Yes, very old and well-known technique, but I think this patent is based on putting the two frequencies to specific use in the logic. Otherwise it would apply to every frequency synthesizer, and this fraudster would be taking on much of the world, not just Intel. HP would of course be a prime target, they still make frequency synthesizers.
But there is going to be lots of prior art anyway, no need to sell Intel shares over this.....
I usually use AMD, no doubt they will be the next target.
In most countries, you just publish the invention, and no-one else can then ever patent it, including yourself after a short time. Copyright issues can also be dealt with at the same time, you could put it under GPL or Creative Commons, or some other licence. In fact, the licence you use could explicitly disallow patenting, if you wanted to be sure, and I think that would have effect in the US, where publication alone does not.
In fact the US system is very wrong, as has been said on/. before, because someone else can come along some time later and patent your invention, which would not be acceptable anywhere else.
You are quite likely to be correct on all, or most, of that, but also the patent makes explicit mention of a board (containing new CPU and logic etc) being plugged into a socket. That is entirely different to having it all on-chip, the patent appears (IANAL) to not include that, so it is irrelevant anyway, and for the reasons you say too. But, the 4MHz is only mentioned in one of the claims, unless I am mistaken, so it is not really relevant, it is the more general claims that would matter.
If I weer Intel, I would have not offered them 1 cent, and billed them for my lawyer's time in answering them. They are just trying it on because of other recent cases, I think that this one, as we seen to agree, has no merit whatsoever.
IANAL, but I did think that a patent had to contain some kind of inventive step. This one does not, it is well-established engineering, with nothing new at all. It was being done well before 1996. I think there must be lots of prior art, no doubt the resources of Intel will be able to find plenty.
They have worded the patent very badly, so that it clearly applies only to a plug-in accelerator. Intel don't make these! Also, it looks to me as if it is easily worked around by soldering the processor in, instead of using a socket, and in any case the replacement processor seems to have to be part of an "accelerator board", like those useful things (Evergreen?) that had a K6/II-350 on board and would go into a P100/P133 socket. It does not seem to cover the case where there is one design of motherboard, and one processor, not on its own board.
This sort of nuisance show why there is all the more reason for urgently reforming the USPTO, which I hope Dubya's elected successor will do as a matter of urgency. This sort of stupid and vexatious claim is becoming far too common. No simple arrangement of a small number of commonplace hardware building blocks should be capable of being patented, there is inevitably prior art, even if it is hard to find, because most things are never published.
Just remembered, did the old Intel Multibus not do something like this, there was the capability for multiprocessing, and they could run at different speeds than the main bus, so of course does VME, and there is nothing to say that you can't dervive all the clocks of all the diverse processors from the VME Bus Clock.
I tend not to like large corporations, especially the greedy sort (Intel is nowhere near as bad as the Criminal Monopoly of course), but Intel deserve to win this one, and I hope it happens quickly, to deter others. Sadly, all of Intel's customers will end up paying, one way or the other, but somehow I doubt that the legal bill will be very big.
Does that mean that the Posix compliance will be real, and not illusory, and that it will read and write "foreign" file systems?
Quick, everyone, patent your little (or big) bit of Linux now, while you can......
Only joking of course, I doubt that what Sir Bill has between his ears is capable of grasping how extensive and powerful the facilities provided by Linux really are.
If they want to make a *nix-like system, they will face serious sompetition, from IBM, Sun (now remember how quickly Bill fell out with IBM, the same will happen there...), the SCOundrel even, all of whom do it better than the Monopolist would ever manage. I have not forgotten Xenix, it was truly pathetic. But, by the time they ever get anything out of the door, SuSE, Red Hat, Mandrake and others I can't be bothered to remember will be selling so well that a new entrant will stand no chance.
The first rule of marketing, so I am told, is timing. Longhorn is late and getting later, if they hack it about and take bits out to speed up the development, it will get later still. So they will fall short on the first rule.
Utter rubbish, from someone who is obviously ignorant and ill-informed. When Marconi was making his substantial contribution to the development of radio, his first equipment operated on similar frequencies to mobile phones (low microwave), and at first his only arrangement for detecting the signal was a spark gap, in fact a very precisely adjusted spark gap. This was long before any kind of rectifying device, no thermionic valve, no galena and cat's whisker, not even a coherer. He was getting a range of well over a mile.
