Slashdot Mirror
P2P Hard Disk System Warns of Tsunamis
An anonymous reader writes to mention an article on NetworkWorld about a free software application that detects Tsunamis by listening for vibrations in the hard drives of computers. The peer-to-peer network uses the technology that allows HDDs to keep read-write heads on track, and passes the information to a network for analysis. From the article: "If an earthquake that could lead to a tsunami is detected, the supernodes inform the other nodes. Computers running the client software and connected to the peer-to-peer network can then warn of such events. The software is able to provide such warnings because the seismic waves produced by earthquakes travel at about 5,000 kilometers per hour, while tsunamis move much slower at 500 to 1,000 kilometers per hour"
I like the science.
I like helping people.
I don't like Windows only java clients which include dll files (they say on the front page its java based, but then on download page you see Mac OS X and Linux version in preparation. I think the DLL is going to pretty much stop all that.
I don't like running random programs accessing none standard data from my harddrive without knowing exactly what its looking at and what happens to my drive if its not got the required capabilities.
There is no technical information about this program or what it actually examines on a harddrive, coupled with the crypto java code buried inside the package (look in bcprov-jdk14.jar\org\bouncycastle\crypto) makes me nervous.
Negating all that it looks cool and if the technical aspects are cleared up I will try it.
I wonder if this functionality could be used to give any standard laptop a similar alarm feature as on the latest macbooks?
liqbase
Whoops, misread TFA!
This should end the need for wasting countless millions on professional seismic research stations. Once again, free software triumphs.
If everybody in China jumped off of a 1 foot step ladder at the same time, would there be a sudden drop in P2P data that might indicate the event?
(of course I didn't RTFA, I'm just trying to avoid work this afternoon by wandering around making incoherent posts)
Is it just my observation, or are there way too many stupid people in the world?
I don't think all earthquakes cause tsunamis. Seems like there may be a lot more false warnings than real tsunamis.
Wouldn't be very effective in third world areas, though.
I wonder how many simultaneous reports it takes to trigger it.
Execute? [Y/N] _
I like the idea of disaster prevention through reading hard disk vibration and all, but isn't this the sort of thing that professionals should be responsible for?
CAn any of you honestly say you would kick a few hard drives just to pis the thing off?
Always back up, never back down. ---- Think you're cool 'cos your uid is prime? Take mine, modulo the one digit integers
Helping people avoid natural disasters is a good thing but I cannot see this program taking off too far. Too many people will fear what this progam could possibly be looking at other then just the vibration data. not to mention the possible exploits available to those looking use is maliciously.
I often have trouble remembering which way is out of bed in the morning.
In the name of Science, of course.
If Jesus wants me it knows where to find me.
*** DISK ERROR ***
[A]bort, [R]etry, [F]lee to high ground? : _
Learn from the mistakes of others. You won't live long enough to make them all yourself.
Same thing here actualy.
-- Lattyware (www.lattyware.co.uk)
If we all jump up and down at the same time we CAN cause an earthquake. I always knew it.
Seismic waves range between 40 and 200 Hz.
Immediately we have electrical interference at multiples of 50Hz. If this is supposed to be a delocalised system eg, SETI@home running in geeks' houses we're going to have much more interference: kicking your computer, bass speakers, fans..
I tried this out on my Macbook.
http://www.suitable.com/tools/seismac.html
Does a similar thing. Once caveat: you can never touch the mac. So it's useless really but an interesting demo of the motion detectors. They are suprisingly accurate. Footsteps nearby show up for instance.
spoonerize "magic trackpad"
There's an earthquake every 5 minutes in Queens under the 7 train!
rooooar
This could shake the very foundations in which hard drive platters are mounted!! Err...
I'm running this thing right now, after downloading from here. It's kind of neat, but it'd be really nice if it gave some sort of feedback to the user to show it was actually operating. I'd like to be able to, say, kick my computer and watch a little seismometer guage move around, just to let me know the thing is working.
;)
Also, to the commenter who was worrying that things like kicks or shifts to a computer would result in false alarms, that's part of why they're using a P2P network. By aggregating the results from many machines, you can toss out false alarms. Of course, if a bunch of people got together on IRC to coordinate times at which they'd all kick their computers, that could probably trigger it...
Too bad it's illegal in Spain.
