make you look good.

Most power companies in the U.S. are not run by the government.

Over a decade of open data culture got us here. As to what next that’s out of scope of my response.

Hydro-Quebec has a web app that works very well to look at outages and service status.

Hopefully they get power back soon. I'm one of the people behind the website, so if you have any questions, feel free to reach out! Especially if you're interested in the data :)

Dope project!

FYI on your LLC page has “queries” spelled “qurries”. Not sure if it’s intentional or not though.

Thanks for catching that! Will get it fixed.

Where I live the City helpfully^H^H^H^H^H^H^H^H^H performatively lets you see where the power is out, but provides no historical information. They pay VertiGIS to provide this service. At least today (I checked, it was something else last time I checked). Woot!

"Performatively": yes. If you can't see historical information, it's performative, it misses the historical case which is the important market. If you're there and the power's out, you know. If you can do something about it in realtime, you will; otherwise you will wait until the power comes back on and disaster recovery kicks in: From when to when? For how long? If you're not there, then go directly to disaster recovery. That's my opinion, and I'm not changing it.

The second important consideration for a democratically governed entity would be: are we equally served? If not, why not? Retrospective information for the entire stakeholder group is required.

Too much ado is made of "security", and that historical information is somehow a threat to operational security. If the power is out now, the live info is the impacted targets of opportunity right now (or the ones to phone / smish when the power comes back on); if the power was out in the past, what's the threat? Are adversaries going to pre-position in areas where outages are predictable due to some foreseeable conditions? Then maybe the City should preposition resources, too. I welcome other viewpoints on this aspect.

On a practical level I have systems which are always on and logging with sufficient granularity so I know when the power went out... and when it came back on. I would think that telemetry from locations within the service area would be ultimately a more reliable way of collecting information about outages, without relying on the utilities which can't be relied on (outside of a contractual arrangement). This telemetry could be active, or passive: geolocating web browser and other internet activity (even pinging or SYNs) would likely do an adequate job, I'm sure that stationary resources could be identified in the dataset. I'm sure this is colored by the fact that I'm an "internet plumber" and telemetry and observability is what I do for a living.

PG&E publishes a real-time outage map with granularity down to the individual neighborhood grid:

https://pgealerts.alerts.pge.com/outage-tools/outage-map/

Georgia Power, also:

https://outagemap.georgiapower.com/

Oncor (Texas) too.

https://stormcenter.oncor.com

As does PSE (puget sound energy) https://www.pse.com/en/outage/outage-map

As does ComEd (Northern Illinois and Chicago)

https://secure.comed.com/FaceBook/Pages/outagemap.aspx?ipid=...

Duke Energy too: https://outagemaps.duke-energy.com

PSE&G which covers parts of NJ https://outagecenter.pseg.com/external/default.html

Perhaps they have improved now, but in the major winter storms of 2-3 years ago, we were regularly frustrated by how PG&E seemed to be spending more time on their innovative ways of reporting outages, but not actually fixing outages.

If your power is out, the "estimated restoration time" is a complete guess. We used to enjoy how it used to jump forwards 24 hours at exactly 6pm every day, during our 8-day outage.

The Entergy outage map is the best one I've used by a wide margin.

https://www.etrviewoutage.com/map?state=TX

Wow, that really is impressive!

Pacific Power does too, though it isn't too specific, no boundaries on the regions, just a hazy overall number[1]

1: https://www.pacificpower.net/outages-safety.html

Does the US map have a different definition for "power outage" than the Canada/UK map on the same website? Unless I've missed news about a tornado or hurricane, the outage statistics for the US seem rather extreme to me.

We had a strong storm come through yesterday afternoon and night. I didn't lose power at home but I'm not surprised that around 85k in my state did. I saw quite a few down trees and leaning power poles this morning.

Part of the problem was that, at least here, we got 4" of rain earlier this week. This storm brought more rain, high winds, and a lot of lightning.

Very bad ice and wind storm across the eastern seaboard this weekend.

Several days of rain (couple inches) and high winds this morning as a front moved through.

According to this report [1] the average time between blackouts in 5 years in the UK, vs 3½ years in the USA and less than 12 months in Canada. (Page 8.)

[1] https://reports.electricinsights.co.uk/wp-content/uploads/20... (via https://reports.electricinsights.co.uk/reports/q3-2019/)

The link is to their poweroutage.us, but they also cover Canada (https://poweroutage.com/ca), the UK (https://poweroutage.com/uk) and the EU (https://poweroutage.com/eu). For some reason they don't link to these from the US site.

Really interesting data, particularly when you compare the very low level of outages in Canada/UK compared with the US.

