Ask Slashdot: After We're Gone, the Last Electrical Device Still Working?

Leomania writes: After watching a post-apocalyptic Sci-Fi short on YouTube (there are quite a few) and then having our robot vacuum take off and start working the room, I just wondered what would be the last electric/electronic device still functioning if humans were suddenly gone. I don't mean sitting there with no power but would work if the power came back on; rather, something continuously powered, doing the task it was designed for. Are we talking a few years, decades, or far longer?

Comments

A.I.?

[turkeydance]

one perspective.

Re:A.I.?

[Anonymous Coward]

It is funny that people believe that we have, or will have "A.I" soon. There has been no progress towards TRUE A.I since the 1970's. ZERO. What you think is "A.I" is not really A.I.

Hint: Siri is not A.I, or a form of A.I, or even a precursor of A.I.

Re:A.I.?

[penguinoid]

Exactly what I was going to say... the AI that exterminates mankind will keep working indefinitely.

Re:A.I.?

[jythie]

Spoken like someone who knows nothing about AI outside sci-fi.

There has been a ton of progress towards 'true' AI over the last 40 years, it just has not managed to produce one. Progress != Success, esp on such difficult problems.

Re:A.I.?

[dbIII]

Nailing down the definition isn't so easy so there isn't anything we can call A.I. yet outside of S.F.

Re:A.I.?

[Immerman]

I don't know - a lot of current weak AI may end up being sub-systems for a strong AI - so in that sense we may well be getting closer. The problem is we have no real idea what strong AI might actually entail, implementation wise, and so have essentially no idea what if any progress we're making in that direction. At the very least we've found a great many strategies that don't work, which is in fact it's own kind of progress.

Honestly though I'm happy with the current state of affairs - weak AI may be able to get us into a lot of trouble (market crashes due to HFT algorithms anyone?) but it's nothing compared to what a strong AI would be capable of.

Re:A.I.?

[Applehu Akbar]

Hint: You are a moron.

The AI goalposts keep shifting over the years as technology improves. Remember when it was claimed: "we'll have strong AI when computers can play chess?" Then it was: "We'll have AI when you can verbally tell a computer to do useful everyday tasks."

As we reach each milestone, we compare the state of the technology to our own human self-awareness and realize that it's time to move those goalposts agin.

Re:A.I.?

[jythie]

Depends on your perspective I guess. One of the other major things going on since then has come from the other direction, with a great deal being learned about existing non-human intelligence and how much more complicated and, well, real, than people had believed. So the simple metric of 'does it act like a human' has become a lot fuzzier when actually looked at.

Re:A.I.?

[Triklyn]

I think the truest definition of AI, the endearing one that we've feared ever since the inception of the concept, is something capable of supplanting us. Both the capacity and potential will to do it. It can't be a tool of ours, it must have its own agency.

Re:A.I.?

[mark-t]

The problem with the AI goalposts that you notice is because it hinges on a term which itself is not very solidly defined - Intelligence. Some would say animals aren't intelligent because they can't speak. Some would say that monkeys aren't intelligent because you can't teach one to play chess.

Fuck the term 'artificial'.... first define what intelligence is and a clearly defined way to identify it where it exists, one that is objective and unambiguous. Then, when something can do whatever intelligence does by something that is man-made (ie, "artificial"), then you have AI. Simple.

Re:A.I.?

[jythie]

I also work in the field, I see new research and discoveries every year. Scientists are constantly learning new things from the attempts and we are constantly constructing systems with greater flexibility and capability.

As for 'database with search capability', what do you think human intelligence is in the first place? It is not some magical ability, is a combination of various capabilities interacting with each other, with data storage and retrieval being a significant component.

Re:A.I.?

[StikyPad]

Remember when it was claimed: "we'll have strong AI when computers can play chess?"

No.

Then it was: "We'll have AI when you can verbally tell a computer to do useful everyday tasks."

No it wasn't.

As we reach each milestone, we compare the state of the technology to our own human self-awareness and realize that it's time to move those goalposts agin.

It's completely the opposite, actually: Our expectations of AI exceeded the capabilities of even the most intelligent humans: "Understand what I mean, not what I say/write/do." We haven't had to move the goalposts because they're farther than anyone can kick.

Re:A.I.?

[g0bshiTe]

Considering that is how the brain pretty much works, you have neural pathways that develop, in which you decision making is based. In a sense your brain is little more than a db with pathways (read queries) hardcoded. if heat SELECT * FROM burn WHERE heat = hot returned 1 don't touch

Re:A.I.?

[g0bshiTe]

If it has a sense of self preservation though it's instinct it's aware enough to attempt to survive.

While it may not be as self aware as "I think therefore I am", it's self aware enough to know "DANGER WIL ROBINSON".

Re:A.I.?

[Falos]

Lots of semantic discussion at this tier. Subtiers. Some of the posts address AI as some linear strength metric that qualifies after some arbitrary point. That's a stupid definition. Better would be to pin it to some tipping point; not the singularity, but self-writing or something. AI isn't deterministic. Chess is deterministic. "Natural" voice navigation will be deterministic.

AI isn't even deductive. AI means it builds new decision trees, adaptive to conditions. Most life forms do it. In this context, humans boil down to execution of instinct code, all the way up to "love (and the expression of) is chemicals in the brain".

Now, we already have code that does this, at a crude level. At the time, Black & White set some niche records for the amount of behavioral code your Creature pet would build for itself. We have programs that don't know how to play chess, but can write a chess-playing decision tree that optimizes over iterations. It will reach a shitty roof well below current, human-written layouts. Programs that "learn" to play Space Invaders (or anything) play like shit.

So, seeing as you read this far, I'll concede that we hit a familiar snag - either we already have AI (of this definition) or we're pegging AI to some arbitrary strength of original, generated thought. But the code equivalent to the instincts of even simple life forms would be a mess, so it'll be a while before self-preserving, self-sustaining, "self-aware" (enjoy ur semaniks.) AI.

Re:A.I.?

[Aighearach]

As long as you can give up on the "artificial" and "intelligence" parts, the rest of it there are solid working academic definitions for.

As to the Question, my prediction is that the LED status panel on a remote solar power installation somewhere will still be functioning hundreds of years after everything else.

Re:A.I.?

[Aighearach]

"I haven't heard of anything new, therefore there is nothing."

What a maroon. I expect more learned intelligence than that, even from an anonymous cowherd.

You might be interested to learn what Watson is, since you claim to be in the field but never read even a basic technical description of the project.

Search, interestingly, is the hardest part of human intelligence too. Almost everything gets stored up there somewhere, but a difference in recall can be the difference between 65 and 150.

Re:A.I.?

[Aighearach]

Human self-awareness isn't a goal of the real research. Better tools for humans is the goal, and the successes can be understood in that light also.

Don't confuse what some pure academic says or wants or speculates with being what the whole "field" is doing, because the people doing experimental work are the ones actually doing anything. And that is driven largely by money from toolmakers, for example IBM.

Research of flight didn't stop at any of the early mileposts, either. The idea that reaching a milepost somehow results in "oh we're there, we need to invent a new name and pretend we're a new field so we can research the next step" is just insane. A common idea, to be sure. Some sort of video-game thinking, where the goals are confused with the checkpoints.

Re:A.I.?

[HuguesT]

Strong words maybe.

My opinion is that AI and more generally CS research allows us to better define what "intelligence" is and isn't. Also, allows us to realize that whatever is between both our ears is still mysterious. The debate is still open on whether we will be one day able to replicate it, and if we should.

Re:A.I.?

[Aighearach]

It is not so simple that the intelligence would then be artificial; the device is artificial, but the intelligence is still human intelligence, translated into a complex tool.

Don't think just because you solve the real problems that the arguments over definitions go away. Defining intelligence is harder than that. There are already lots of competing definitions. Arguing over definitions is for philosophy and English majors, not AI majors. It probably won't be seriously attempted, even while it remains the major focus of journalists. (English and pol-sci majors)

Animals can indeed speak, that is recycled propaganda that was refuted long, long ago.

As a tournament chess player I can tell you, most of the humans aren't any good at chess either. As far as chess ratings go, many monkeys would have a similar rating to the average human, e.g. unable to win a single game against the lowest rated player at a tournament. If we had a "chess in the zoos" program to teach chess to baby primates, I'll bet we'd have some good orangutan players. Adult humans can barely learn the moves, but even if they study for decades few are capable of reaching the level of play than an 8 year old reaches in a couple months, if you can get them to put a couple months of focus in.

Re:A.I.?

[Aighearach]

Who said it?! That talking head on the TV. What are yousaying, that she didn't talk to real AI researchers for the descriptions of the field?! That she sat down with her editor and they made up whichever words sounded interesting?! Shocking, truly shocking.

Maybe when AI gets good enough, I can get an expert system that turns on a red warning light whenever there is media drivel on the TV. Oh, wait, that's just called a closed circuit, not AI.

Re:A.I.?

[mark-t]

An artificial reproduction of some kind of intelligence, whether or not that intelligence exists naturally, is by its very definition artificial intelligence, because it is man-made. Something that is artificial is not inherently any less of a genuine article than something that it is not... it is just not something that was produced by nature. The notion that an artificial intelligence might someday be utterly indistinguishable from natural intelligence does not alter that.

Also, I suggested that the entire reason we are dilly-dallying over whether we are getting any closer to AI is because the word 'intelligence' isn't very well defined in the first place. In many cases one person's idea of intelligence simply amounts to another person's idea of a parlor trick. We need a line to be drawn somewhere because without one, we will be arguing forever... even if we were to achieve it, we would never know because we are too busy arguing over what constitutes it.

Re:A.I.?

[Aighearach]

There is a big hole there, though

1. not actual intelligence
2. man-made
3. ...
4. man-made _intelligence_

Being man-made doesn't make it intelligence, or insulate it from being a man-made tool that appears clever because the builder was intelligent. If we don't have a good definition of intelligence, it is basically impossible to make an objective claim that the intelligence is "in" the system, and not just something that had to be present externally to design the system.

Luckily, none of that impacts the attempts of natural philosophers, and later scientists, to encode human intelligence into machines in order to better complete human tasks.

Re:A.I.?

[Archangel Michael]

I'll suggest an AI test that should be sufficient to solve the question of AI.

1) AI is self aware
2) AI is capable of learning something* it has never seen before. e.g learning to play a newly invented game
3) AI is capable of doing many complex and unrelated tasks at once. e.g. Playing Chess while having a conversation and learning something new ...
4) AI is capable of creating its own interpretations of things like art, literature, philosophy and even religion. e.g. Starry Night critique.

Re:A.I.?

[mark-t]

Being man-made is what makes it "artificial"... whether you want to call "it" intelligent or not. If it is considered intelligent, then by definition, it is AI.