Now, two bits of metal, the nozzle and the car, moving relative to one another, make a spark gap potentially much finer than anything Marconi could achieve by adjustment, as they break contact momentarily, the gap is as near zero as possible, a very small voltage will jump the gap.
As for your useless and ill-conceived experiment, you have totally failed to see the point. If you lightly brush the antenna against the plate, which this source of ignition does depend on, you will in fact see sparks, sams as if you use DC and a small battery. it is the intermittent contact caused by the movement of the nozzle into the tank that creates the conditions for the spark. Yes, Marconi could get the right conditions with careful micrometer adjustment, it is far easier to get it momentarily by random movement. But, I see that all this is well beyond your intellectual capacity.
And, you are wrong about the metal in the hose, at least in the UK. It is deliberately arranged to have a high, but not infinite, electrical resistance so that in the other high-risk case of someone who is charged picking up the nozzle, the discharge current is limited to a milliamp or so, not the 100 amps peak that can, and does, happen when you touch bare metal. And you will not believe that current either, as you clearly have no comprehension of anything above DC, but it is roughly what an ESD gun, which mimics the human body, will produce. About 100 amps peak, for a few nanoseconds.
And, furthermore, if you had been anywhere near any serious RF work, or EMC testing (which I have BTW), you would know that the side of the car crudely resembles a ground plane, while the long, flexible hose, even if it was copper, is utterly useless as a ground at anything above a few MHz. I never suggested tha car was a "perfect" ground plane,of course it is not, but the length of the hose is many wavelengths, it is nowhere near a ground conductor. In fact, a metal hose would, depending on how it was arranged, likely have a resonance on at least one of the mobile channels, it could make the problem very much worse.
But as you wish to remain ignorant, and are certain that without a single scrap of knowledge or experience to back up what you choose to believe, I suspect that you will blow yourself up at a petrol pump quite soon. Just don't do it near me.
Interesting, and thanks for the link. I wonder how reliable it will be in the long term. It will fail if there is a bad electromagnetic environment (noisy) at either end. If on eend is in a city, I think channel usage by others would be a serious problem, you really need a clear view from your dish with nothing that might interfere anywhere near your line of sight, or any of the sidelobes.
If you can get an infinitely narrow beam (you can't of course, but in general the larger the dish, the narrower the beam) you can go as far as you like, in theory. The practice is not always so easy. Nevertheless these guys have done very well indeed, maybe what you might expect from O'Reilly.
A word of warning though, in most countries (don't know about the FCC in particular) it is your "effective radiated power" (ERP) that is licensed, not yout transmitter power. By using a high-gain antenna, the ERP is increased (you don't get power from nowhere, it is mostly concentrated in the beam instead of being isotropic), so you will in many cases be illegal, even with the modest, and relatively harmless, actual power of the transmitter. You can put as high a gain antenna on a receiver as you want, of course, but when it gets too big, planning laws and other considerations might apply.
Apart from that, I can see there being a good market for dish antennae..... But, if too many people do this, it will upset the big boys of the telecom industry, there might be changes to the regulations. Enjoy while you can!
It just occured to me (my memory is a bit slow sometimes) that about 15 years ago, laser links were in fashion, for example to go between buildings where you could not run cables. Now a laser is, in clear weather, an extreme example of a directional antenna, there is no inverse square law loss, if the spread is sufficiently small that all the beam hits the detector, so it will go a long way.But, you would need very stable alignment, maybe not across the San Andreas fault, for example, or from a tall building that sways in the wind. IIRC the equipment they used to sell, and likely still do, cost about $3000, which is not cheap, but not too silly either. But, not good in a location where adverse weather is expected!
I think there needs to be world-wide agreement and coordination on issues like this, because the net is world-wide. Any sane person except certain unscrupulous marketing people, and some politicians, who can't grasp the concept, would agree that spyware should go.
It is unfortunate that the IETF, or W3C, or some other democractic body who understands the issues can not be delegated the task of setting internet law. It should not be left to local or state level either, it is an international thing, it needs an international solution. Nevertheless, this is a good start.
