Hacker News Reader

Comment by brudgers | Hacker News Reader

There's a reason that smoke detectors operate the way they do. That reason is that people die when they don't.

The issue I have with the Nest smoke detector is that it is sold as a luxury consumer product. That puts it in the class of "the customer is always right" sales. The Venn diagram of that and things that are designed to save people's lives regardless of their tendency to statistically evaluate the risks and rewards of rare but deadly events inaccurately is disjoint.

If Nest did not comply with the regulations, they could not market their product as a smoke alarm.

View on HN

Actually, people die when they incapacitate their smoke detectors because of false alarms that can't be silenced any other way.

Citation: http://www.nfpa.org/research/reports-and-statistics/fire-saf... which says that 60% of people who died in a housefire were in a building that had no operating smoke alarms. A few more quotes:

"Surprisingly, the death rate was much higher in fires in which a smoke alarm was present but did not operate than it was in home fires with no smoke alarms at all."

"Smoke alarm failures usually result from missing, disconnected, or dead batteries. People are most likely to remove or disconnect batteries because of nuisance activations."

> people die when they incapacitate their smoke detectors because of false alarms that can't be silenced any other way.

Not sure nest would win if it was easier to disable. Have you ever had a smoke alarm that goes off every couple hours? You can't sleep. Doesn't matter if you can turn it off with a wave.

And let's say you reach the point where you instinctively wave so it doesn't bother your sleep, now what happens when there's a fire?

False alarms are the problem. From the video and reports, it appears the Nest Protect has real problems with this.

Pulled the battery from my kitchen fire alarm after it went off three nights in a row while making dinner. Nothing was smoking or even boiling, much less actually burning. It's an electric stove to boot.

You are contradicting yourself, if the alarm was triggered when making dinner then it must have generated a smoke, how would the detector even know you are making dinner.

With the nest, the alarm turned by itself for no reason, you don't cook in bedroom, he even said that he got notification when he was at work.

Two different reasons, your situation was that smoke alarm was too sensitive, perhaps you placed it close to the oven, here the alarm fires for no reason, and it is not the kitchen one, but bedroom one where normally you don't have smoke.

The alarm was triggering during a non-alarm situation.

That is, in fact, a false alarm.

The behavior described by cowsandmilk is highly predictable: provide false alarms, and people will almost always disable or bypass the alarm.

Yes. Incapacitated smoke alarms are a contributing factor in fire deaths. Hence smoke alarms are deliberately difficult to incapacitate.

Be aware that battery powered smoke detectors are a retrofit, and do not meet building code requirements. They are better than nothing, but failure due to disconnected or missing batteries is already in the realm of worse than the worst thing that can legally be built today.

You're conflating "incapacitate" with "temporarily silence".

They're two separate concepts. We should make it really easy to temporarily silence a smoke alarm (for, say, 15m or 4h), so that people almost never have to fall back on incapacitation.

And mandating wired ones does not address the problem. People still incapacitate them, by unplugging them. More generally, people will always find a way to silence a smoke alarm, and the regulators' goal should be to ensure that they do it in a temporary way.

My wife has silenced three smoke detectors by physically smashing them from the ceiling with a broom handle. It's a bad idea to persistently disturb her sleep.

That's what happens when you can't easily silence a smoke detector. Needless to say, after this treatment they didn't work so well - the only good thing about such a design is that it sells more smoke detectors. For a long time we simply didn't have any that worked though. We do now, but none of our replacement ones are linked together anymore - too much of a nuisance to replace the wiring she ripped out while silencing them.

Is there a link to the regulation that requires this? I was under the impression that most states require a hush/silence button on new smoke alarms. False alarms are inevitable and a hush button reduces the temptation to permanently disable the device.

Underwriters Labs make you pay to see that.

I can show you the table of contents, though: http://ulstandardsinfonet.ul.com/tocs/tocs.asp?doc=s&fn=0217...

But... every smoke detector I've bought in the last decade has had silence button. So how it is against any regulation?

According to a quick google, all smoke detectors in Maryland are required to have a hush function by 2018. So... pretty sure that isn't prohibited by regulation there. FEMA says the best smoke detectors have a temporary silence button. I guess maybe there are some states that prohibit it? Seems awfully foolish though: temporary hush buttons save lives.

If the device detects a sufficiently-high level of smoke, the button is ignored.

Again with this statement, accompanied by absolutely no citation.

I posted a link to the citation in the root comment of this thread.