And as I said above... what can appear as intelligence to one person can be another person's idea of coincidence or just a smoke-and-mirrors parlor trick.

Re:A.I.?

[Anne Thwacks]

There is a decent chance we will all be killed by a super-intelligent AI some day.

However, by the same definition of decent, there is a decent chance we will be killed by teenage mutant ninja turtles first.

Re:A.I.?

[afidel]

my prediction is that the LED status panel on a remote solar power installation somewhere will still be functioning hundreds of years

Only if someone makes one with non-RoS solder and uses solid electrolyte capacitors or only glass/mica dielectric capacitors since there's no way regular electrolytic capacitors will survive anywhere near that long.

Re: A.I.?

[saloomy]

A.I. Is like the uncanny valley in design and facial recognition, applied to decision and understanding. I feel bad for researchers who work in the field and suggest they rename it to something else. People will seemingly forever move the goal posts on intelligence forever. Chess-playing computers, computers that can have a conversation, play jeopardy or answer questions like google search or wolfram alpha, analyze videos like the predator system, recognize songs like shazam, spell check better than most, and soon, drive us around are all intelligent. But we hate to believe that machines are as or more intelligent than we are, so we move the goal posts. The reason is... Intelligence is like magic. Once it's understood, it's no longer special, or magical, the concepts become easy, well understood, and the method puts the skill in "well that's obvious and isn't AI". Take whatever you want to call AI functionality, wait a decade and read it back to yourself in 10 years. We will have passed it, but the definition you will use then will change. To answer the poster's question... Satellites in high stable orbits come to mind, the rover opportunity, maybe some soon-to-be drone like the one flying in interstellar? For certain anything solar powered has a good chance, especially if it's in infra-stellar space. I bet the moon-buggy still works. I think they recently sold the rights to it. It's not turned on, but it doesn't have to be to be called "working", if your definition is "in a functional state".

Re:A.I.?

[Immerman]

Who says a "true AI" has to match the way the human brain works? It has to get at least vaguely similar results - but that's likely an entirely independent criteria. And in point of fact, and from what little I know of neuroscience, your description bears absolutely no resemblance to the way the human brain works.

Re:A.I.?

[blippo]

No, I don't think so.

We have improved our understanding of "the problem."

A lot of things that we thought was hard, was based on the fact that it's hard for a human.
There are a lot of progress on the things that we think is "simple", such as looking at a spoon,
and classifying it as a spoon, moving and navigating, etc, but it's almost nothing compared to
the simplest animal, even insects.

As an example: The ratslam algorithm is really interesting, and a rather neat and surprisingly simple algorithm for location awareness
based on biological processes. Still not good enough for vacuuming my house though.

Pure "logic" in a box is "solved" but useless since in reality we need fuzzy-logic and semantics and human-like analysis to clean up the data before it's analysed -which is hard. I think that Watson and other expert systems are getting fuzzier, but there is along way to go there.

Re: A.I.?

[iamacat]

Actually there have been tons of progress, it's just in the areas where there is a practical need rather than what looks cool in a Sci-fi movie. There is no need to clone a human as a robot, we already have lots of humans. On the other hand, lots of work is being done on taking an image and automatically generating a short description of what it shows so people can find it among a billion others.

Re:A.I.?

[brantondaveperson]

But this does very much depend on what one might mean by "super intelligent". I don't think we're in danger of ever producing a human-like intelligence, but I don't think the AI-kills-us-all scenario really requires one. Just some robots that are "intelligent" enough to co-operate, fly around, recognise us from the air, and kill us.

They're already building them. I'm not certain I see it ending well.

satellites

[Anonymous Coward]

Probably satellites would last the longest, with maybe Pioneer or Voyager probes for however the RT batteries last.

Re:Satellites

[Anonymous Coward]

Nope, the orbits decay.

I'm not sure of the rate, but likely hundreds of years, not thousands.

Re:satellites

[AmiMoJo]

A Nokia 3310 will outlast them all.

Re:Satellites

[MarkWegman]

The orbit decays and ceases to be geosynced. But it's got a long way to go before it hits the atmosphere and burns up. Remember how much energy was needed to get it up there from low earth orbit. The satellite has to give up all of that because of the tidal effects of the moon before it's close enough to be slowed by the atmosphere. There may also be a very minor effect from the sun's radiation. I think it's safe to assume that the solar cells will deteriorate before the decay of the orbit causes a problem.

Re:satellites

[alex67500]

Only if it's sitting on its charger... It only had a week's worth of battery. Which is massive by today's standards, but still.

Re:Satellites

[nanoflower]

Okay, since you say satellites are out of the question what do you think will still be working after the last satellite falls out of the sky. Keep in mind we are talking hundreds of years (using your own words.)

Re:Satellites

[ldobehardcore]

The 10,000 year clock? If they ever build it.... Although, there's certainly a chance that a satellite sufficiently high enough in altitude with durable solar panels etc. would stay in orbit much longer than 10,000 years, if not functional due to it's batteries going dead. Perhaps there's already a satellite up there that will turn back on if it's panels are exposed to sunlight even if it's batteries are dead. I wouldn't know.

Re:Satellites

[msauve]

Not only are you critical of reasonable answers, but your own answer doesn't even match the question. The 10K clock is powered mechanically, not electrically.

Re:Satellites

[ldobehardcore]

I guess I wasn't paying enough attention... I didn't mean to be critical, just trying to contribute. I'm starting to regret I even said anything.

Re:satellites

[Dereck1701]

Running out of mono-propellent probably wouldn't be an issue because satellites simply don't attempt those kind of maneuvers without human input (at least I'm pretty sure), while that would definitely be a problem for LEO and Geostationary craft between those two orbits or are further out don't need regular boosts to stay going. And while reaction wheels do go bad pretty quickly (a decade or so) they aren't necessarily needed for a spacecraft to remain "operational". Sure without them they usually can't continue their primary mission but even if the craft has directional solar arrays slow rotation will give likely give them enough power for the satellite to remain in standby mode for decades, possibly a century.

Re:satellites

[hackertourist]

The Voyager RTGs are decaying, NASA expects output power to drop below the point where it can keep a single instrument going around 2025.
The Pioneers are already long past the point where they can't send a strong enough signal to be detected.
The latest nuclear power plants for the US Navy have been designed to run without refueling for the life of the ship. That's 50 years for aircraft carriers, so the USS Gerald R Ford (CVN 78) is capable of functioning until 2065. Now I don't know how stable a nuclear power plant is when left on its own, but potentially this'll live much longer than the Voyagers.

Re:Satellites

[Minupla]

How about the SOHO satellite? A L1 orbit should stay pretty stable even without any further assistance.

http://sohowww.nascom.nasa.gov...

Min

Re:Satellites

[Anonymous Coward]

I'm starting to regret I even said anything.

Welcome to the Internet.

Re:satellites

[Lumpy]

God I miss the days of good cellphones. My Nokia N82 was epic in battery life. Back when Nokia made the absolute best cellphones in the solar system.

Re:Satellites

[Lumpy]

Solar powered craft at L1 or L2 can probably go for a few milennia.

Re:satellites

[Antique Geekmeister]

The Volkswagen Beetle from the Woody Allen movie, "Sleeper"

        http://www.tin.org/bin/man.cgi...

More realistically, some chemical batteries, such as good lead-acid batteries in cool, dry climates will retain a slight charge for years. But they all have a notable self-discharge rate of at least a few percent a month. The notable exception among battery technologies seems to be this:

        http://en.wikipedia.org/wiki/O...

That device has been running off its battery, at an extremely low rate, since 1840. The bell is much softer now, but it shows no signs of failing.

Re:Satellites

[AikonMGB]

Solar cells will degrade to the point they can't supply keep-alive power to the spacecraft; batteries will degrade to the point they can't sustain the spacecraft through eclipse season; electronics will accumulate more and more total ionizing dose, single event upsets and latchups will become more and more frequent, and things will basically stop working. I don't think anything we've launched will come within an order of magnitude of a millennium.

Re:Satellites

[AikonMGB]

Sorry you said L1 and L2; ignore eclipse season comment.

Re:Satellites

[AikonMGB]

L1, L2, and L3 are weakly stable; think being at the top of a parabola. It doesn't take much effort to keep yourself there, but you do have to reject orbit perturbations. L4 and L5 on the other hand are actually stable, which is why trojans collect there. Note that there aren't any natural equivalents to trojans at L1, L2, and L3.

Re:satellites

[Idarubicin]

One should be very wary of the distinction between "run without refueling" and "run without regular maintenance". Even assuming that the reactor's fuel would last, the ancillary equipment associated with the reactor's operation (coolant pumps and such) and electricity generation (steam turbines) certainly wouldn't be expected to operate unattended and unmaintained for months, let alone years.

That said, the fifty-year planned lifespan of the Nimitz-class includes, if I'm not mistaken, a mid-life refuelling and complex overhaul (RCOH). To be fair, the reactor's fuel would likely last longer than the planned 20-25 years if the carrier weren't actively steaming--but I wouldn't trust the other parts to last anywhere near so long.

Re:Satellites

[Antique Geekmeister]

Even L1 and L2 lunar orbits are not that stable without attention. Solar forces, such as light pressure and solar wind, tend to destabilize them over time and require station keeping, especially as satellites have low mass and large solar panels.

There are other difficulties for satellites. The increased radiation of all types and even magnetic effects from the Van Allen belts can be very hard on circuitry. Adding shielding is very expensive, and over time even the shielding can become slightly radioactive itself. So I'd be very surprised indeed if any of our current satellites lasts even a single human generation.

Re:satellites

[Z00L00K]

Not in Geostationary orbit. They will be there "forever", they may still work for a long time, but eventually debris impact and radiation will kill them.

Re:Satellites

[Z00L00K]

Geostationary satellites collects in orbit at some places quite some distance out, and can be there for a very long time. Moon pull and solar wind may impact them, but it can take a long time before they leave their positions.

Satellites at the Lagrange points are probably going to stick around even longer.

Re:satellites

[Charcharodon]

There is no reason your phone couldn't last a week now. Just need a battery pack the size of a deck of cards attached to it.

Re:Satellites

[RivenAleem]

What about them garden lamps that are solar powered? They charge during the day and give off light at night.

Re:Satellites

[Headw1nd]

You have a grandiose definition of "a while" if hundreds of years doesn't make the cut.

Re:satellites

[jythie]

I wonder if they would outlast terrestrial RT battery powered devices like autonomous lighthouses.

Re:satellites

[XxtraLarGe]

God I miss the days of good cellphones. My Nokia N82 was epic in battery life. Back when Nokia made the absolute best cellphones in the solar system.

I highly doubt it. If you were serious, you'd still have one, since there are still *plenty* of old-style phones available. Would you really trade whatever you have now for one of these?