The UK regulator introduced the Interruptions Incentive Scheme in 2002 to encourage distribution networks to reduce customer interruptions and minutes lost. It triggered a large wave of investment in network automation (remote switching, auto reclosers etc.)

So basically the difference between 9s is investment/cost.

At least here (Canada), utils push all of their costs to end-users & IIUC have an incentive to have high capex/low opex networks because of regulated return.

As a Canadian residential electrical customer, we pay a lot in base fees from what I hear relative to US customers. Sure it's more reliable, but tbh, it's not worth spending much to get 5 9s (5 mins of downtime a year) vs 4 9s (50 minutes/yr). Heck, even 500 minutes/yr would be fine for me.

But commercial/industrial users won't feel the same way, and managed to successfully spread the cost of adding 9s among users that largely don't care.

Btw, at least for Ontario and Quebec, current average downtime is below 4 9s, and quite close to your quoted 500m/yr time.

Ontario's Energy Board has a dashboard (https://app.powerbi.com/view?r=eyJrIjoiNmY1YjU0NmUtMTJhYi00N...) that says in 2023, total average outage time was ~5hrs (and that's somewhat typical of the last 10ish years).

Hydro Quebec says that in 2023 (https://www.hydroquebec.com/data/documents-donnees/pdf/hqd-0... FR sorry) the average downtime was ~4.5 hours.

Its the same in the US. Except for texas which does some slightly more innovative things. Its really hard to do anything else than overbuild, Texas was strongly criticized for underbuilding its network when there was an ice storm a few years ago.

It's actually amazing looking at this map on a percentage basis.

States with 10 million plus customers that only have a few thousand without power.

Incredible achievement if you think about it.

Wow what happened. Was there a bad storm?

Severe thunderstorms this morning

https://www.atlantanewsfirst.com/2025/02/16/live-updates-tor...

Can't speak for the entire country, but rural Virginia had a nasty ice storm pass through earlier this week [1]. I lost power for about 40 hours.

[1]: https://www.wdbj7.com/2025/02/12/36k-plus-virginians-lose-po...

Not too bad, but here in Memphis most power lines are old and above ground, so any moderate system coming through will knock out a significant chunk of the grid.

South of Atlanta we had a tornado warning around 5am. Several downed trees and power outages but I haven't heard of any actual tornadoes that touched down in my area yet.

I'm in Georgia. There was a bad thunderstorm front that came through last night and briefly took out the power to my building. High winds and tornado watches across much of the state. We really need to bury our power lines.

It's very typical of this season as we experience several rounds of rapid temperature fluctuations between the 60s and 70s and the 20s and 30s. Late winter and early spring typically bring storms and tornadoes to our region.

It's funny to see everyone rush outside donning springtime apparel and then be hit with the coldest days of the season just after. And it happens every year. "Fake spring", we call it.

We've got a forecast for snow this upcoming Wednesday too! Fingers crossed.

Like the other Georgia commenters, had a severe storm starting at 4AM trailing off at 6 and ending at 7. We lost power for about 20 minutes, all my essential gear stayed up on battery power. I’m on underground power but we have trees down in Atlanta and a lot of people lost power for a few hours.

What’s amusing is this is all on the heels of another discussion about electrical grids in the US vs Europe and a lot of points made are exactly what we experienced. Our grid can be technically as competent as anywhere but a giant storm sending a tree through a transformer or distribution line is going to make for a bad day.

One note I haven't seen mentioned here yet, we got a lot of rain last week prior to this storm.

Here in Alabama we got around 4.5" in my part of the state, that's nearly our average for an entire month. The ground was already very soft before this storm, making trees more susceptible to high winds. We also had a lot of lightning here last night.

I'm actually surprised I didn't lose power. I'm in a fairly rural area with mostly above ground power lines and lots of trees within falling distance of power poles and lines.

Airplanes keep falling out of the sky and infrastructure keeps failing as Trump fires more and more FAA, DOE, and FEMA employees, and hands the reigns of government over to Musk and Putin.

But just wait until Trump pardons Jeffrey Skilling, and puts him in charge of the Department of Energy Office of Electricity, to overhaul grid modernization, cybersecurity, and resilience running the national power grid, the Energy Information Administration, to provide power grid data and analysis, and the Federal Energy Regulatory Commission to regulate interstate electricity transmission, wholesale power markets, and reliability.

https://en.wikipedia.org/wiki/Jeffrey_Skilling

why not have a more granular heatmap as the main viz? coloring by state boundaries is really unintuitive and often misleading, even without Mercator projection criticisms.

It depends on what your goal is, I suppose. If it's just to see the organization of data by administrative boundary, this is fine. For the people this company is selling to, that's probably even the ideal.

