FEMA and FCC Plan Nationwide Emergency Alert Test for Oct. 4, 2023 | Hacker News Reader

FEMA and FCC Plan Nationwide Emergency Alert Test for Oct. 4, 2023

fema.gov

65 pointsBender53 comments

We had a national test in Norway that mostly worked OK a few months ago. But then they wanted to use it in a real emergency in early August(some specific areas were about to be flooded due to heavy rain) and that didn’t work as well. First of all cell phones way out of the flood area were getting emergency messages, some of these got several messages in different languages and different wording over several days. It was very strange. I’m way out of the flood area and got 4 messages during 25 hours. But my wife got 0. We have the same phone and same provider. A long way to say that this system should be tested in a “production” setting as well.

Well we did recently have a pretty spectacular failure of the emergency alert system in Hawaii.

Bureaucratic failure though

Test messages are spam.

After the pandemic, I think we’re quite a number to not trust the public services (either act too late, or try to revive the panic when it’s over). When I received the alert test messages, which, I remind, will not only be used to amplify terrorism scare but also for crowd management during flood, tell you to avoid protests, and then to tell us that we should urgently take the vaccine…

No thanks. Even one test message that was irrelevant to me was enough to preemptively block all scare levels them at the OS level (which is possible, at least in iOS in France).

It's the "boy who cried wolf" problem - test messages should begin with TEST TEST TEST and there should be an option to filter out test messages.

I blocked all emergency alerts until they finally gave you the ability to block just amber alerts.

Not just the word TEST but also flag it as a test. There is a field that differentiates test messages from real ones. Some cell phones have the option to enable/disable the test broadcasts. buried in the menus under emergency broadcasts

In this case however I believe they want everyone to receive it regardless of being a test. I believe S. Korea also just did this test including locations of nearby bunkers. The US may not have that many accessible bunkers.

The US has some bunkers but unless you are some high government official or just work at the bunker, you're not getting in. Some people have their own bunkers but it's incredibly rare. Many people that live in tornado-prone areas have tornado shelters. In many parts of the US, below-grade basements are not a thing. Half of my basement is below ground but unfortunately the part that isn't, faces nearby Washington DC making it pretty useless for that.

iOS already allows you to filter out test alerts and amber alerts, while still getting (for instance) tornado warnings.

The tornado warnings have gotten really annoying; it used to be they'd only give you a warning when there was a real chance (so 50% of the time you got a warning, a tornado touched down somewhere) - now it feels like anytime there's a thunderstorm, you get the alert.

> I blocked all emergency alerts until they finally gave you the ability to block just amber alerts.

Likewise. I still have a tendency to forget whenever I reset my phone or get a new one until inevitably the next amber alert will be at 3:30 AM on a night that I really need every minute of sleep.

There is no logic to amber alerts being delivered as screaming alarms, and everyone responsible for that behavior being the default should never be allowed to make meaningful decisions again. Normal notifications that respect silent and DND settings would be perfectly fine.

Did you receive those emergency texts (the topics you listed) or merely speculating what they are used for?

Why wouldn’t they be sent, after the system is in place? Think of the children that can be harmed by a deadly virus. Of course it’s a case of emergency to get everyone vaccinated. Have you ever seen the government be reasonable with a public resource?

I would think they wouldn’t just utilize the system so broadly or maliciously as you speculated. I have seen the government be reasonable with a public resource at least in US. Are there times when it is incorrectly used? Yes, but it is used far more reasonably than not.

I wonder how this works technically? and to what extent it taxes the bandwidth of the system / backhaul / towers / channels?

Is it:

Cell tower: "Every phone currently connected to this tower -- display this message now <emergency alert xyz123...>"

Cell tower: "subscriber 202-555-1212 known to be connected to this tower, display this message now <emergency alert xyz123...>" , "subscriber 202-555-1213 known to be connected to this tower, display this message now <emergency alert xyz123...>", etc...

AT&T central: "deliver to AT&T subscriber 202-555-1212 special text message <emergency alert xyz123...>", "deliver to AT&T subscriber 202-555-1213 special text message <emergency alert xyz123...>", etc...