Re:Satellites

[dbIII]

In geostationary they end up bunching up over Indonesia a few years after they lose station keeping ability. Maybe some will run into each other or maybe not before they end up in different orbits.

I think it's safe to assume that the solar cells will deteriorate before

There's still photoelectric light meters from the 1950s in working order with the original cells. Unless the things get hot up there how are they going to wear out?

Re:Satellites

[dbIII]

What's the failure mode of those solar cells?

Re:Satellites

[dbIII]

Capacitors are the go in such devices these days. I don't know how long the current solid ones last.

Re:Satellites

[AikonMGB]

I don't know the underlying physics, but radiation (including solar radiation), especially at high temperature, causes the junctions in the solar cells to become less efficient over time. Less efficient cells generate more heat, which increases the rate of deterioration. Eventually the open-circuit voltage of the cell drops so low that it is below your spacecraft power bus and you stop being able to pull power off the array.

Another failure mode is when individual cells short-out, which happens when the junctions just straight up burn out. This generally results in the loss of individual cells, which lowers the voltage of the array or, depending on architecture, may take down that cell's entire string.

Re:Satellites

[dbIII]

Sounds like diffusion. Hence those 1950s photocells still working since it's sort of related to melting point, as in stuff with high melting points like silicon has things diffuse through it slowly a long way from the melting point. At least that's the failure mode I know, which is going to be decades to centuries.

which happens when the junctions just straight up burn out

Sorry to be difficult, but how? It's not exactly going to arc at such low voltages.

Re:Satellites

[Minupla]

Maybe the graveyard solar orbit of something like WMAP then... http://map.gsfc.nasa.gov/missi...

Min

Re:satellites

[Immerman]

So charge it once a week - nice easy schedule to remember. I have my Samsung flip-phone give an "charge me" alarm every Wednesday evening, and if I still forget it'll probably last until the next Wednesday anyway.

Re:Satellites

[AikonMGB]

Local overheating, hard radiation hits; voltages can actually be pretty high depending on string length and your orbit. Spacecraft experience high differential charging depending on the plasma environment through which they are passing. Outright cell failure is relatively rare though; typical failure mode is degradation below the operable voltage.

Re:satellites

[Bing Tsher E]

I have one of those LG flip phones showed on that link, not sure which of them, it's one I got for $4.95 at Walgreens. It's not my only 'phone' it's just the only one that I pay for phone time on. My small pocket sized wifi cellular router only costs $45 for 3 GB no matter how long it takes to use that 3 GB (no time span requirement, just however long it takes to use the traffic amount), so I use it alongside a non-activated Android phone for my pocket Internet needs.

Yes, I carry three mobile devices in my pocket. I'm a nerd, not a wanna.

Re:Satellites

[Rei]

LEDs have lifespans of what, 50-100k hours? So maybe a couple decades. And some will significantly outlive their design life, as is always the case with failure curves. The solar cells should be good for decades, until the contacts corrode.

One *could* design devices to last for thousands of years. But that's not usually a design constraint ;)

Re:Satellites

[meta-monkey]

I'm starting to regret I even said anything.

Story of my life.

Re:satellites

[Strider-]

Either my understanding of orbital mechanics is completely wrong or that is completely incorrect. Geostationary satellites need very regular station keeping otherwise they either fall to Earth or are ejected out into solar orbit. If ejected it could remain operational for a while but if it fell back to earth the results would be obvious.

Your understanding of orbital mechanics is totally wrong. Geostationary satellites do need frequent stationkeeping maneuvers, but that is because the satellite is required to remain in a 30km box. If these maneuvers cease, as would happen with the sudden disappearance of humans, they will start to drift off their stations, eventually collecting in a couple of regions, one over the Indian ocean and the other over the Pacific. (This is due to the earth's slightly uneven gravity). Because of the vastness of space, the probability of them actually running into each other is fairly low.

A geostationary satellite would need almost the same amount of energy to come down as it takes to put it up there, and probably twice as much to escape the earth's gravity well. At the end of life of these satellites, they use the remaining fuel to boost them another 200km or so in altitude, then vent all remaining fuel (so they won't explode if there's a fuel leak), and then blow the electronics to make sure they don't interfere with anything else. They will remain in that graveyard orbit forever.

Re:satellites

[Enigma2175]

Either my understanding of orbital mechanics is completely wrong or that is completely incorrect.

It's the former. Hint: the moon is beyond geosync distance but somehow it manages to stay in orbit. The "balanced" orbit you are thinking of may be the Lagrange Points, where the gravity of a body balances with the gravity of another body.

Re:Satellites

[chihowa]

The GPS satellites will continue to function for a while, but the positioning system itself depends on the satellites being tracked and managed by ground stations. It's not an autonomous system.

Also, they're not in geosynchronous orbits.

Re:Satellites

[chihowa]

The contacts will probably be the first to fail, followed by the storage capacitors. The panels are generally overspecced, and fail by degradation, so they'll likely outlive the LEDs. It's kind of depressing that such a simple and solid state device would fail so quickly.

Re:satellites

[dj245]

One should be very wary of the distinction between "run without refueling" and "run without regular maintenance". Even assuming that the reactor's fuel would last, the ancillary equipment associated with the reactor's operation (coolant pumps and such) and electricity generation (steam turbines) certainly wouldn't be expected to operate unattended and unmaintained for months, let alone years.

That said, the fifty-year planned lifespan of the Nimitz-class includes, if I'm not mistaken, a mid-life refuelling and complex overhaul (RCOH). To be fair, the reactor's fuel would likely last longer than the planned 20-25 years if the carrier weren't actively steaming--but I wouldn't trust the other parts to last anywhere near so long.

As a steam turbine engineer, I am fairly confident that, given a well maintained system to start with, the first failure would probably be in a stuck steam control valve. Over time, oxides build up on the valve stem, which would cause it to become stuck at some point. This would probably take 3-6 years. When that happened, the instrumentation control loop (need more steam, open valve, need less steam, close valve) would have a hiccup since it would ask for more or less demand and the valve wouldn't move. Valves stuck-open have historically caused many turbine overspeeds and resulting disasters.

Depending on exactly how the system was set up, the stuck steam valve should trigger the control system to automatically close a different valve, shutting down the plant. However, it is possible that it would result in a large kaboom as the turbine entered overspeed and the turbine blades liberated.

As for the last electrical device operating, my money would be on a solar powered yard light. The quality of those devices is generally terrible but the law of averages suggests some of them have to be on the long tail of a MTBF curve.

Re:satellites

[TheCastro1689]

In a recent most life dies scenario it was pointed out by experts that the east coast would become a nuclear wasteland within a few months. Basically nuclear reactors would explode rather quickly. Sad news in a zombie apocolypse.

Re:satellites

[magarity]

Hint: the moon is beyond geosync distance but somehow it manages to stay in orbit

In the long term this is incorrect. The moon's orbit is not permanent. Since the original question is what devices will keep working without humans present then the correct answer is likely satellites, but those are not forever.

Re:Satellites

[ComputerInsultant]

Actually there are small bits of dust clustered around L2 (Gegenschein). https://en.wikipedia.org/wiki/... L2 is not stable, but it takes a long time to clear out the things that are there.

Re:satellites

[fahrbot-bot]

The currently available Kyocera Kona (basic flip-phone) reportedly has 6 hours of talk and 29 days of standby.

Re:satellites

[fahrbot-bot]

... as the turbine entered overspeed and the turbine blades liberated.

See. It's not just software. Turbine blades want to be free too.

Re:satellites

[Z00L00K]

Probably because in space there's no corrosion. Anywhere on earth you have chemical aging of material and corrosion to consider. You have some of it in space due to hard UV, but it takes longer.

Re:Satellites

[MachineShedFred]

The issue then would be degradation of the systems on the satellite from micrometeor impacts, radiation, heat, etc.

They wouldn't last forever - eventually the solar arrays would stop supplying the minimum voltage needed and it would turn into another orbiting chunk of metal. Or the electronics would burn out from too much radiation damage - shielding doesn't completely eliminate the problem forever, it just delays it really well.

Re:Satellites

[MachineShedFred]

Lots of these use rechargeable NiCad or Li batteries, which would fail well before the PV cells or the LED.

Re:Satellites

[smaddox]

Pretty much all space solar cells are based on GaAs due to their higher efficiency. The rate of dopant diffusion at operating temperatures should be negligible. Most likely the electrical contacts would fail well before the cell itself; however, radiation damage might win out, and degrade the efficiency to effectively zero. This NASA report shows that efficiency approaches zero after exposure to 1 MeV electrons at a dose of ~1e17 cm^-2. I don't know what the exposure rate would be at L1/L2, but if you can find that, you should be able to estimate the lifetime due to radiation damage.

Re:satellites

[avgjoe62]

No, I think it will be a smart house in Allendale, California.

See the proof here.

Re:satellites

[ikedasquid]

Most of the US Navy's nuclear ships are setup to be refueled at least once in the expected lifetime...but that's not the main thing here. The steam plants on these ships leak (clean, not radioactive) water/steam when they operate. So much so that they have built-in make-up systems to replace the lost water. Eventually the tanks that store the make-up water will run dry, and new tanks must be placed in service. Doing this is a manual operation - sailors turning valves.

I don't know how often this must be accomplished, but there's no way it's more than a week before it must be done. If it is not done, I'm pretty confident you have less than a day before some alarm/safety system detects low water levels and stops feeding the boilers. After that happens it won't be long (minutes?) before a different safety system shuts down the reactor.

Re:satellites

[Aighearach]

There is either corrosion or an analogue, depending how you define it.

There isn't Earth weather, but there is space weather. Much of the electronics on a space craft are susceptible to damage from that weather. Just as the stuff in space doesn't have much Earth weather to worry about, the stuff on Earth doesn't have the space weather to worry about. I'm not sure why an IC on Earth would be expected to have a shorter lifespan than the ones in space. The ones in space have to be specially constructed just to keep them from failing rapidly. And there is still a high failure rate. Satellites don't last long and have to be replaced frequently.

Re:satellites

[hackertourist]

Most of the US Navy's nuclear ships are setup to be refueled at least once in the expected lifetime

Yes, that's why I specifically referred to CVN 78 which no longer has that requirement. The latest nuclear submarines have also been designed to do away with the midlife refueling, since that's a horrendously expensive 2-year-long drydock job.

Re:satellites

[Aighearach]

A good argument has been made that the metal shells of satellites will orbit for a long time, but no reasonable case has been made that their very sensitive, short-life electronics will survive longer than things on Earth that are isolated and have decent power generation, like remote solar powered scientific sensors.

Re:satellites

[Aighearach]

I'm using a small dumb-phone for emergencies and a wifi tablet for modern apps. My dumb phone is just a $20 Samsung, and it holds a charge for over a week.

The thing is, if you wait that long every time, the battery has memory effects. You still need to charge every 2 or 3 days.