Sorry; I meant a citation to the actual law people are referring to, the one that says smoke detectors shall not be capable of being manually silenced in the presence of smoke.

Smoke detector user manuals might vaguely allude to such a law to justify crappy engineering, but that's not the same as a citation to specific Federal, state, or municipal codes, or even an insurance underwriting code.

The actual law is "Do what UL says". Or, to use a direct quote:

"The U.S. model codes and installation standards require smoke alarms and smoke detectors to be listed in accordance with the Standard for Single- and Multiple-Station Smoke Alarms, UL 217, and the Standard for Smoke Detectors for Fire Alarm Signaling Systems, UL 268."

http://osfm.fire.ca.gov/firelifesafety/pdf/Smoke%20Alarm%20T...

I'm missing the part that justifies what we see in that video. He's unable to mute the detectors, but there seems to be a distinct lack of smoke.

Why are people carrying water for this idiocy? Nothing in the document you linked has anything to do with the subject at hand. Fine, whatever, you're not supposed to be able to mute the detector in the presence of heavy smoke. But there is absolutely no prohibition against a mute feature in general.

I found a citation that I think even you will accept:

https://community.nest.com/message/36003

True, and the 4% business was also mentioned in the summary document earlier. That doesn't change my two underlying points: (1) it's an incredibly stupid regulation that seems to have been designed to cause people to remove the battery or otherwise disable their smoke detectors permanently to avoid nuisance alarms; and (2) there was no smoke at any level in this scenario.

Poor engineering on Nest's part led to a situation where a poorly-thought-out regulation came into play. Neither condition is defensible... yet look at all of the people in this thread trying to defend one or both.

(Yes, I'm moving the goalposts to a certain extent, as I was mistaken about there being no law against silencing a smoke detector in the general case.)

Worse, municipalities are starting to pass stricter code, requiring the battery to be unremovable:

http://rules.cityofnewyork.us/tags/smoke-alarm

Since I can't actually see the document, does it say that the fire alarm can't be muted even in heavy smoke? Or are you just guessing that it might? I would find that surprising.

I don't know. UL doesn't make their rules openly-available. You have to pay them to read them.

But Nest's website says UL's rules forbid the silencing, and I see no reason to think Nest would lie about it.

I found the relevant citation: https://news.ycombinator.com/item?id=9064731

I recently purchased a non-Nest smoke alarm. The documentation explained that if there was an alarm, the hush button would silence it temporarily, but if there was a high level of smoke, the hush button would not work and the alarm would not be able to be silenced.

Since both Nest and other smoke alarm companies say this, I conclude there is some regulation to this effect.

The Nest device in the video did not appear to be experiencing high levels of smoke.

Found a citation: https://news.ycombinator.com/item?id=9064731

Yes, this is infuriating. It happens on my networked system of smoke alarms. It ignores the hush for nearly a minute while the smoke clears out. Ridiculous. Makes me want to just disconnect all of them.

UL is not a lawmaking body, so if they have regulations against being able to hush a smoke detector -- and I highly doubt that -- Nest is technically under no obligation to obey them.

In effect, they are. Many regulations reference their standards, require devices meet a specific UL standard, or recommend that devices be stamped with a UL approval.

Here's the gotcha though...

A smoke detector company sells a safety product, if that safety product fails they could easily be sued for the lifetime earnings of a dead individual (or even family in some cases).

So that requires that these companies have insurance, and in order for insurance to cover them they must meet certain insurance dictated requirements, which can be anything, but UL specifications are an easy thing to point at (to both protect the company from liability and insurance from payouts).

So while in some locations there might be no government regs, insurance is making regulations all on their own, and these companies can either play along OR risk paying out up to millions of they get sued.

Luxury consumer product? Is that way of saying, you must be smart if you bought our product? It does seem to have some pretentiousness built into the marketing.

I prefer KISS when it comes to my life and smoke alarms are not something I am going to fool with. With regards to their thermostats, I rarely am in the rooms my house has thermostats but programmables were cheap and easy to setup.

> Luxury consumer product? Is that way of saying, you must be smart if you bought our product? It does seem to have some pretentiousness built into the marketing.

It says no such thing. You're just putting your own biases on it. Luxury means: "the state of great comfort and extravagant living."

Nest products are premium products. They cost more and you're meant to get a better quality of product out the other end.

Nobody thinks they're "smarter" because they buy luxury, although they might try to justify their purchase on the grounds of its USP.

Reply on news.ycombinator.com