Anagram("United States of America") == "Dine out, taste a Mac, fries"
Do you know how many false alarms my wife will set off when she's stomping through the house to yell at me? (you never said anything about not letting the kids play with Mr. Chainsaw, dear)
Or how many my coworker will generate as he shakes the freaking floor when he walks around?
Ooh jEeZ. HeRe hE cOmEs aGaIn
--- This
This will work well until a few idiots playing with "MacSaber" make me have to flee town due to the forth coming Tsunami!!
Unstable Apps: Our Android Apps Don't Suck
In the absense of further knowledge, I'm somewhat skeptical about the hard drives being sensitive enough, but I mean that in the original meaning of "skeptical", as in, updated pending further evidence, not forever committed to not believing in it. Clearly, this guy thinks they are sensitive enough.
But if that hurdle can be cleared, processed correctly the data will be very useful. Most objections Slashdotters are going to raise will be irrelevant. Local aberrations will be cancelled out at the supernode, because the aberrations will only appear at that one node. Simple interference at constant frequencies is also easy to detect and mask out with "Introduction to Signal Processing"-level signal processing.
Merging the data together is a bit more challenging but should be doable.
The only thing I don't see is talking about knowing where the machines are in the real world, which would be very helpful, and that may be coming later. The other thing is that the system probably won't work very well with a simple "IsEarthquake" signal coming out of the clients; the supernodes really ought to examine all the data from its clients and then decide if there's an earthquake. Otherwise, several correctly-timed local abberations could all look like "earthquakes", even with completely different characteristics, if all that is going to the supernode is "IsEarthquake". Of course, the real system may already have both of these things covered and the article merely oversimplified.
Upshot is, signal processing can do some very surprising things with data that seems to consist almost entirely of noise, if you have enough data coming in.
What happens when everyone running this thing jumps up and cheers (or yells) after the first touchdown of the Superbowl? (or this weekend's Ohio State vs Texas game?)
What about the random noise that could be caused by rushhour traffic past someone's apartment/office building? Or even just your furnace/air conditioner turning off and on?
Karma: NaN
It probably gathers a user profile (SSN, banking details, passwords etc) to build a statistical user model that is used to generate cross-correllated weighting data for the best fit matrix.
Engineering is the art of compromise.
By suggesting that free software isn't "professional" software (and that the creators/suppliers of free software, since they don't get paid for it are amateur rather than professional) you've invited a firestorm from the FOSS weenies; they'll be along any minute to roll a nine-sided die and lay a +4 smack down on you quicker than you can say "flux capacitor"!!
So SEED please, or people will DIE!
Does a similar thing.
No, not really. That application just shows a graph; this system collects and correlates data from many systems.
Once caveat: you can never touch the mac.
Again, no, not really. The system described (not Seismac) correlates data from many systems, and an earthquake will affect many systems. Your typing, jumping up and down, or even a big truck rumbling by, will not. Nevermind that earthquakes have a very charachteristic vibration, so individual nodes are unlikely to be fooled easily in the first place. The supernodes would look for correlation.
To really dumb it down: if an earthquake happens, PCs will see the same/similar vibration in an expanding circle pattern. Similar systems are used with microphones in some cities for gunshot detection- many of them can, with just a few 'listening stations', pinpoint gunfire to within a dozen feet.
Please help metamoderate.
Tsunami detection systems have no problems whatsoever with detecting eartquakes/big eartquakes. The tricky part is determining which earthquakes produce dangerous tsunami wave.
In some place sea bed/shore line may magnify tsunami coming from specific directions, it complicates things even more for predicting very local tsunamis.
There have been false tsunami warnings on the record when big eartquakes have "failed" to produce predicted tsunamis.
Part of the source code has been leaked, looks like the program is a little less sophisticated than we were led to believe:
." seconds.\nIt is suggested that you save whatever you are currently working on, including your life if possible.\n";
while(1)
{
if( &hitByTsunami() )
{
my $tsunami_speed = &determineTsunamiSpeed();
for my $user ( &getAllOtherUsersOnContinent($my_continent) )
{
$user->sendMessage( "You will be hit by a tsunami in ". ( $tsunami_speed / $user->distanceFrom($me) )
}
}
}
How long before a group of 400 users covertly plans to shake their desktops at a precise time some day? What happens then? Is there a check against other clients running in the same geogrpahical area to determine if they are experiencing the same vibrations?
"I threw up my hands in disgust and wondered if it had been such a good idea to have eaten my hands in the first place."