In some places this is arguably covered by existing laws, for example in the UK the Data Protection Act might have some effect, but it also might criminalise the unfortunate person whose PC got infected with spyware and so disclosed someone else's personal data. That is why a uniform approach is vital.
That was very good value for money! I live in the UK unfortunately, our telecom monopoly (OK it isn't any more, but only in some places) could not supply a short-term connection for that price, far less all the servers and things.
For servers, if you do this kind of thing often, it would pay to accumulate obsolete PCs (a lot of people seem to be disposing of 733MHz Pentiums thse days, they don't cost very much at all), for use as servers, they are perfectly adequate for Linux or FreeBSD. That way you could keep them all properly configured, just need to change the gateway/router config each time, and it is a good use for what might otherwise become landfill. I am collecting some for a different purpose, large offices are regularly disposing of them, no doubt because Sir Bill's latest bloatware is too slow and needs 2 or 3 GHz. What a waste! If it has only a 10MHz network, putting in a 100MHz card will not exactly break the bank, in fact if you put in several cards, filling all the PCI slots, you have an instant router/bridge/firewall (depending which you need). You sometimes also see 100MHz ethernet switches being disposed of, maybe they have moved to GHz or need more ports, or whatever. They are definitely worth acquiring.
I would advise allowing plenty of setup time, something will go wrong. Best to be ready at least a day before people arrive, if possible.
Don't know about regulations in the US, but in the UK we have some provision around 458MHz and a few other places, but the bandwidth is really only 9600, or maybe 19200. We also have bands around 1300MHz and IIRC 2400MHz, for video links etc, which could carry data, but the radiated power is limited because the band is shared with other things. You will not be able to extend a broadband thing like WiFi (or even worse, Bluetooth) very far directly , even with directional antennae, so you need the other options. There are mircowave links of the type used by telecom operators, most likely vastly too expensive for a one-off.
Why not try negotiating with the site owner about having the system installed permanently, that way the costs can be recovered from users over a long time, so an expensive solution might become viable?
If you can get an ADSL line to the site, it gets easier and you only need standard stuff, firewall, router, WiFi access points and so on. It would be best to have a caching proxy server, it can take a lot of load off the line as many people might be looking at the same sites. All of this is straightforward Linux/BSD/Windoze (if you really must) territory, the only thing difficult is the telecom infrastructure. Oh, and watch out for electrical safety regulations, there are lots of potential difficulties, and you reall do need to know your local regulations, and comply with them, or use low-voltage batteries for the lot.
I would ask the local telecom provider how much a decently fast line would cost before messing with satellite systems. The people who supply networking, cable TV, or voice comms in that geographic area are the people who know. In the UK, which does not help you very much, BT can and will provide a voice line or lines (you could aggregate the bandwidth with multiple modems) anywhere, or an ISDN line in most places, ADSL gets slightly harder because of the distance limit. All this is at a cost, of course......
You have totally misunderstood the point. It is nothing to do with walking through fumes with a phone (although that exposes you to a different risk). It is entirely to do with a situation where a spark is made to pass between the nozzle and the tank, by any means, whether RF or static. The pump hoze is designed with a high resistance, so if you pick it up while charged, a small current will flow harmlessly to earth. The car should also discharge itself to earth potential, there are conductive additives in the tyres nowadays. But you can't get rid of RF energy so easily.
You can walk about in fumes all day, or use the phone all day (assuming no sparks inside the phone, batery contacts, switches, etc) without anything happening. It is the fact that two bits of metal, the pump nozzle and the tank neck of the car, are in very light (i.e. imperfect) contact electrically that creates a place where sparks can be generated.
If you took a small battery, say 6 volts, and connected one terminal to the pump, and the other to the car body, with a resistor to limit the current to a sensible level, say 10 to 100mA, and then poked the nozzle about, in and out of the tank, how long would it be before there was an explosion?
Probably not the first time, but 10, 100, 1000, 1000000 attempts? Would you try such a thing, while filling the tank?
The fact that RF from the phone is energising the thing instead of DC from a small battery, is not of much importance to what would happen, except that the connection to the phone is less obvious, and most people seem to be unable to see how the voltage arises. They seem to think that I have suggested direct ignition by the RF field, well no, that will not happen, and I never suggested it would.