FEMA central: "deliver to mobile subscriber 202-555-1212 special text message <emergency alert xyz123...>", "deliver to mobile subscriber 202-555-1213 special text message <emergency alert xyz123...>", etc...

Is it like a loudspeaker or a high priority special messaging channel?

I have no idea but I would guess there were choices about whether to optimize for speed of delivery versus guaranteeing that every single phone would get the message. And does the alert get stored for phones that weren't on? Or you only get it if your phone happened to be on and connected at the time of the alert or within 1 minute of it? How do they tell whether a phone received the command, or should try again?

It is probably using cell broadcasts: https://en.wikipedia.org/wiki/Cell_Broadcast

Ah, thanks that explains it -- (follow the link also for https://en.wikipedia.org/wiki/Wireless_Emergency_Alerts )

Fun tidbit, the UK until this year didn’t have a cellular broadcast system. During the pandemic the government worked with each mobile provider individually to deliver notices to every subscriber [0].

[0] - https://en.wikipedia.org/wiki/UK_Emergency_Alert_System#Coro...

And it took like a full day.

I've often wondered how they measure the efficacy of the tests. Is there some sort of telemetry on the phones that reports back that the user received and viewed the alert? Is there a sample group of test users who have agreed to report whether they received the alert?

Or does that even matter? Are they really more concerned with the infrastructure at the provider level?

The release also mentions that it's primarily intended for emergencies at the national level. I wonder what sorts of emergencies they're envisioning. Earthquakes, hurricanes, and other natural disasters can have wide impact, but are there any that really have nation-wide, proximate impact? Are they anticipating nuclear strikes?

Here's the test report for the 2021 EAS and WEA tests:

EAS: https://docs.fcc.gov/public/attachments/DOC-378861A1.pdf

WEA: https://docs.fcc.gov/public/attachments/DOC-378907A1.pdf

EAS is based on reporting (like a questionnaire) from EAS participants - so the parties that are supposed to receive and retransmit alerts.

WEA comes from two sources - the wireless carriers, and a survey of end-users (volunteers). In addition, the 2021 test was effectively opt-in - the upcoming test will actually target all cellphones (which was how the 2018 test was performed).

In the 2018 test report, they said they mostly got feedback through a variety of sources (direct public feedback, as well as information from surveys from different government agencies).

As for the actual useful purpose of national level alerts? I'd agree that outside of a nuke strike... not a lot. (In)famously, EAS was not activated on 9/11. Seems that a national level test is probably most useful as a way for the federal government to keep an eye on the total system.

Separately from the utility of the national test, the national test is required by law.

If you have a phone you don't want found, be sure to silence or turn it off that day.

Turn it off completely. It’s hard to test what combination or settings/silent modes will actually suppress an alert like this

This is a concern for people who are living in difficult circumstances, particularly with domestic violence, with companions who might steal valuables, or similar circumstances, where the test would reveal presence of a concealed phone.

Even if you're not directly affected, should share information about the test and possible impacts with those around you who might be, so that they can disable or move any devices elsewhere on the test date.

Funny FEMA should just now come up, I just now spent some time considering FEMA!

It is very surprising to me that much emergency infrastructure exists, without providing very basic things to households like regionally relevant emergency kits and something like one month of MRE + Water. Rather than it sitting in some warehouse... because distribution becomes a very immediate problem.

Not forced but, offered to anyone who needs it and to be made aware of it.

Seems like that would be a win considering how much they spend per year - almost $100 USD/yr per US citizen. Freaking 30B USD.

I suspect that if you distributed emergency supplies ahead of time, some people would treat it as a convenient handout and simply consume them.

And even aside from the temptation to use the supplies, not everyone would be able to find a place to store a month of rations & water.

Not to mention hurricanes and tornadoes will likely destroy the rations for those who need them most.

Right? Here's some MREs to store in your house that's getting washed out to sea in the hurricane or burned up in the wildfire!

Great planning work.

This is where awareness and planning help.