And larger scams have bigger still, and so ad infinitum
Can anyone say "social engineered virus on a Slashdot scale?"
Do not mock my vision of impractical footwear
"If an earthquake that could lead to a tsunami is detected, the supernodes inform the other nodes. Computers running the client software and connected to the peer-to-peer network can then warn of such events. The software is able to provide such warnings because the seismic waves produced by earthquakes travel at about 5,000 kilometers per hour, while tsunamis move much slower at 500 to 1,000 kilometers per hour"
"'Ere! That swappable hard drive just ejected itself."
"Really?"
"Yes."
"It works! It works!"
Oh, say does that Star-Spangled Banner entwine / The myrtle of Venus with Bacchus's vine?
Free Seismology Programs for Windows:
Seismic/Eruption, Seismic Waves and data retrieval. View earthquakes and volcanic eruptions in close to real time. Developed for the Geology Hall of the Smithsonian.
AmaSeis, a program to obtain seismographs from the AS-1 Amateur Seismometer. The AS-1 is based on a classic project from Scientific American's "The Amateur Scientist."
EqLocate. An interactive program to locate earthquakes.
Teleseismic waves recorded to monitor the earth's seismicity have energy primarily in the 1 to 100 second period range (1 to 0.01 Hz). Higher frequencies are attenuated; attenuation is a function of wavelengths travelled. Lower frequencies/longer period waves require bigger earthquake source regions like those associated with tsunami generating events.
The HD sensors will need to be able to accurately monitor long period signals. I'm not sure they are capable of doing so, and secondly, the ambient noise around the computer would be a big problem.
Others have noted that not all earthquakes generate tsunamis. This is correct. For long-distance tsunamis that pose a risk, a magnitude 8 or greater earthquake is generally needed; there are about one earthquake of this size per year. Additionally, the location and characteristics of the earthquake (e.g., shallow focal depth, slow rupture speed) play a role whether a tsunami is generated.
Locally generated tsunamis can be triggered by smaller earthquakes (e.g., 1973 Kalapana event in Hawaii). Additionally, massive underwater landslides can also generate very big tsunamis (e.g., Storrega slide off of Norway).
There is already a global network of seismic stations that monitor earthquakes. First the event is detected, then a preliminary location is computed when enough stations have recorded the event. The magnitude and focal mechanism (seismic moment tensor) of the event can then be estimated based on this location and the characteristics of the seismic signals that have been recorded at the various seismic stations.
Those monitoring tsunamis base their initial assessment on these information. They can then more closely analyze the various seismic waveforms to better refine the preliminary information. If warranted, they then contact previously designated POC's to issue a tsunami alert (this was a problem for the 2004 tsunami, there were no designated POC's for the Pacific Tsunami Warning Center to contact).
How a P2P network could accomplish all of this remains to be seen.
On a side note, I used to play golf with the current head of the PTWC many, many years ago.
That is the normal range for velocities of propagation of acoustic and elastic waves through rocks. Qualitative table at http://web.ics.purdue.edu/~braile/edumod/waves/Wav eDemo.htm . For a typical cross-section through crust (output from seismic tomography) check http://www.ess.washington.edu/SEIS/PNSN/REPTS/Sum9 7/G03084B.jpg . Look at the color scale on the left for values.
Of course this is for the definition of "much slower" that includes giant waves travelling at 500 to 1,000 kilometers per hour.
Erm... Hong Kong South Korea Taiwan Bankok Manila Australian/New Zealand cities Call centers all over India That should be enough to monitor the Asia-Pacific coast and Indian ocean, if the technology actually works. You know, as well as Warp Drive, we'll also have to improve the level of our geo-political awareness or the Vulcans will think we're too stupid to explore the universe and bring peace to the galaxy...
Actually, I've got a flight leaving in four hours to start a consulting gig for an island nation the South Pacific, and the situation there is pretty much exactly as I described it. My comment was mostly informed by my travels to the Pacific Islands, where I've spent some time, but I've traveled fare more widely than that, and I can tell you what I state is true at the very minimum for most of Indonesia, China and India, i.e. about half the world's population.
And I've gotta say, if you consider South Korea, or any other top 30 HDI country, to be a "developing," then it's you who doesn't get out very much.
I think there is a world market for maybe five personal web logs.
It's just the vibrations from a million childen cranking their $140 laptops!
Some settling may occur during posting.