Those smart phone batteries may last a few days, but if you're not charging every day the battery won't last. That's just the deal with these newer battery technologies.

If you want one with long life and little battery memory, I recommend a Motorola from the early 90s. Under 10 lb, and comes with a car adapter!

Re:Satellites

[camperdave]

I think he meant the Earth-Sun L1 and L2 points.

Re:Satellites

[Lumpy]

http://en.wikipedia.org/wiki/G...

In general we think of a generation being about 25 years - from the birth of a parent to the birth of a child. We also generally accept that the length of a generation in earlier periods of history was closer to 20 years when humans mated younger and life expectancies were shorter.

So we have them that have outlasted a human generation and are still in operation to this day.

Re:satellites

[ArsenneLupin]

satellites would fail quite quickly as their reaction mass ran out or their reaction wheels failed...

Actually there's no reason to believe that their reaction mass would run out more quickly than with humans still present. The contrary would be true: no ground station to send station-keeping commands, and so no consumption of reaction mass.

But they might end up bumping into each other, or spiraling up or down out of their orbits eventually without any commands to keep them in place.

Re:satellites

[ArsenneLupin]

And while reaction wheels do go bad pretty quickly (a decade or so)

With a "decade or so" the satellites would already be a handsome winner. Little other technology will survive for even a month without being powered or recharged.

Re:satellites

[ArsenneLupin]

Not in Geostationary orbit. They will be there "forever", they may still work for a long time, but eventually debris impact and radiation will kill them.

Nope. Even geostationary satellites do drift, and need "station keeping manoeuvers" to keep them in place. Especially the colocated ones (more than one satellite in one orbital position).

And the commands to perform those manoeuvers are sent manually, from ground.

Re:satellites

[ArsenneLupin]

Since the original question is what devices will keep working without humans present then the correct answer is likely satellites, but those are not forever.

... and even if a satellite stayed in place forever, they wouldn't stay operational that long. In less than twenty years, all its transponders would go silent, and it would just be an inert mass, just like my roomba without electricity and a full dust pan.

Re:Satellites

[SuperKendall]

Out of four garden lamps I put in about five years ago, only one functions and that very poorly.

The LED lights themselves may still work, but the solar collectors on top get clouded or dirty (or both) and stop providing power.

Re:satellites

[Bangback]

So as a certified Naval Nuclear engineer (though a decade stale...) These nuclear plants are designed for significant manning for routine operations. One of the key systems controls distribution of electrical power and continuous control of voltage through the various electrical "grids" in the ship. This system is highly manual because its the type of thing that a trained person does really well and there are all kinds of niche conditions that rely on a lot of judgment. As the voltage is manually tweaked multiple times/hour, it is pretty certain that the voltage would go out of spec in a matter of days or weeks, eventually triggering cascading failure as one redundant system after another falls offline due to electrical transients. Highly unlikely it would make it a year. With one really trained person you could probably make it 5-10 years until equipment failure due to a lack of maintenance (either skills or parts) caused shutdown.

Re:satellites

[ikedasquid]

I'm pretty sure 78 will still go through RCOH...but again, that doesn't matter. It's make-up water and manual alignment of the tanks that is the limiting factor.

Re:satellites

[ChrisMaple]

Most solar powered yard lights rely on NiCad cells, which are seldom good for more than 2000 cycles or 10 years if not cycled. Most of them use CdS photoresistive cells to turn on and off the light, and CdS cells are some of the worst devices ever fabricated. Solar powered yard lights aren't sealed, so moisture and temperature variation cause corrosion. They don't stand a chance.

Re:satellites

[Z00L00K]

They will still be in orbit for a very long time, just not the intended orbit.

And if they aren't manually shut down they can be active even when they have run out of fuel used to keep the position.

Re:Satellites

[dcw3]

Remember how much energy was needed to get it up there from low earth orbit. The satellite has to give up all of that...

Wasn't that already given up to some sort of reverse thrusters placing it into geosync?

Re:Satellites

[dcw3]

which is why trojans collect there.

Wondered where all the used ones ended up.

Re:satellites

[Cramer]

That would be true of satellites as well... if they don't recharge everyday (solar), they'll be dead in short order.

Our long range space probes (voyager, etc.) have already won this contest. They've been running for decades off their RTG, well out of range to be solar powered.

Re:satellites

[Mr. Slippery]

My small pocket sized wifi cellular router only costs $45 for 3 GB no matter how long it takes to use that 3 GB (no time span requirement, just however long it takes to use the traffic amount)...

Who??? Give! You can't tell us about that and not tell us where to sign up.

Re:satellites

[toddestan]

It's an interesting idea. I agree those things are about the cheapest crap you can buy, but given millions of things out there, is it possible that one might beat the odds and make it a few decades?

Instead of those self-contained units, what about an off-grid system someone set up with a regular solar panel, inverter, battery bank, and some low-powered LED lights that automatically come on at dusk? The batteries would eventually fail, but assuming they were sized to run the house, they could probably power some yard lights even in a very degraded state for a very long time.

Re:satellites

[toddestan]

My guess is most any nuclear reactor would make it might be a few weeks before something triggered it to automatically shut down. If it's connected to the grid probably only a few hours max. Nuclear power plants will automatically shut down when they lose grid power, and I don't see the power grid staying up very long on its own. Non-grid connected reactors might last longer, but nuclear reactors are pretty high-maintenance devices and when some valve gets stuck, or some sensor goes out of spec, or coolant levels get low and no one is there to address it the safety systems will shut the reactor down.

On the other hand, it's not unreasonable for a satellite to last 20-30 years with essentially zero maintenance.

Re:satellites

[Neil Boekend]

Memory effect is exactly what happens if you recharge those old NiMH batteries before they were empty.
Modern phones have Li chemistry batteries. They behave quite different: discharge them till empty and they die. Your usual behavior should be to charge them before they are half empty if you want them to last. Emptying them to 5% (when the phone tells you the battery is empty and will shut down) can be done once in a while, sure, but for battery life it should not be the default way of working.
Treat a NiMH the way you should treat a Li-ion and it dies soon. Treat a Li-ion the way you should treat a NiMH and it dies soon.

Re:satellites

[Aighearach]

You regurgitated some stuff, but didn't fill in the details I alluded to.

The older ones that are as I described are not NiMH, they were replaced by NiMH. Feel free to regurgitate another batch if you want to play again.

(NiMH just suck as to this discussion, they didn't need to be mentioned)

Re:satellites

[Neil Boekend]

If you don't mean memory effect then don't use that term.

Re:satellites

[Aighearach]

You whooshed pretty hard on that one. Try again sober.

I reaffirm what I said. You missed the point. And no, I didn't use a term wrong.

The original Planet of the Apes

[Anonymous Coward]

lightbulbs in the abandoned subway station...

XKCD answered this

[slart42]

A very similar question was in an XKCD "What If?", but only in the printed book version (which has a bunch of extra chapters compared to the blog): "What would be the last artificial light source to glow when all humans were gone".

IIRC, the conclusion was that it would be status LEDs on space probes or radiation glow from buried nuclear waste.

Re:XKCD answered this

[Anonymous Coward]

Gnu Hurd build server??

Re:XKCD answered this

[CheeseyDJ]

IIRC, the conclusion was that it would be status LEDs on space probes or radiation glow from buried nuclear waste.

Why would space probes have status LEDs? Think about it.

Re:XKCD answered this

[someone1234]

Unit testing before launch?

A nuclear power plant (and its control room)?

[antiperimetaparalogo]

A nuclear power plant (and its control room)?

Re:A nuclear power plant (and its control room)?

[antiperimetaparalogo]

When i wrote "nuclear power plant (and its control room)" i had in mind a German design that i think operates in South Africa - i don't remember the details but i think it covers everything you wrote.

Re:A nuclear power plant (and its control room)?

[antiperimetaparalogo]

Assuming you mean proper operation and not melt-down many nuclear power plants have too many dependencies- they require cooling and maintenance to keep working.

When i wrote "nuclear power plant (and its control room)" i had in mind a German design that i think operates in South Africa - i don't remember the details, it was experimental few years ago, it had some kind of cover for its fuel that protected it from melt-downs.

In contrast these can run for many decades:https://en.wikipedia.org/wiki/...

Even longer if you design them accordingly.

That is great... but for a sci-fiction stupid question my answer is better, since it has a control room - imagine that you are the only surviror, go inside it, and in some sign you read: "NO SMOKING".

Re:A nuclear power plant (and its control room)?

[Spazmania]

Nope. It would scram when the rest of the electric grid collapsed a few days in. The plant has to constantly output power. When the grid fails, the plant automatically goes into safe mode to avoid tearing apart the turbines. Diesel generators then start up to run the plant until grid power returns but they'd only last as long as the fuel.

Nothing associated with the public electric grid would last long without humans present.

Re:A nuclear power plant (and its control room)?

[antiperimetaparalogo]

Hmmm... i wrote about a nuclear power plant because i had in mind an experimental German design -that i think it operates now as production ready in S.Africa- that has some "pasive" safety features as a pebble-bed reactor - i don't remember details, but does a pebble-bed (if theoreticaly is designed for a scenario like the story suggests) reactor needs such measures?

Re:A nuclear power plant (and its control room)?

[chr1sb]

I suspect that you're thinking of a pebble-bed reactor.

Re:A nuclear power plant (and its control room)?

[antiperimetaparalogo]

I suspect that you're thinking of a pebble-bed reactor.

Yes, you suspected right Sir, i already mentioned it as pebble-bed (but i was not sure!) in some other comment earlier, but you provided me with the answer about this German design that was used in a demo plant - thank you!

Re:A nuclear power plant (and its control room)?

[Lumpy]

The nuclear pile still puts out a decent amount of heat after scram, and there are thermocouples down there for emergency power.

Re:A nuclear power plant (and its control room)?

[necro81]

A nuclear power plant (and its control room)?

I would recommend reading The World Without Us, which examines what would happen to the relics of our civilization if we, humans, suddenly disappeared (i.e., not extinct via war or disease, but just hypothetically got raptured away). Nuclear power plants don't fare so well. In fact, without human attendants to control them, cleanly power them down, and then decommission the plant several decades later, there is every likelihood that some nuclear plants would experience catastrophic accidents within weeks to years that would spew radionuclides all over the countryside.

Re:A nuclear power plant (and its control room)?

[dbIII]

Neutrons blast the absolute shit out of anything near an intense source of them so nuclear power plants are designed to have some reactor parts replaced after a number of decades. Somebody is bound to take that as a criticism but it's just a thing that has to be done to keep things running like replacing a car battery every few years.
Yes there are materials resistant to the damage and that's what's used so replacement can be put off for longer.

Re:A nuclear power plant (and its control room)?

[camperdave]

I would recommend reading The World Without Us, which examines what would happen to the relics of our civilization if we, humans, suddenly disappeared...