If you are charged with static, and walk up to a car being filled, and poke your finger at the tank, you might well cause an explosion, but if the same person gets out of the car, removes the filler cap, picks up the nozzle, and puts it in the tank, the chances of a static spark are minimal because everything will by then be at the same potential.
You may be an electrical engineer, but your knowledge clearly is less than zero above the 50Hz or 60Hz power frequency.
At UHF frequencies, the car body is almost ground, by virtue of its capacitance. Yes, at 50Hz, it may well be insulated by the tyres.
The pump nozzle is on the end of a rubber hose, with some built-in conductivity (high resistance) to reduce the effects of static discharge. It is nowhere near ground at RF, even if the hose was metal, it is long and has a significant inductance.
Your assumption sabout free space impedance are also utter junk, the thing that matters is the apparent impedance at the end of the bit of metal, which can go a lot higher. What is the impedance at the end of a dipole antenna, at resonance, for example? What is the voltage at the tip of a resonant CB antenna, with 4 watts input? By your reasoning, it is 38.8 volts, it can actually go over 200 quite easily. The free space impedance is a mathematical fiction to enable wave propagation to be worked out, it does not have physical reality in terms of current-carrying conductors, but it does give the right answers in free space.
And furthermore, as I have said all along, the sparks you get are because of making monentary contact between the nozzle and the car, if you were at all competent, even at DC, you would know that you can make a spark with a single cell battery in these conditions.
When people like you resort to bad language, it is because they either have no coherent argument, have lost the argument, or are just plain incompetent. Any proper electrical engineer would be aware of capacitance and inductance, resonance and transmission line effects.
Furthermore there have been explosions thought to be caused by police radios, in many places they are banned from using them in petrol stations for that reason.
It is interesting that none of the abusive comments have come from RF or even EMC engineers, but only from the ignorant, ill-informed and in your case, just plain rude.
Yes, your statistics might be about right, within a order of magnitude either way, but if a simple action can prevent an accident, it should be done.
Some types of accident are easy to eliminate, this being one. But where statistics and reality diverge is that some things you can control, others not. You, as an individual, can choose to drive more carefully, and I hope you do, but you can't control the other guy, who has a fairly good probability (in terms of the sort of numbers we are interested in here) of being drunk, or asleep, or having faulty brakes. So, you have reasonable control over one subset (your actions) of the inputs into the accident scenario, but no control over the others. At the filling station, you have full control over one input which can totally avoid the risk.
Now, curiously, people usually have not the slightest clue about managing risks, it is well established that in the UK, rail is the safest form of transport, with air second, by a factor of about 2 to 5. Worldwide, air probably wins as you suggest, in sparsely populated regions the rail infrastructure can't be maintained to the same standards for economic reasons, in a very densely populated country like Japan, rail does even better relatively, because it economically viable to achieve a very high standard). Now, in the UK, which I know best, following a rail crash (typically there are only 3 or 4 fatalities, and then less than one per year, so the picture is easily distorted by one major incident), huge numbers of people take to the roads, when all the statistics show that it is far more dangerous, as a result about 100 extra fatalities might occur (3500 per year on UK roads, 1000 due to drink alone). All because people do not understand the risks. they are willing to accept a much higher risk, because thay imagine that they are in control, than the obvious but much very lower one, where thay are not in control. They don't see the difference between avoidable and unavoidable risks, and they assume that the ones where they have some semblance of control will not happen.
All I am saying really is that if one risk can be totally eleimiated, why not do it? It is easy. Our Health and Safety Executive makes laws and rules which use the term ALARP (as low as reasonably practicable) quite a lot. That means that you are without excuse if you did not get rid of the easy ones, some more difficult risks cannot be changed very much. Allowing mobiles in filling stations does not comply with ALARP, because it is "reasonably practicable" to not allow them. It costs nothing, and needs no new technology.
And yes, looked at objectively, we should ban cars, for many reasons, but I hope we do not. But, there is no argument that I could advance to justify their existence, except for emergency services etc. Maybe I would love to be able to get rid of my car, if public transport could get me to work efficiently. Approximately 1/4000 of a life would be saved each year if my car was to go. But, my employer is not located near public transport, so that is one risk reduction that is not "reasonably practicable".