This is a flawed system. It has too many assumptions. It assumes that a little problem with the hard drive is automatically an earthquake. It could just as easily be from too much bean-dip and burritos for lunch. It also assumes that connections stay up during a quake and that the roof doesn't come down. Neither of these is true in a magnitude 6.0 or greater quake.
One must take care with this sort of thing. Many years ago I worked for an organization which, among other things, kept track of geomagnetic activity at high latitudes (the ground signature of auroral activity). This was done by analyzing data from a small network of ground-based magnetometers (measures changes in the local magnetic field). Over time, we noticed that once site, locate on an Air Force base in Maine, always showed a disturbance around 2:30 each Friday afternoon. The people who ran the system for us, including the maintenance folks, were weather experts and were clueless about magnetometers, but they were adamant that everything was working fine. We finally discovered what was happening when one of the group was in Maine on vacation. He dropped by the site for a visit on Friday afternoon and found the culprit. The sensor sat out in a large grass-covered field which was mowed using a large, electromagnetically noisy, lawnmower every Friday afternoon about 2:30PM. Just because you have a signal doesn't mean you understand what you have.
I see more potential in P2P networks for warning users for vulcanos or tsunamis than in detecting an earthquake. Exploring hard drive sensors may be interesting but not very reliable. However, a P2P network, combined with geotagging, may be a good way to target computer users of an imminent event. At least, it could be an additional media to disseminate the information and complementary to TV and radios.
it'd be interesting to see what happens the first time this thing goes off. how would people react? would they trust it?
even if i programmed it myself, i dont know if i'd trust it enough to risk mass panic by notifying others.
This software is a cool idea since the information can come from many nodes so error correction can sort out the many false positives from things that make the floor shake. On interesting false positive story comes from the University I attended - a University that was fortunate enough to acquire an abandoned silver mine less than 5 kilometres from it's main campus. The geophysicists took advantage of this to set up a seismograph some distance underground - but after recording what looked like several large earthquakes nearby in the first couple of hours they looked at the train timetable.
I'm a computer wrangler and not a geophysicist.
So, do we now have a justification to put a compatible disk in the children's laptop?
This is a really creative and possibly practical idea. I'm quite impressed! Whoever thought that one up deserves the Hacker label, in the good sense!
"I object to doing things that computers can do." -- Olin Shivers, lispers.org
only if it traveled on the serface.
OTOH, if you have an earthquake that is going to be felt byt he whole world, the follow tsunami will be the least of your problems!
The Kruger Dunning explains most post on
Sure. Except: The number of households and businesses worldwide with high-speed DSL Internet connections has hit 26 million -- with South Korea, Taiwan and Hong Kong leading the way..
From the same article "The full top 20 looks like this: 1. SOUTH KOREA 2. TAIWAN 3. HONG KONG 4. Belgium 5. Canada 6. Denmark 7. Germany 8. SINGAPORE 9. JAPAN 10. Sweden.... "
Other data puts NZ, Australia, Japan and S. Korea in the top 20 per head of population.
"It doesn't cost enough, and it makes too much sense."
Indeed. The problem with this project is that the countries where computers are very common (say, the United States) already have effective tsunami systems. The IOC offers tsunami warnings as well in most of the Pacific, and are extending their coverage. The problem is that the third world governments often have problems with disseminating the information, and it's these very same countries that also don't have a lot of modern computers with motion detectors. The places that have effective end-to-end tsunami warning systems don't need this software, and those that don't won't have the hardware necessary to use it.
There's also the concern about sensitivity, of course, including false alarms by people moving their laptops, bumping desks, walking around... some serious testing would be needed to ascertain both the precision (reliability) and accuracy of this system before believing what it says.
The Freelance Wizard
I can't *believe* the number of straight lines in this post. Just stop now, ok?
Do not mock my vision of impractical footwear
A week or so ago there was a thread here about stolen computers calling home. The problem still remaining was that most of them did not have GPS and could tell anybody where they were. However, if there were some central database of siesmiographic info, a computer could figure out where it was by triangulating the data from known locations.. Then it could call home and say "This is where I am. Come get me."
How about if you have a wobbly desk or live in an apartment? Yeah.... This is a good idea... umm hmm...
RUN FOR YOUR LIVES ALL THE CITIES IN THE US ARE GOING TO BE HIT, OMFG!
Great Intellect...
is this a joke? oh.. come on, listening for HD vibrations? I can't believe it.