Or watching the series Life After People which portrays the same thing.

Re:A nuclear power plant (and its control room)?

[Cramer]

The current 40 year old fleet of US reactors? Those wouldn't last a week before some alarm trips and it automatically shuts down because the operators aren't present to deal with it. (relatively minor events happen all the time, and while they are highly automated (honestly, there are many places humans don't want to go), they aren't 100% autonomous.)

A "modern" helium pebble-bed reactor, would fair better, but something would eventually trip it, too, without any operators present.

In theory, a reactor has the fuel to run for years. But as many such sci-fi shows point out, without load from the grid, they'll automatically shutdown.

Satellites

[Gre7g]

Solar-powered, geosynched satellites will keep going for a while.

Solar powered wrist watch...

[advocate_one]

n/t

Re: Solar powered wrist watch...

[pecosdave]

My Casio is up to year 8 on the original battery. If my corpse fell where there was plenty of sunshine but not so much dirt as to wash up and bury my wrist I could see it going quite some time more. It's good for 200 meters diving, it should handle exposure for a good long time.

Re:Solar powered wrist watch...

[skids]

Well, we know it's not going to be a Pink Floyd "Pulse" CD at this point.

I was going to say one of those solar lawn lamps, as at least one of them would have gotten lucky and gotten an exceptionally long lasting battery and solar cell.

But this trumps it. Or a solar calculator, though whether that is "working" when nobody is pressing buttons is up for debate -- and if we grant that, any stupid lithium cell powered musical greeting card can still be consdered to be "working"

Re:Solar powered wrist watch...

[mspohr]

I got a lithium cell powered greeting card for my birthday a few years ago. My granddaughter loves it (silly song and dancing hamster). She plays it several times a week (yes, I'm getting tired of the song). However, still going strong! I believe it will live forever.

WALL-E

[ei4anb]

you should watch WALL-E next, while pondering that question

Solar powered parking meters

[captainpanic]

1. Solar powered parking meters, obviously. Humans may be all gone, but you still gotta pay for your spot downtown.
But seriously though, these are designed to be robust, and to keep working even if the solar panel gets dirty. I don't see any reason why it would fail at any time.

2. The other one I can think of are (again, solar powered) satellites in higher orbits. But I am not sure how much damage the solar radiation does to those on the long run.

3. Wouldn't it be sad if the last electric device to work is one of those crappy solar powered moving plants (made of plastic)?
One of these: https://www.youtube.com/watch?...

Re:Solar powered parking meters

[ErroneousBee]

http://en.wikipedia.org/wiki/R... has no moving parts, no coin slot to clog up. One somewhere where plants won't swamp it, but no too exposed that the weather gets it.

Re:Solar powered parking meters

[93 Escort Wagon]

I would have said solar powered LED highway road crew signs.

Preferably one where the message has been altered to read "ZOMBIE APOCALYPSE AHEAD".

"There will come soft rains"

[Cutterman]

Ray Bradbury asked the same question in 1950.

http://en.wikipedia.org/wiki/T...

The Cutter

Re:"There will come soft rains"

[Kiyyik]

Yes! I was hoping somebody would mention this :) I read it when I was a kid in school, and it chilled me to the bone. I wasn't really into sci-fi at the time, but Bradbury always knocked me out with his stories. He really got into the human side of things in a way a lot of the "space opera" types never did. This particular story is very haunting, and I think comes very close to how it will be--heck, come back and read this thing now, and see how plausible most of it is as our homes get smarter and smarter.

Re:"There will come soft rains"

[rwise2112]

There was also a documentary that I quite liked called Life After People which discusses this in part along with other things.

Re:"There will come soft rains"

[cetan]

I too read this when I was young, as a part of a science fiction anthology book we had in school. It is the one story from that time that has always stuck with me. The over-shadowing sense of futility and loss in the story really triggered something in my brain.

Re:"There will come soft rains"

[Hattmannen]

That was the story my mind leapt to as well.
It was initially written as a short story and later incorporated in "The Martian Cronicles". There was also a great dramatisation of it on Dimension X, which can be found on The Internet Archive.
It is hard to believe it was written in 1950. Sure our vision of what the tech looks like might have changed a bit, but not the essence of the story. It is a truly chilling vision and it looks increasingly plausible with every year.

There will come soft rains and the smell of the ground,
And swallows circling with their shimmering sound;

And frogs in the pools, singing at night,
And wild plum trees in tremulous white,

Robins will wear their feathery fire,
Whistling their whims on a low fence-wire;

And not one will know of the war, not one
Will care at last when it is done.

Not one would mind, neither bird nor tree,
If mankind perished utterly;

And Spring herself, when she woke at dawn,
Would scarcely know that we were gone.

--Sara Teasdale

Re:"There will come soft rains"

[Mike Buddha]

https://youtu.be/LzhlU8rXgHc

Voyager 1 and 2

[scsirob]

These puppies are way out there, running on neclear power. No-one to bug them, nothing to break them.

Re:Voyager 1 and 2

[geekmux]

These puppies are way out there, running on neclear power. No-one to bug them, nothing to break them.

Nothing?

Because there are not millions of objects hurtling through our universe at any given moment, as we sit here and theorize what large object might have wiped out all life on this planet before?

The universe is nothing but one big pinball machine. Luck runs out eventually.

Re:Voyager 1 and 2

[Coisiche]

NASA gives the thermoelectric generators on the Voyagers 10 more years".

Re:Voyager 1 and 2

[aliquis]

"The universe is nothing but one big pinball machine. Luck runs out eventually." geekmux

Should be added to the quote DB ;D

Re:Voyager 1 and 2

[Anonymous Coward]

Do you realize how fucking HUGE and EMPTY space is? On average, it's ONE ATOM PER CUBIC METER.

ONE *ATOM*.

Re:Voyager 1 and 2

[Lumpy]

Xleggorp and Zullupt have been tagging that thing every time they make the Kessel run in less than 11 parsecs.

Re:Voyager 1 and 2

[necro81]

Given that New Horizons is several decades (Pu-238 half-lives) younger, I would place my bets on it.

Re:Voyager 1 and 2

[DarthVain]

...or so we thought. The first alien selfie was a real kick in the junk.

Re:Voyager 1 and 2

[Cramer]

The odds of voyager getting hit by a space rock are about as good as you getting hit by one.

Interesting topic...

[davidmcg]

The problem is that any country or area relying on renewable energy, such a wind turbine, hydro-electric etc will still continue to run, but the power stations will be unmanned and as a result, the system will shut down despite the original source of that energy still running. Only devices I can imagine continuing to run would be things like solar powered devices, like watches, calculators etc. I don't think anything complex would be able to continue without human intervention.

Re:Interesting topic...

[bughunter]

I still use a "solar" powered calculator, although most of the time the light comes from an electric lamp.

Yes, I have several smartphone calculator apps. But only physical keys are suitable for fast, repeated, accurate calculations.

High powered electric heaters

[tempmpi]

They are just a piece of wire, often embedded in some kind of ceramic. Without power and stored at a place well protected from the enviroment it would likely last for 100,000 years or more.

Re:High powered electric heaters

[necro81]

They are just a piece of wire, often embedded in some kind of ceramic. Without power and stored at a place well protected from the enviroment it would likely last for 100,000 years or more.

I think that the expected lifetime of Ohm's law is roughly the age of the universe.

Re:High powered electric heaters

[GrantRobertson]

Doesn't meet the OP's requirements. It has to be functional and still functioning even with grid power off.

Re:Easy

[Kokuyo]

As opposed to a non-nuclear nuke?

Easy

[mwvdlee]

The last electrical device after humanity ends will be the deathray which the aliens used to blast humanity out of existance.

Re:Easy

[lisaparratt]

As opposed to a Weapon of Vaginal Destruction.

RTG's

[Anonymous Coward]

I know there are some remote Russian lighthouses powered by RTG's that could conceivably work for a few decades unattended, I'd imagine there are similar things along the lines of radio repeaters etc that could be similarly powered and keep going for a century or so.

Probably the Oxford Electric Bell

[Chrisq]

The Oxford Electric Bell has been running since 1840 and will probably carry on for a long time yet

Re:Probably the Oxford Electric Bell

[del_diablo]

Isn't that just powered by the magnet field?

Re:Probably the Oxford Electric Bell

[Bing Tsher E]

Only in the same sense that a Tesla automobile is powered by the magnet field.

Both have chemical batteries driving an electromagnet.

Re:Probably the Oxford Electric Bell

[camperdave]

Nope. Static electricity. One bell is positively charged, and the other negatively. The clapper is attracted to one of the bells, gets charged and is thus repelled. It is attracted to the other bell, where the charge is dissipated and the clapper picks up the opposite charge. The cycle then repeats.

Re:Probably the Oxford Electric Bell

[fuzzywig]

No, it uses a very early form of battery. However, no one is quite sure exactly how it was made. Each ring of the bell only takes a tiny amount of power, so it should be going for a good long while yet.

a data collection device in antarctica

[lkcl]

andrew trigdell told me an amazing story back in 1999 about how he helped install Linux 0.99 on a solar-powered data collection computer in antarctica. Linux 0.99 was known to be highly stable, hence why it was chosen. it has a 56k modem which is enough to get the data back, and to check (very slowly) that it's still operational. so i think anything that's designed for long-term with those kinds of harsh remote and inaccessible conditions in mind, powered off of sustainable independent power, would be a good candidate for a device that would still be functioning even decades later.

Re:a data collection device in antarctica

[necro81]

so i think anything that's designed for long-term with those kinds of harsh remote and inaccessible conditions in mind, powered off of sustainable independent power, would be a good candidate for a device that would still be functioning even decades later.

It's design is probably robust for decades, but anything out on the ice that sits still for more than a few years is destined to get buried in snow, solar panels and all.

Re:a data collection device in antarctica

[dbIII]

There's some spots that don't get covered with snow. That's where world wind speed records are set.

If humans still exist... A lot of stuff.

[jellomizer]

Unless.... You expect a targeted strike where the only survivors are liberal art majors. Spin some magnets in some copper wire you get electricity. You can spin it with a water wheel, wind turbine, combustion engine, or steam power.

You reverse the process and you can make a motor.
Place some resistance you get heat and light. ...
This dystopia future just isn't practical unless there is some lead time where science and engineering has been some how removed from our cultures.

Re:If humans still exist... A lot of stuff.

[Lumpy]

Or worse, Graphic arts majors.

Re:If humans still exist... A lot of stuff.

[Barny]

You might want to re-read it. "After we are gone" in the sense of "no more humans."

Re:If humans still exist... A lot of stuff.

[Bing Tsher E]

Even worse: web developers.

"Now, here is where we just load up the Javascript we've written to get that thing to start a fire...."

Re:If humans still exist... A lot of stuff.