I don't think anyone yet knows how they will compare. Different frequencies, modulation patterns, and duty cycles, all change the ignition hazard, the health hazard, and the EMC hazard, and not many people have 3G phones yet. There are also very likely to be significant differences between the 900MHz and 1800/1900(depending on country) 2G phones, most people I know who have an opinion, think that the 900MHz will present the worst ignition hazard, while the others may be the worst health risk. If the 3G frequency is higher, it "might" be even more so. The health risk is another matter altogether, all the tests so far, at least most, have been worse than useless because they have not used the correct modulation patterns, some only unmodulated carrier, and the results have been all over the place, with evidence of detectable effects (whether they are good or bad, they can't tell), or with no evidence, or with evidence of effects beyond what could reasonably be expected.....
My opinion is that there will eventually be shown to be a small health risk, minimal to those who use the things sensibly, as is thought to be the same for all RF fields. The Russians, not known for concern about public safety, used to, and presumably still do, have microwave safety limits 30dB (or was it more?) below those of the western world, that in itself is what makes me think there is a small risk, but the issue is now so badly obscured by vested interests on all sides that it will be a long time before the truth is known. Meanwhile, I use mine (not 3G) occasionally, maybe once a month, for a few minutes, because that is all I need, it is unlikely, I hope, to present a serious hazard. But, it is a dual-band Nokia, it does not tell me which band it is using at any time. It would be interesting to know.
What I do know, is that using an earpiece is bad news, the extra coupling of RF from the phone to the head, as a result of the wire, was fairly convincingly demonstrated some time ago. My colleagues, especially the EMC engineers, were not surprised, we were well used to the phenomenon.
Make sure you use a Farady cage, such as being fully inside a metal-bodied car with fixed roof, otherwise you may discover the effects of ground resistance, and the voltage gradient which can be lethal some distance from the strike, especially on soil of poor conductivity or bare rock!
You can safely sit in your car in the worst thunderstorm, it might even be fun, but you might get burned paintwork, possibly even dented metal, and damaged tyres, if there is a strike. And beware of damage to the electronics, you do not want a protruding antenna of any sort, if you must do that sort of thing.
Given proper precautions, observing thunderstorms at close quarters is a very safe thing indeed, otherwise very unsafe. There is no middle ground here.
You miss the point completely. Statistics for such things are not gathered centrally by anyone. If you research newspapers worldwide you will find a significant number of cases. But everyone that I know who work in safety-critical industries (say 50 engineers) all agree that these fires and explosions are a very real risk, although they only happen on a statistical basis. Not everyone who smokes will get lung cancer, that is also, as far as can be determined, on a statistical basis, but it would not thereofre be sensible to advise people to smoke becuase it is safe, when it clearly is not.
The evidence of ignition of fuel vapour as a result of RF energy causing sparking has been embodied in safety legislation for many years, the research would have happened from the early 1900s onwards, as soon as the explosions started happening. I don't have access to a comprehensive library of past research papers, but some people reading this probably do. There are data published about ignition energies, flash point temperatures, and so on in reference books, available in good libraries. There are even things like the statistical distribution of the ignition energy, it is a probability distribution, on rare occasions an event of well above the ignition threshold might not cause ignition, on other occasions something well below the threshold will. The measured value is a mean, the limits statistically may be quite broad. I don't have access to a university library nowadays otherwise I would look all thi sup for you. What I do know is that in most countries (not, maybe, the US, where they even control oil refineries with Windoze NT, according to a previous/. post some time ago) there are laws, or rules established by the industry themselves, or both, which have evolved as a result of laborious research, and sound theoretical calculation. These laws apply to things like oil refineries, chemical plants and so on.
Now in a refinery, there are strict rules about electrical bonding between sections of pipe etc, to prevent the possibility of sparking at the interface, and exposure levels to RF are set accordingly.