However for us in the cheap seats, we most likely as a group want to see number of people impacted. For that, administrative boundaries are the inverse of what we want given administrative boundaries tend to shrink as population increases.

Just goes to show the gap between B2C and B2B. In this case it seems this company knows their target audience, and it's not us.

im not sure i follow the logic of state boundaries being a useful level of granularity for any audience. if some suburb of Austin has a power outage affecting 10k people in a 3mi radius, that means i get to see 8.7% of the contiguous US land area turn yellow unless i specifically select to zoom to TX to discover that the affected area is 10,000x smaller?

Texas is an especially interesting example you've chosen. For political reasons, people will often focus on the entire state due to its completely independent grid from surrounding states. Whether that tells the story you or I want to read or not, is something else entirely though.

Because our priority is having the page load quickly, and having to load data for 3k+ counties on a national map would slow things down. The state view is a compromise between speed and giving the user enough info to then go more granular by checking the state view they're interested in.

Suggestion to the void: update the map so that it shows the counties and districts without power.

If you click on the state, you get a county view.

Looking at the solar data pages: Fascinating that the average power usage in Alabama is more than double California. (10kw vs 4kw)

I assume that's lots of air conditioning? Or perhaps downstream effects of appliance regulations?

It's absolutely air conditioning. It's a very humid place.

This map is spot-on! It shows Virginia, where I live, in red with over 100k customers without power.

Here is a link to a video I took in my back yard about 45 minutes ago when the wind was fierce: my power went out about 10 minutes later.

https://youtube.com/shorts/N1rcUk-1kNM?si=1s8uCdvfqa5VmrXL

Haven't lost power here in my neck of NC yet, but I was sure we would. Blew through here around 9 am this morning:

https://ibb.co/zhrxwx93

Still pretty gusty out but nothing like this morning.

That doesn’t look too fierce… the worst storm I’ve seen live had the entire treeline bending over like there was an ent king approaching from downwind.

Live power outage data is intentionally obscured by power providers as a matter of public safety. They report outages, and numbers of affected customers, etc. But efforts are made to ensure the locations of those customers cannot be too easily weaponized. It remains a question how effective those efforts are.

Is this to prevent burglary? I'm not sure I follow the logic here.

Yes.

Surprised that Puerto Rico isn't included, which has a lot of power outages of different scale.

The electricity provider there specifically asked us to not publish the data, and has gone to extensive lengths to block us from getting it.

That sucks. Would you be interested in adding states/territories with no data to the map?

Relevant: https://www.arcgis.com/home/webmap/viewer.html?layers=d40907...

PG&E with less than 10,000 customers out of power? Lies.

Because we were all burned out or forced out with absurd home insurance rates.

Three unlucky people in Wyoming without power :/

Customers, not people. Could be residential or commercial. Could even be low priority, unoccupied customers in the middle of nowhere. The advances since the adoption of smart meters is grid operators know exactly who is using how much power and who is out in near-real time as the cellular modems (assuming there is cell service in the vicinity) tells them right away when there's a problem. I'm curious if large customers have satellite to cell nanocell access points to keep in touch with remote sites and for their critical vendors, like electricity utilities, to also gain access where there isn't ordinarily cell service.

Oh so three customers could be thousands of people if it was a factory that was closed or something like that.

It makes sense to prioritize repairing the biggest dollar signs first, I suppose.

Or it could literally be three houses - depends on how accurate the reporting is when there’s a minor fault.

I’ve seen a wire melt which took out three houses - not sure if it ever appeared in state-wide data.

is that what happens when you vote for the Distressful Outages (of) Generated Electricity?

Gonna be some yucky weather going over the Northeast today. Winds, rain, then cold. Eww.

Yeah heard there is ice coming down in some places. Stay safe!

I'd love to have the opportunity to compare this to other countries.

How comparable is that though? Germany for example has less severe weather than the US

Not all of the US has the same weather either.

Is this outage at all related with the recent gvmt personnel reductions?

very cool love tools that make the game more accessible to the less smart folk like me fires and power outage wonder what else is used

Sad that the US has all these power outages. It comes from rampant capitalism. No obligation to provide redundancy so the cheapest solution it wores on wooden poles and one tree down leads to outages. In every weather event in the US the news reports xxx people without power. In the rest of the world it isn't a stat that makes the news. I presume this is because redundant grids don't often go down and when they dothe places where it happens has bigger problems.

It needs to show % instead of n-k+.

This is a lot isn't it?

Not really, 100k homes in Georgia is roughly 3% of homes.

If 3% of Sweden was without power it would be ground breaking news, a media shitstorm and remediations to not let it happen again.

We're further north so power is more essential to human life here, but 3% is a huge number to me when we're talking something as fundamental as electricity.