[Bing Tsher E]

You're attributing too much humanity to the fratboys and sorority girls in college.

Oblig xkcd?

[Trogre]

Sounds like a question for Randall's "What If" series.

Fascinating question though. I should imagine one of the voyager probes perhaps, or a solar powered beacon on a hill somewhere. A lot of stuff on Earth is very dependent on regular maintenance so I wouldn't imagine much stuff still working fifty years from the hypothetical extinction event.

Re:Oblig xkcd?

[Golden_Rider]

Sounds like a question for Randall's "What If" series.

It actually has been done, although as far as I can see only in the "what if" book. It has a chapter "the last human light" which pretty much is this question, just limited to "lights" instead of "device".

Something burried inside mountain

[gnujoshua]

Whatever it is, I think it will be inside a mountain. Maybe there should be an digital computer version of the 10,000 year clock (mechanical not electrical) or something like the Svalbard Global Seed Vault.

We already know

[Coisiche]

It's actually on an adjacent planet rather than Earth, but Opportunity seems like it will just keep going.

xkcd joke about it.

Help Stamp Out Care Bears!

[TheRealHocusLocus]

Topic: Ask Slashdot: After We're Gone, the Last Electrical Device Still Working?
Reply Subject: Who cares?
Response: I don't care. I think that it isn't a interisting problem.

Analysis: Looks like this dim bulb has gone out.
Tally: Currently at 999 Points of Light
Who: A Child Left Behind
What: Passionate declaration of indifference.
When: 42 years after men last walked on the Moon.
Where: March For Apathy 2015 [cancelled]
Why: Dissonant aggressive demotivational pathos.
What: 's the use.
How: Did we get here?
Further Reading on this Topic: Failed Slashdot submission,

Breakthrough: Manned Space Travel Achieved Using 40-Year Old Technology

TheRealHocusLocus writes

"Paul Rosenberg has uncovered some surprising new evidence that manned space travel is not only possible, it has actually been achieved using decades-old technology. Some 40 years in the making, a tale too amazing to remain untold. With a few quaint photographs he asks, could we build this? The answer is no. Or is it? It is uplifting to read that "Productive humans have been delegated to mute observance as their hard-earned surplus is syphoned off to capital cities, where it is sanctimoniously poured down a sewer of cultured dependencies and endless wars..." for it must take something really compelling to prevent us from reaching the stars, and he has nailed it. This essay makes the case that the headliner of 2052 may well be: Breakthrough: Manned Space Travel Achieved Using 80-Year Old Technology. I can hardly wait! Down with robots."

Re:Voyager

[Coisiche]

The thermoelectric generators on both Voyagers have a finite lifetime. I don't think it's very much longer until they are both inert lumps of metal. 10 years according to NASA.

My Pink Floyd Pulse Album

[Anonymous Coward]

The little LED on my pink Floyd pulse album will still be flashing long after we are all gone.

If the power comes back on

[houghi]

Everybody is missing the "if the power comes back on". That means we are talking about things that are electronic and can be stored for a LONG time. What you need to consider is more rust and material decay.

Metal? That will rust, so after a while you will not be able to just plug it in an electric source. Vacuum, like in a lamp, will most likely not exist after a while, also due to rust.

You PC? Let it stand and rust and oxisation will kill it when not used. Anything with batteries? They are dead even sooner.

So either a lap, an LED or a simple coil for heating. Thick enough and it would need a lot of power even now to do the heating.

My guess is that there would be things that when plugged in after 2 ceturies will still work. After that? With some cleaning.

Re:If the power comes back on

[Lumpy]

Rust only happens in the presence of oxygen. and I have seen sealed systems that are flooded with noble gasses that will absolutely stay in like new condition for 10,000 years unless the polymer seal fails massively. oxygen intrusion would be so slow that they would probably still be OK 5000 years after a seal failed.

Re:If the power comes back on

[Barny]

No, they are ignoring it because that is what the question specifically tells you NOT to consider.

Here, broken apart by more than just a semi-colon.

"I don't mean sitting there with no power but would work if the power came back on[.]"

"rather, something continuously powered, doing the task it was designed for."

Factual

[Anonymous Coward]

The problem is that not many devices have been arround long enough. The oldest lightbulb:

http://en.wikipedia.org/wiki/L...

I would expect that if you would keep a simple device, like a lightbulb, in a vacuum and store it in bubblewrap, that it would be able to provide light for millennia, as long as it is not used in between.

Re:Factual

[Neil Boekend]

You didn't read the full question, did you?

Ironically?

[tlambert]

Ironically?

Given the way things are going, probably the Opportunity Mars Rover...

There will come soft rains

[renergy]

Guess Mr. Bradbury has already described the situation rather well in Martian chronicles. [Offtopic]the short story was actually compulsory reading in communist Czechoslovakia, i have read it over so many many times while waiting for a train[/offtopic]

Probably small static battery powered devices

[Mostly a lurker]

I doubt the ones that immediately occur to me will be the winners. However, there are smoke detectors guaranteed to work for over a decade, and there is a pacemaker with a minimum lifespan of 14 years. Some digital watches are also 10+ years. Tadiran batteries are supposedly good for 40 years in some applications (remote monitoring devices?)

The lights...

[Anonymous Coward]

at Hoover Dam.

Re:The lights...

[pr0t0]

I wish I had mod points. IRC, Hoover Dam would continue to make power for 50 years after we're gone. The only reason it will stop, is the water intake valves will get too clogged with marine detritus without divers to periodically clean them.

Re:The lights...

[Phronesis]

There's a lot of wear on the turbine systems. They require weekly to monthly checks of bearings and regular lubrication and maintenance. I doubt that Hoover Dam would operate without maintenance for 50 years just as you could not run a car for a decade without ever changing the oil.

Re:The lights...

[FranTaylor]

the water intake valves will get too clogged with marine detritus without divers to periodically clean them.

marine detritus? since when is there seawater behind hoover dam?

Re:The lights...

[pr0t0]

Not seawater, but Lake Meade isn't exactly a swimming pool either. It's a vast biological marine ecosystem; full of life. The life in question, is Quagga Mussels. They're an invasive mussel species that, left unattended, would clog the cooling pipes.

The Clarendon Electric Bell.

[Noryungi]

It has been ringing continuously since 1840, and will probably continue for a long time.

Read all about it here: http://www.atlasobscura.com/pl...

Re:One thing working, I know!

[TheRealHocusLocus]

The atomic clock might be the only thing working after we are gone.

Isotopes will be decaying, but there will be no one and nothing to count the decays. But what ever is the last thing actually running, you can be sure it will be running Windows NT. For more on this and a brief history of the circumstances which brought our modern world to its knees, see my own little parable about technological angst and global catastrophe, The Time Rift of 2100: How We lost the Future

Postscript to the story: Internet of Things security concerns! Must. Implement. Granular. Crypto.. Yep, we're on track.

BS question

[gweihir]

The power goes out permanently about 15 minutes after human input stops. That is the planning interval in the large electrical grids. No humans results in emergency shutdown of all power-plants a few minutes later and that is it.

Re:BS question

[camperdave]

And how, exactly, does that affect the many solar powered devices around the planet? Or the microhydro sites and small scale wind turbines? Or the RGT powered lighthouses and research equipment?

There's lots of stuff designed to be off the grid, or to work specifically when the grid fails, so the loss of the grid is definitely NOT "it".

Solar

[Metabolife]

I would assume it would be a simple device connected to a fully solar electric living space. If the solar panels and tech behind them keep functioning, the simplest of devices (maybe even a phone connected to an outlet) would have the most longevity.

The Oxford Electric Bell

[DrJimbo]

The Wikipedia explains:

The Oxford Electric Bell or Clarendon Dry Pile is an experimental electric bell that was set up in 1840 and which has run almost continuously ever since, apart from occasional short interruptions caused by high humidity.

[...] The Oxford Electric Bell does not demonstrate perpetual motion. The bell will eventually stop when the dry piles have distributed their charges equally if the clapper does not wear out first.

Nuclear power

[olterman]

Probably some secret and isolated nuclear powered monitoring system somewhere.

RTG of some kind

[swb]

The power output of RTGs declines over time, but according to Wikipedia Americium-241 has a half-life of 432 years and is an experimental replacement/alternative to Pu-238 RTGs.

Since shielding and weight requirements are a non-factor for terrestrial RTGs, it wouldn't surprise me if there was some secret bunker someplace with a huge (1-2kW) RTG in place as a kind of emergency power source capable of powering a control system or something to bring up other power systems.

Re:RTG of some kind

[Neil Boekend]

Obligatory Userfriendly

That would be useless wiring weight

[aepervius]

Before launch you have connector to which you connect a computer and you can do a self diagnose on the satellite using that connection to the on board system. There is no reason to dedicate leds and leds wiring for that especially that you will need to check for many fail conditions.

Re:That would be useless wiring weight

[necro81]

Before launch you have connector to which you connect a computer and you can do a self diagnose on the satellite using that connection to the on board system. There is no reason to dedicate leds and leds wiring for that especially that you will need to check for many fail conditions.

Having spun a number of boards in my career, I can tell you that it is trivial to add an 0402 LED indicator, just as an indication that the 3.3 V logic rail is powered. And because it was easiest (via inertia) to keep it in than to cut it out (even as a do-not-populate instruction to the board house) that little LED stayed in the design, even though in production no one would ever see it.

Given the complexity of most satellites, I would be deeply surprised if there wasn't at least one LED on one of those boards.

Re:That would be useless wiring weight

[TheCarp]

> And because it was easiest (via inertia) to keep it in than to cut it out (even as a do-not-populate
> instruction to the board house) that little LED stayed in the design, even though in production no one
> would ever see it.

And someday down the line, in a situation he never imagined, someone is going to see that LED and be glad it was there. Possibly after its been scrapped, rescued from the scrap heap, and repurposed to replace the fried control in someone's coffee maker....but appreciated none the less.

Re:That would be useless wiring weight

[AikonMGB]

We don't design LEDs into our own boards, and we explicitly remove them from COTS boards that we use. Generally speaking the diffusers on LEDs outgas, meaning a) they are depositing materials on your spacecraft surfaces (bad) and b) could result in a shorting risk (also bad). There may be space-grade LEDs that big-space (think Hubble, JWST, Voyagers, etc.) use but I would be surprised. There's simply no need.

"Is it plugged in? Is it turned on? Is it on frequency?" solves about 99% of basic device connection issues. An LED will make one very short portion of that slightly shorter, and then only when testing on the bench, since you can't see it as soon as you box it up. As soon as you can talk to a device, you are able to run a long form functional test on it, exercising every part of the design and ensuring everything is working correctly. If it passes, you're good. If it fails, you pull the unit.

For ground support equipment, yeah sure, throw an LED on every rail and switch output.