An emergency food stash that is part of a go-bag / bug-out-bag is useful. Individuals or households would have their food, clothing, and vital documents in a single piece of luggage (duffel bag, backpack, suitcase), which could be grabbed and carried or quickly thrown into a vehicle.

The challenge, at a wide-scale (national, regional, even city-wide) disaster-preparadness level, is of communicating this and establishing suitable levels of conformance.

To the extent that there are preparations which individuals can self-provide (food, water, and clothing among them) and those which require or benefit from central coordination (command-and-control, warning and information systems such as the FEMA test, emergency responders (rescue, etc.), and coordinated mass transportation (establishing and managing evacuation routes, busses, trains, etc.) it might not make sense to specifically provision such supplies, though if specific items aren't readily available commercially government coordination or distribution might help.

The best "emergency supplies" are consumed over time, to keep them fresh. If you're planning ahead, you just make sure that the consumables you have that can last a month, you have a month supply of.

And that is expensive (something like $1k per household, perhaps, based on https://www.costco.com/chef's-banquet-one-month-emergency-fo... ).

The US is amazing at transportation in general, and it makes more sense for each state to maintain emergency supplies they can move in a moment's notice.

But households should in general be prepared to survive about a week without outside assistance; if you're more wealthy you can increase that.

The DoD procures MREs at about 80 bucks a case of 12. You could probably stretch one MRE per day for survival situations (~1250 calories), though 1.5 or 2 would be much more comfortable. Taking it to be 1 per day, one month is 200 dollars.

MREs have a 3 year shelf life at room temperature, so each person you supplied would need to be topped off every 3 years - so like 65 bucks per person per year. So fits in their current budget, but would displace like... a lot of what they're currently doing.

One month worth is probably also not really worth it from a value perspective? Push down to 5 days of MREs + water per person, and you'd probably capture like 98% of the value.

> So fits in their current budget, but would displace like... a lot of what they're currently doing

Yeah, right? Like sure, we could just eliminate FEMA and instead use that budget to just hand out MREs. But then they wouldn't be doing their emergency management training and standardization that they do. They wouldn't do disaster housing assistance. They wouldn't be working on hazard mitigation activities.

There's a lot of things FEMA does that isn't just hand out some crackers after the storms pass.

Personally, if my house gets destroyed in a tornado I'd prefer FEMA to come through and cover a hotel for a couple weeks while we figure out what to do rather than just hand me some MREs a year ago that also got destroyed in the tornado.

MREs are not meant to be eaten for more than 21 consecutive days; even eating them for a couple of days will make the diner uncomfortably aware of why they've been given the nickname "Meals Refusing to Exit"

Great points.

Just thinking of MRE and water is narrow, but in my region the most likely broad critical event is total supply chain cut, so it would be a long term shelter in place situation. So I got to think for myself on that point.

For example people in Amish country aren't going to give a crap in the first place and have some of their own supply chains.

I am glad about the alert system test and think it's a great use of funds!

The recommended shelf life may be three years, but the meals can almost certainly be used long past that point, with only modest degradation.

I'd agree that a month's supply is excessive. Three days to a week would be far more reasonable in most cases. The goal is to tide over until other resources can be found, which in most cases will be a few days.

Where longer-term nutrition is required, far more conventional long-lived foods will be more appropriate: bulk grains, dried beans, canned goods, powdered or dried foods, and freeze-dried foods. These can come from normal household stocks, and simply be rotated as part of normal food purchase and preparation.

Refrigerated or frozen goods will of course spoil quickly without power in most cases (though could be viable for cold-weather climates where cold storage ... comes with the territory).

Not to put undue weight to YouTube's reliability, but there's at least one channel devoted to eating long-out-of-date MREs and other military emergency / survival rations. From as long ago as 1906 (117 years ago):

<https://yewtu.be/watch?v=562nQKR3_3M> (1h35m runtime, first bite at 16:24).

There are multiple selections from the 1940s (WWII era), and selections from numerous militaries (US, UK, Canada, Vietnam, China, Japan, Sweden, Israel, South Africa, ...).

<https://yewtu.be/channel/UC2I6Et1JkidnnbWgJFiMeHA>

And ... having developed a new hobby ...