Power outages are pretty common at least where I have lived in America. It's not uncommon to lose power for 20-30 minutes and sometimes up to a few hours. Outages with storms are extremely common.

More rare act-of-god scenarios like the 2009 ice storm[1] left my family without power for 2 weeks. I'm sure there was some action taken, but I continued to lose power at that house at least 2-3 times a month easily.

[1] https://en.wikipedia.org/wiki/January_2009_North_American_ic...

In Italy I think I had a single power outage in the last 10 years and it was planned ahead of time for maintenance work.

That's cool that you haven't been affected, but they happen frequently in Italy.

https://www.thelocal.it/20230719/blackouts-hit-italian-citie...

I posted it already in this discussion, but [1] shows a power outage once a year or more is typical in Italy.

People in France, South Korea, Switzerland and the United Kingdom can go 5 or more years between power outages.

[1] https://reports.electricinsights.co.uk/wp-content/uploads/20...

The article you posted reports about "hundreds of people in the dark for hours on end". That makes the news, compared to hundreds of thousands of customers in the dark in the US not making any sensation.

Compared to other countries, America has normalized a different level of deviancy. Perhaps 5 things are worse in America while 3 things are better when making a single country comparison. It's important to understand the tolerance and need for service varies. For example, single family homes without fireplaces and firewood storage areas and also lacking natural gas service and off-gridable power storage are completely reliant on grid electricity power for heat.

I live in California and have had at least one power outage every year, usually lasting a few hours. Somehow we just got used to it, I think it's an example of normalization of deviance.

https://en.wikipedia.org/wiki/Normalization_of_deviance

I mean, here is pretty much 2% of Stockholm losing power a few days ago. Barely a blurb, and there wasn't even a good weather related reason for it. It was just an "accident"

https://ground.news/article/power-outages-in-stockholm-more-...

Alaska, North Dakota, Minnesota, and Maine all get weather colder than Sweden. Other northern states hover at around Sweden-ish levels during winter, varying based on elevation and lake effect.

We just don't really give much of a shit about our people or infrastructure. So we all just make do. When my power goes out every winter, my furnace doesn't work, so I break out the portable propane heater, sleeping bag, extra blankets, winter gear. I keep telling myself I'll buy a generator, but the expense and complexity aren't really worth it for a few days. A bunch of my neighbors have wood stoves.

After the 2005 Cyclone Gudrun in Sweden (730,000 households without power) there was a huge push to make power and telecoms infrastructure more resilient by burying it underground https://en.wikipedia.org/wiki/Cyclone_Gudrun

If power to 3% of the population was lost around here, it would be considered a heads-will-roll catastrophe.

If I understand right, the largest power failure in the UK (not counting the coal miner strikes) was the 1987 storm, and that didn't reach 3% in the UK*, more like 1.15%. I can't find exact figures for the percentage affected by the 2006 European blackout, but assuming the region boundaries and population hasn't shifted too much since then, that was also a bit less than 3%.

* nor did it reach 3% in France, which was affected worse; but a quick search didn't give me a list of French power failures to compare sizes of.

Remember that 3% of Georgia is not 3% of the entire U.S.

And? 3% of the UK isn't 3% of Europe.

Georgia is big enough that percentages is the right measure. If this had been Rhode Island or Hawaii then law of large numbers no longer applies and correlation of failures means that seeing even 30% affected isn't crazy. But Georgia is 11 million people, doesn't get that excuse, 3% is still a lot.

You said that the 1987 outage didn't reach 3% of the UK. The 2025 outage in Georgia didn't reach 3% of the US, either. You compared a state with a major power outage to multiple countries, as though it were a meaningful comparison—I don't see how it is.

US states may not be sovreign like EU states are, but they're similar in a lot of other ways.

Cool. 100% of Italy lost power in 2003. That's over 50 million people.

Resulting in a major investigation, whose conclusions were acted upon to try to ensure it didn't happen again.

If you sit on your backside and say "eh, not too bad, these things happen", then you normalise failure, and things keep getting worse: https://en.wikipedia.org/wiki/Normalization_of_deviance

Agreed. On that note London is also big enough that percentages are the right measure. Here is 7% of London being knocked out.

https://en.wikipedia.org/wiki/2003_London_blackout

100,000 is a lot of homes... they say there are a lot of people on earth today but it's only 6-7% of the total of people who ever lived :)

That poor dude in Maine

Wait, you guys have power outages in the US?

With our aging infrastructure and orange man administration trying their damn best to dismantle the federal government in preparation for billionaire kleptocracy to extract as much value from USA. I don’t see it getting any better.

Now we know where Elon’s xAI data centers are.

Looks like someone in Atlanta tried to turn on an RTX 5090. /s

Reply on news.ycombinator.com