Re:That would be useless wiring weight

[Immerman]

"Hot damn, an LED! We can sell this and eat for a month! Too bad we don't have the technology to actually make these anymore...."

Re:That would be useless wiring weight

[meta-monkey]

That's kind of an interesting point. Let's say 500 years from now humans have warp drive and can zip between stars. Should the Voyagers be collected* and placed in a museum? Should it be left alone as a historical "site?" One could visit it. Get the sense of what ancient man accomplished, sending this tiny thing so far from home. But that would be lost putting it on display at the Smithsonian.

I argue a "traveling" museum should be built next to Voyager, sharing its journey. School kids could warp in on their Space Bus, view Voyager from the observation deck, yawn through holoexhibits that tell the story of ancient space exploration, and then have lunch at the Carl Sagan Cafeteria.

* I used as neutral a word as I could, "collected," rather than something that implies a relationship. "Recover" isn't right, because that implies retrieving something that rightfully belongs elsewhere and was lost, and Voyager is not lost. It's exactly where it should be doing exactly what it was designed to do. "Rescued" is right out. That's recovered + emotion. "Interrupt its mission" or "abduct" are emotionally loaded from the opposite perspective.

Re:That would be useless wiring weight

[adolf]

Two reasons why LEDs aren't on satellites:

1. Weight. (A few LEDs might add up to maybe even a whole gram, all said: And you can use that gram somewhere else.)
2. Power. (Which is not unlimited. And using more power requires larger solar panels, and more storage for dark hours.)

Given the complexity of most satellites, I would be deeply surprised if there weren't a few zero-cost test points...maybe with a provision for a POGO test rig to test things with if the thing is ever seen intact by humans again (which it won't be).

Re:That would be useless wiring weight

[TheCarp]

I was more thinking of other boards than the ones on space equipment, but even then, for each specific board out there, you know there was at least one or more prototypes and possibly a few rejects that got tossed in a scrap heap.

But aside from emotionally loaded, all of them imply something that doesn't make sense... the idea that Voyager has some intention or plan of its own. It is equipment we put out there. Aside from the emptional, there is little difference between collecting it and reeling in a tape measure, with the exception that you will actually reuse a tape measure and the cost of reeling it in is far less than the cost of not doing so in most situations.

Re:That would be useless wiring weight

[wooferhound]

Voyager is carrying a Message From Earth to any possible civilization that may find it . . .

Re:That would be useless wiring weight

[meta-monkey]

Yeah, I'm just saying it would have historical significance centuries from now. It'd be a lot more than a mere tape measure.

Re:That would be useless wiring weight

[Kiaser Zohsay]

V*GER? I hardly know her!

Uncle D

[Mr D from 63]

Walt Disney's head freezer is designed to keep running indefinitely.

Re:Uncle D

[Bing Tsher E]

It's fueled by lawyers and the hope of small children. It'll never burn out.

Western Electric Type 500 phone

[AntronArgaiv]

Those things are pretty much indestructable.

Flashlights

[MagickalMyst]

Not sure if this counts, but electro-mechanical devices such as a shake flashlight would probably still work.

Re:Flashlights

[toddestan]

The problem would be that with no humans around, there would be nobody to shake the flashlight to make it work.

My money is still on either solar power satellites or some automated, off-grid solar powered device. Either one would likely function at some level for a few decades left on its own.

Solar powered emergency phones?

[putaro]

Kind of mundane, but they're built to get installed in the middle of nowhere and keep working.

Re:Easy

[Dunbal]

Surely you mean a snuke?

Franklin Ace 1200

[DarkKaplah]

Built like a tank and designed to last. It would boot up with a copy of "The Oregon Trail" left in the drive.

Solar calculators

[moeinvt]

With moderate use, those things last forever.

Road Signs

[egyas]

Probably one of those "Use Caution" signs on the roadside. Powered by its own solar panel/battery, in a warm weather state.

Re:Road Signs

[McGruber]

I think it would be one of those flashing warning lights they mount atop signs. Most of them are solar powered, their flashing lights are banks of LEDs, and there a ton of these signs, spread all over the country, so the odds are in their favor!

The Other end

[JimSadler]

A wind turbine or water turbine could continue for decades. Since part of the power generated is fed back to coils within the turbine they both use and create electricity. I do wonder how much of a windmill farm can survive one good tornado.

Re:The Other end

[Phronesis]

With no one around to lubricate the bearings and gear trains, I'm guessing that turbines of any sort would freeze up in a few years.

Solar powered electric fence charger

[kamakazi]

A solar powered electric fence charger is designed for neglect. The fence itself will be useless, weeds will ground it fairly quickly, and anybody who maintains them knows a fence won't last a year unmaintained, but the solar powered charger will keep ticking as long as the battery lasts, and will probably keep trying even after the battery fails. The cheap little solar powered yard lights also should keep working for quite a while, at least the ones that aren't DOA when they are purchased.

But all devices that rely on a battery will be outlasted by devices using RTGs for power, or direct solar devices that don't use a battery, like those car ventilation fans you put in your car window.

The type of devices built with RTGs (Satellites and Mars rovers) are the absolute highest quality components assembled and tested with the best quality control, while the solar powered car ventilation fan is built by an 11 year old Chinese kid working an 18 hour shift, so I am betting on the satellites.

Ocean or Ice Bouy

[philip.mather5551]

Something like... http://en.wikipedia.org/wiki/N...

Re:Ocean or Ice Bouy

[Trax3001BBS]

Something like...

http://en.wikipedia.org/wiki/N...

My first thought was a ball showing a river elevation but it's not electrically controlled.

So hit on any electrical devices powered by waves http://en.wikipedia.org/wiki/W..., or Wind http://en.wikipedia.org/wiki/W...

But would have to be post-apocalyptic, as after the oil runs out it's all going to be different. As off the wall as any electricity produced going to national security, Your T.V. ain't.

fish, plankton, seagreens ...

[TheGratefulNet]

and protein from the sea.

I am 'box' !

Not anything from Apple

[QuietLagoon]

The last device still working definitely won't be anything made by Apple. The battery life in Apple devices is abysmal.

Life After People

[braksator]

There is a documentary named "Life After People" which has some ideas on how long it would take things to fall apart. They make special mention of Hoover Dam working for a long time, until the pipes get clogged up with clams or something. We need to make robots that clean out those pipes.

My Daughters Furby

[paradox11]

I swear I could take the battieries out of that thing, and it will just keep creeping me out

I am thinking satellite

[EmperorOfCanada]

I am thinking various space probes floating about. Some of those things will keep going for centuries. There is also some solar powered crap on the moon. That could keep going for stupendous amounts of time. Maybe millennia depending on the degradation of the solar panels.

But maybe there is some geothermal system that will just keep going for eons as the power generation aspect of a thermocouple should last indefinitely and is buried so it won't be subject to any weathering.

Atomic clocks don't rely on nuclear decay.....

[Ellis D. Tripp]

Nuclear decay being a chaotic process and all.

So-called "atomic clocks" utilize the RF absorption of various isotopes, typically Cesium or Rubidium. Heated to a vapor in a sealed chamber, the vapor is excited by a microwave RF source, and at a highly specific frequency, the vapor absorbs the RF energy. This phenomenon is used as part of a feedback loop to keep an electronic oscillator disciplined to whatever frequency is desired.

Atomic clocks won't work without electrical power, and would be subject to all the same physical rust and breakdown as other electronic devices over the years.

Re:Atomic clocks don't rely on nuclear decay.....

[Phronesis]

Atomic clocks won't work without electrical power, and would be subject to all the same physical rust and breakdown as other electronic devices over the years.

Indeed. I used to work at the Time and Frequency division at NIST, where they keep the master atomic clocks for the US. They had a big room full of 12-volt car batteries to provide backup power to the clocks in case of a power outage.

RTG's decay by design

[dbIII]

RTG's decay by design and current ones are not built to run for a century or so.

Re:RTG's decay by design

[dbIII]

You'd need a different design of RTG with a different radiation source material for it to work at all over a very long time. Currently they decay very quickly (to give you a lot of power for the mass) so it's going to be many orders of magnitude less output between construction and a century later.

Nuclear Power plants, solar powered things ...

[Qbertino]

Nuclear Power plants and solar powered things with no moving parts will probably be the last things still working. Perhaps a solar and condensator powered lamp in some ruggedized military component or something. Correctly built, a device like that could last thousands of years.

If we count decay of artificially saturated nuclear fuel as a man-made "machine" or "device", then we have 200 000 years of "worling devices" ahead of us.The problem here is that they "work", wether we want them to or not. Which sort of is the problem with nuclear waste.

RTGs in lighthouses

[Phronesis]

The former Soviet Union built hundreds of automated lighthouses in remote locations powered by radioisotope thermoelectric generators. Those use 90Sr, which has a half-life of 30 years so they can go for many decades. They were installed in the 1970s-90s, so most of them are around one half-life out. They could well continue operating for several decades, but some small solar-powered devices might well outlast them if they aren't damaged too badly by weather over the years.

Solar calculators

[Gnascher]

All solid state electronics ... very low power requirements.

There will be probably millions of these thing lying around for a long, long time if humanity suddenly goes poof. Many of them will be siting in protected, stable environments ... all that will have to happen is for one to get exposed to light and the little 0 will show up on the screen, ready to take input.

Electrolytic capacitors

[jzu]

Electrolytic capacitors have a poor life expectancy, typically decades. Cap jobs are common on old audio amplifiers. I heard that recent ones are made using "optimized" processes, which doesn't bode well for durability, and indeed I've had to replace a motherboard due to exploding capacitors. It gets even worse when they are not powered from time to time - it reduces drastically their lifetime.

So don't expect to get most electric gizmos in working order straight out of a time capsule. Their power supply would probably need newly made electrolytic capacitors, which could prove difficult in a post-armageddon environment.

Re:Electrolytic capacitors

[ChrisMaple]

The common, cheap, wet aluminum electrolytic capacitor has wearout mechanisms including drying out and deforming. Other types of electrolytics (tantalum, niobium, and some polymer variations) do not have a known significant wearout mechanism and should last more than 100 years if not abused.

Vibrators

[kdub007]

They are designed for extreme conditions and durability.

Stream/Wx Sensors

[clifwlkr]

These things are all over and in remote areas. They are solar powered, and designed to be left alone for long periods of time. They are built such that they can go down at night if the battery dies, and come back to life with the sun. So as long as the sun shines, these things will keep sending their data. I would bet they would just keep quietly ticking on for a very very long time.

The last thing functioning...

[jc409]

A little know and highly secret NSA monitoring system, codenamed "Losira"....

Androids...

[nowsharing]

The conscious AI robots that hunted down every last one of us.

Well

[koan]

Does a crystal radio count even though there's nothing meaningful to receive? What about a Casio watch... I just found one in my storage (over 5 years old) and its still going.