Tsunamis are caused by large quantities of mass moving and causing a displacment of water. This is most likely to be due to earthquakes, but also includes landslides or volcanic activity.
Purpose built seismometers can detect minute vibrations caused by the passage of seismic waves from an earthquake on the other side of the planet. The stations where they are based are specially constructed to isolate the seismometers from any sort of local environmental noise.
What will actually help with real-time tsunami warnings is faster and more accurate analysis of the triggering earthquake. Knowing the displacement that occurs along a fault means you can make a more accurate estimate of water displacement, and hence tsunami potential. From that point, you can model the propogation and size of a resulting tsunami, to allow better warnings to be distributed to shorelines. A magnitude 9 earthquake could occur underwater - but if it's a transverse fault, and the two side of the fault are moving past each other horizontally, then there's practically no chance of a tsunami being triggered.
Trying to analyse hard drive vibrations and is somewhat optimistic. For a start, seismometers have a very wide frequency-response range to vibrations, and are purpose-designed for this - the same can not be said for hard-drives.
While there are clearly many, many times more hard drives out there than dedicated seismic stations, the supposed benefit of having so much data would most likely be cancelled by the massively increased time required to extract any sort of useful signal. Especially from such a narrow frequency response, that cannot distinguish noise from signal (eg; a truck driving past your house, and actual seismic signals). And this is without even getting to the fact that tsunamis are vibrations in a liquid - when vibrations from tsunami reach your location, that's because the tsunami is already there.
This seems like a case of someone with technical knowledge/ability not grasping the physical science basics behind the actual problem. Someone's noticed that PC's have (very limited) vibration sensors in them, and jumped on the shared p2p bandwagon to solve the problem.
Improved tsunami warnings will come from improvments in dedicated seismic monitoring for early warnings of a potential tsunami in the local region near the earthquake. Followed by fast, accurate modelling of tsunami propogation, to allow reliable warnings to be sent to coastlines at greater distance. One of the problems with early warning systems for natural hazards, is that if they aren't accurate, then their effectiveness in the future is reduced.
Sending data from actual seismic monitoring stations to a p2p network might help with improved earthquake models. But the time-critical nature of the problem means that such analyses are probably better left to dedicated supercomputers or clusters.
You run the software and imagine you will help save the world. But it won't help if no government officials are monitoring it, and they have their own systems already. What it does to the internet is flood the net with useless p2p data, and sparks a few communities or forums where is nothing to talk about. "Look, yesterday's peak vibration was at 66Hz!". Blogs full of one liners "Tsunami is coming tommorow morning!! Run!"... Those can't help, but will create panic and chaos.
What is worse, it creates a negative effect of increased fear. As if we were not being scared enough with superfluous information about disasters and crimes all over the world, terrorist threat and suspected nuclear developments in "unfriendly" countries. You run the software all the time, because you fear tsunamis. Effectively, you are expecting a tsunami anytime. But they happen what, a few times a year in different regions of the Earth? It's not only that false alarms can ruin your day, just the fact that you run the software will increase your paranoia degree. Despite the intended purpose to make you feel safe...
I'm sure I'm not the only one with simple hard drive isolation in the form of rubberized mounting brackets. I wonder if they took this and more extreme examples into account; the very nature of them being vibration isolation (for noise reduction).
Always-on DSL is available in most large Chinese cities, although the price is such that most people wouldn't opt for it (prefering, instead, cheaper, metered plans that better fit their usage habits.)
Of course the price isn't really excessive -- it's between 130 and 200 rmb per month -- but if your monthly salary is only 2000 rmb it might be expensive.
Still, I think there are enough Chinese computer geeks for this sort of thing to work, assuming the technology works.
Presumably the same is true of India. Remember, just because there are lots of people who are living in poverty in these two countries doesn't mean there isn't a relatively large (and growing) middle class. Add to this the fact that both cultures are pretty computer-obsessed, and I think that you'd have pretty good coverage. It's not like every person has to have a computer running this thing. Even if only 10% of the population can afford it, that's still a huge number of people. Remember the populations we're talking about here.
I can't comment on Indonesia, as I haven't been there. I lived in China for four years though.
Just out of curiousity, how is it supposed to correlate IP address to physical location in the world?
I thought that all the attempts to connect IP to physical locations had pretty much died of non-maintenance, and impossibility of getting all the location information from ISP's in anything remotely resembling realtime?
that seems like an awful big hurdle to the operation of this thing to me...
has something changed?
ìì!