Not all experiences are positive. A 1986 MRE prototype, tested in 2018 (32 years old at the time) is decidedly stale:

<https://yewtu.be/watch?v=qhCmUoDnzmk>

Watching a sampling of videos might give you an idea of what to watch for in terms of damaged / stale / degraded food. There's also good information on nutritional choices in emergency situations, including "metabolic water", the amount of water that's consumed or released in consumption of carbohydrates and fats (net positive) vs. protein (net negative), as well as caloric density by unit weight (4 kcal/g for protein and carbohydrate, 9 kcal/g for fats), vitamin, and fibre content.

A friend also reports that organisations often sell or auction out-of-date rations which can remain usable for some additional time.

With careful storage (cool, dark, dry place, low humidity, no degradation of packaging), emergency food should store well.

The items which would be useful for an emergency situation are largely readily available commercially, and/or depend on both local and personal circumstances. There are guidelines for packing a go-bag or bug-out bag, e.g.,

<https://www.state.gov/global-community-liaison-office/crisis...>

<https://www.outsideonline.com/outdoor-gear/how-to-build-the-...>

Many of the specific supplies that are useful are dual-purpose, and may include preserved or shelf-stable food, camping, recreational, or work-out supplies.

The most critical needs are usually fairly brief, the day or two before rescuers can reach people, or during which evacuations are ongoing. Those in more remote regions may want to keep longer-term stores, perhaps a few weeks worth of supplies, but will likely have much of this on hand already.

Keep in mind that a go-bag is to go, and if it weighs much over 25 kilos / 50 lb or so, becomes a challenge to movement itself.

In Canada each province has two tests scheduled each year, one in May and one in November:

* https://www.alertready.ca/testing-schedule/

I wonder if this will happen if there is a government shutdown.

Good. Need to do that every once in a while.

We had the same all the way out here in India. Wonder what do these governments know in order to coordinate like this.

Alternatively, it could be a case of:

"Oh, how long has it been since that emergency alert system was tested? Oh, that long? Maybe we should schedule a test to see it works"

In germany we now have a "Warntag" [literally "Warn day"], which is scheduled to be once per Year in September.

This is mostly a result of a previous test, which revealed that in case of a catastrophe, most germans would not be warned (Sirens not functioning, cell broadcast was not used, ...)

Sadly the website of our Federal Office for Civil Protection and Disaster Assistance is not translating every page to english. https://www.bbk.bund.de/DE/Warnung-Vorsorge/Bundesweiter-War...

In the Netherlands we test the systems every first monday of the month.

Also the alert on your mobile phone?

Seems the mobile version, NL-Alert, triggers first monday of every quarter

BC is similar [0], though it’s twice a year.

[0] - https://www2.gov.bc.ca/gov/content/safety/public-safety/emer...

Also: https://www.alertready.ca/testing-schedule/

>This is mostly a result of a previous test, which revealed that in case of a catastrophe, most germans would not be warned (Sirens not functioning, cell broadcast was not used, ...)

And failed yearly tests being ignored. It took the Ahr flood to actually do something about that.

https://www.tagesschau.de/inland/gesellschaft/katastrophensc...

San Francisco used to test their sirens weekly which was frequent enough that both the siren and the incoherent verbal bits just kinda blended into the background.

Right, the "familiarity breeds contempt" problem with testing alerts too often.

With sirens it would probably make sense to have a different but similar tone/sequence for tests vs production alerts. Everyone would instinctively know that something was wrong.

Which runs right into the problem that when there is a difference between test and production. production will probably fail. Some days you just can't win. Probably the best compromise would be monthly(I agree weekly might be too much) infrastructure test, with a full blown yearly test/holiday where everything goes to hell for a day as you test actual procedures.

SF's emergency siren system is being upgraded, not retired, though the project is delayed and over budget:

<https://www.sfchronicle.com/sf/bayarea/heatherknight/article...>

Nice documentary about how the Swiss system works:

https://youtu.be/zybgpvd3ExE

Reply on news.ycombinator.com