Circuit boards, maybe not

[responsibleusername]

I didn't see this mentioned by anyone else, but there is also a phenomenon called creep corrosion (and other things, the only common word is creep that I've seen) that applies to a lot of situations that PCB's are also subject to. It can cause solder joints and traces in close proximity become shorted. I think temperature has an affect, but I think it can also happen at close to room temperature in low-temperature solder situations. This would most likely spell death for the majority of our PCB based technology (it not others as well). It could cause a system running continuously to fail eventually or I think even dormant non-powered electronics to fail even if not stored improperly. There was an article here recently about a darpa program for self healing software http://www.tomsguide.com/us/da... that was interesting, but I feel like we would have to design more redundant hardware to avoid problems like this or potentially metals that aren't subject to these problems that could run long after we are gone. It wouldn't be economical though since so much hardware outlives its usefulness in our fast paced economy. Here are some pictures of creep in action: https://www.google.com/search?...

Re:Circuit boards, maybe not

[ChrisMaple]

This is a familiar problem, known under some conditions as "tin whiskers". It's a particular problem with high voltages. To help prevent occurrences, use solder mask, big spaces between conductors, and air spaces so that the creep distance is much greater than the strike distance. Sometimes provisions can be made so that there's enough power on a line to evaporate a whisker if one forms.

Re:Circuit boards, maybe not

[responsibleusername]

Thats the term I was trying to think of, so am I correct in thinking this could be a problem with a lot of circuit boards, especially those that are more cheaply manufactured? It seems possible to design hardware for very long time, but to me redundancy and intelligent software would both be needed to keep a machine/computer running for an extremely long time (assuming a stable long term powersource of course).

Solar offgrid with NiFi battery backup.

[Ungrounded Lightning]

A solar offgrid (or grid-tied with standalone capabilities) would provide power locally until too much stuff failed.

Lead acid batteries last for several years, recent lithium probably for a couple decades. Nickel-Iron batteries are more lossy, but last for centuries, if provided with water to replace evaporation, potentially decades if they have catalytic fill caps to recombine lost hydrox or, say, a reservoir-based automatic watering system. (If their chemistry has a long-term unavoidable failure mode I'm not aware of it.)

Even with the batteries dead (NiFe or otherwise) the system will have power when the sun is out until at least one panel in every series substring is too degraded, shaded, or smashed to provide adequate power.

Semiconductor controllers might go for a decade to centuries, depending mainly on whether the conductive interconnects of the semiconductors are sized to avoid electromigration at the current levels used and what they're using for large capacitors.

Wind generaors have several moving parts to screw up - how many depends on the design. For a simple homebrew one you have the main bearings, yaw bearing, and tail furling-system bearing. Any one of them failing will take it out. (Even the furling bearing: Once that screws up it doesn't furl right and tears apart in the next storm.) There's also the get-the-power-past-the-yawing mechanism (typically a long cable being twisted and manually "unwound" every few years, or a brush mechanism.) Call it a decade without maintenance at the outside.

So some of 'em may run until a nearby lightning strike fries something.

Sensor buoys?

[SuricouRaven]

Sensor buoys, perhaps? Solar power on top, overnight power via supercaps which have a lifetime far longer than any battery. They could carry on transmitting for decades, until the cells degrade too far to meet their very small power demand.

If not those, solar powered calculators. Just sitting there, doing their thing: Monitoring the 'on' button for a press that will never come.

EZ ANSWER - POINTLESS ARTICLE

[sexconker]

Anything that just runs current through a stick of metal to achieve the desired effect.
Heaters, electric stoves, electric blankets, light bulbs, etc.

Light bulbs are easily broken, but the other three can take a beating and have enough metal in them to stand the test of time. The blanket has less, but it's insulated heavily to protect it from corrosion. For all of these devices, you can bypass any temperature controls, safety mechanisms, etc. and get the device to work by simply applying your voltage directly to the heating element.

The Centennial Light

[Reeses]

Like all things, the simplest solution is the best one.

http://en.wikipedia.org/wiki/C...

Built to last. No moving parts. Encased in its own perfect environment.

This will be the first thing to power back on. It may burn out immediately after, but it'll fire right up.

hydro power plant

[jclaer]

See the 1957 book, On the Beach by Nevil Shute, nuclear fallout apocalypse.
at Amazon.
Spoiler alert:
Nuclear submarine commander looking for remaining people on the surface tracks down the last electrical signal. Turns out to be static from the neon light in a shop window.

Re:A nuclear power plant -(actually hydroelectric)

[shoor]

I saw one of those documentaries about what happens if humans disappear, and, as part of the show, they visited a hydroelectric plant to find out what would happen there. It seems that one of the problems hydro plants have is shellfish growing on the turbine shaft. As I recall, without regular cleaning and removal it would stop turning after a few months.

Everyday technology can outlive us...

[__aaclcg7560]

A British woman found dead inside her apartment two years after she died because she withdrew from society, charity paid her rent and her utilities were automatically paid. If humanity drops dead suddenly, "stay calm and carry on" will rule the day.

Re:Easy

[Aighearach]

I was assuming it was a nuclear(5) as in http://dictionary.reference.co... with "nuke" being nuclear(1).

Maybe I'm giving credit too freely, but I thought it was a mildly clever word play that was also awkward and unfunny.

IT's birdie chase

[AndyKron]

The last device working will be that goddamn ball in my cat's birdie chase toy. https://www.youtube.com/watch?...

Last device standing...

[bwcbwc]

based on current tech, will be the generators themselves. My best guess is that it will either be a solar generator (no moving parts) or else possibly one of the other renewable/low-fuel options: Wind, hydro or nuclear. None of them would last more than 20 years or so without maintenance (Fallout series not-withstanding).

Filament/glow lighting

[swschrad]

as long as the vacuum or glow-gas holds out, a function of the amount of Kovar seal around the leads and whether it has a glass cap, those puppies would come right up if power came back.

Long Now Foundation - check it out

[vinn]

If you haven't heard of the Long Now Foundation, check it out - http://longnow.org/ I think the whole paradigm is pretty cool - civilization has been in existence for 10,000 years, so let's build stuff now that will last for another 10,000 years. Surely by then any civilization will wonder what the hell we were thinking. Anyway, one of their projects is the cryptically named "Long Server". Now, assuming humans just disappeared tomorrow, it's completely possible that Hoover Dam could run for 50-100 years and thereby entire data centers could stay up and running. That's a blink of an eye in geologic terms though.

A cheap digital watch in a plastic casing.

[ledow]

A cheap digital watch in a plastic casing.

Waterproof. Shockproof. Not conductive. Can't rot. Can't degrade. Battery might die but will work fine with a new battery. Simple. No moving parts. Flat. Small. Cheap and ubiquitous. Lots of them discarded when the strap breaks or the battery dies. Likely to be left in a container of some kind and thus protected in even landfill.

10,000 year clock

[softcoder]

It's not electronic, or even electric, but if they finish it, I hope that it is still running 10,000 years from now.
http://www.10000yearclock.net/

Still in control

[Soft Filter]

Simple input devices. Switches, keyboards, mice, joysticks etc. will likely remain capable of the tasks they were designed for long after the machines we connect them to fall into disrepair. If they're in a reasonably dry location out of sunlight, made with durable materials, and there aren't any people plants or animals around to break them, they should last for centuries at least. That being said, what's a sufficient level of complexity for the sake of this thread? A switch can be as simple as two wires that complete a circuit when touched together.

My smoke detectors will still beep periodically...

[studpuppy]

But only at 2:35am. And from random spots in the house.

Re: Easy

[Demolition]

I immediately thought of solar-powered wristwatches. I'm guessing that it will be a Citizen Eco-Drive or Casio Tough Solar.

Citizen Eco-drive Wristwatch

[Thelasko]

I nominate the Citizen Eco-drive wristwatch. It will run as long as it's exposed to light for at least a few minutes every 8.7 years!

Re:A nuclear power plant -(actually hydroelectric)

[ChrisMaple]

Another problem is that any particles in the water will wear and pit the turbine blades. Eventually they'll just erode away, assuming bearings don't fail first.

Re:Speak and Spell

[Trax3001BBS]

My kids play with my original 1981 Speak and Spell ( and the Speak and Math one ) on a daily basis.

Almost bought one of those for myself, I was under the impression it would pronounce words input into it, it doesn't.

I thought this was a place for geeks...

[schamarty]

yet no one seems to have mentioned Asimov's "last question"... WTF?

http://www.multivax.com/last_q...

OT sig reply

[Kiaser Zohsay]

All my liberal friends think I'm a conservative, all my conservative friends think I'm a liberal.

Then you're doing it right.

My watch

[vikingpower]

is a Seiko with solar cells as the dial background. The manual explicitly tells the owner to expose it, once every half year. to direct sunlight for about 8 hours. Which I didn't. The watch runs fine, within very small deviations from "official" time.

If my dead body falls in a place where the watch can receive some light from the sun, it will probably run on for a century or so, before some of its components ( capacitor ? ) fails in such a way as to keep the entire watch from working.

The only annoyance: it will display an entirely wrong date, as after each non-31-day-month the date needs to be adjusted manually.

Been done

[AlanObject]

Randall Munroe worked out this question based on my question in his book. My question had to do with what the last artificial light source working would be. The answer isn't what I expected.

Re:A nuclear power plant -(actually hydroelectric)

[Cramer]

Actually, they build up on the intake and exhaust shafts. Eventually (years?) no water would make it through the plant.

Something...

[cwsumner]

Most likely, something that has not been invented yet. And that we probably would not recognise or even understand.

A good bet would be an engineered intellegent organism. Whether mechanical or organic would probably be a moot point, since advanced forms of either would be indistiguishable from the other.

Five Fives company

[safives05]

Five Fives company site is a specialized site to view all products carefully child and family and all the personal care distinctive products, considered five Fives of the oldest and most prestigious Arab companies working in the field of cosmetics company in Egypt, there are personal care products, which contain the best products used in the care body and skin, and there is information of interest to you and which contains many articles which contain many useful information for your hair and skin, and contain important tips for skin care and hair, there is also a child-care products, which contain special products for children Kalbibi Oil, cologne and lotion and all Featured Products take care of your children and sensitive Bbcherthm. Five Fives of Egypt's leading companies company which has experience in the provision of high quality for consumers, as the company has evolved to become famous inside and outside Egypt, the aim of the company is that you provide an effective, safe and high-quality products at an affordable price, also offers the company's skin care products that contain moisturizing products for the skin and creams for the skin, as the company also offers Cologne multiple types, Cologne Marie there are, Cologne Lavender, Cologne lemon and again, you can browse the Five Fives company site to learn more about premium products offered by the site, as you can communicate on the site by means of communication different social, visit the site at the following link: http://www.5-fives.com/