Some years ago, a small local radio station in Scotland, which had occupied their site for a number of years, was forced to move by an oil company which wanted to build a refinery nearby, because of the RF threat to a fully bonded installation. The field strength at the oil company's boundary was similar to that of a mobile phone at 10 to 15 feet, yet it was perceived as a hazard to a fully bonded installation. You do not have electrical bonding between the pump nozzle and the car, which makes ignition much more probable, and is precisely the reason for my opinion. Ask yourself, are you allowed to use a mobile in an oil refinery, and if not, why not? They don't allow it even although teh structure provides good electrical conductivity, because they know and understand the risks, and so does their insurance company, and the legislators. A gap between two bits of metal adjacent to the flow of petrol vapour out of the tank as you fill it is a very bad scenario, compared to any regulated and controlled continuous structure of pipework, the riska sre far higher. Analysing risk involves understanding what can happen, even if the risks are statistically improbable, but not zero. It does not mean dismissing a risk because you "think" it can not happen. The risk here is fairly improbable, but not zero. Lots of equally improbable things can, and do, happen, and cause accidents.
I drive by the laws and advisory guidelines which apply in the UK, it so happens that per head of population our road accident rate is a fraction of that in the US, in fact the lowest of any developed country (it could and should be much better of course, there are still 1000 people killed by drunken scumbags evey year for a start), despite our speed limits being higher, so we must be doing something right. Yet we have had oil refinery explosions, which might suggest that the existing legislation is too lax.......
I am glad we seem to agree, and even more glad that some of us remember the days when they tried hard to do things properly, even though the technology was not quite ready. Now they have the basic technology, but use it to make junk, and more junk, and I mean program material as well as hardware......
It does need an adverse set of circumstances, another poster has referred to 7 sigma. Statistically, it does not happen very often, but one preventable accident is one too many, which sums up my perspective on this matter. 100 or more accidents worldwide is too many, and you can't tell when it might happen, therefore don't do it, simple as that, end of controversy, and end of long, boring, loud, irritating conversation from the person in the next car. Unfortunately there is no technical reason for banning phones in most places where we might like them to be banned.
Not the whole IEEE, not a large proportion of safety professionals with RF experience, and certainly not most such professionals on this side of the Atlantic, including those who advised our legislators, quite correctly, based on the evidence available for many years.
And as to your other questions, I have always worm my seatbelt when driving since 1971, when I first had a car with belts. No exceptions, ever. I do not measure the typr pressures every day, they do get a visual check for signs of severe loss of pressure. I try hard not to drive when I am tired, and I use my turn signals, as you call them, in accordance with the rules which prevail in this country, always if traffic is around, probably not if I know that the road is entirely clear. Yes, all of these things have caused deaths, which is why I started with the one which can be dealt with most easily, the seatbelts. It involves neither opinion nor speculation, and was known since maybe 1950 or earlier to be a worthwhile safety measure, therefore I adopted it absolutely at the earliest possible opportunity.
Likewise there is no controversy among informed experts, not those in the pay of one vested interest or another, about using mobile phones near petrol pumps, so I do not do it, ever.
I am glad you mentioned the cat, which only my most recent car has, and that only a little one, because it is a turbodiesel. The handbook contains dire warnings about parking on long grass......
The only time I ever splled petrol over a hot manifold was in a very old car, 1968 vintage IIRC, the fuel then was leaded and possibly not so easily ignitable, but I have heard of it happening to other people also. The sight of a small quantity of furiously boiling petrol is quite frightening.....
I am a bit surprised that one of the several possible sources of ignition did not get it, either the distributor or the dynamo brushes, although the engine did stop when the fule pipe came off, maybe it was stationary with no sparking before the vapour got to either place.
Now concerining sparks with apparently low energy circuits, you could try shorting a fresh 3 volt battery with a piece of wire (just lightly brush it against the contact.) You will see sparks. If you hold the wire on the terminals, it may get hot (depending on the gauge of wire), the battery itself may get hot, and if you persist, in some cases, may eventually explode. This is provable fact. Some cellphone batteries if shorted might momentarily supply a current 0f 50 or 100 amps, some maybe only 10 or 20, but still a lot of current, also provable fact. (Don't blame me if you ruin your battery, or worse). Ask an airline about their regulations for shipping batteries, and why..... (they are relatively safe in your laptop or whatever, but no way, in the cargo hold.)
If you repeat the dark room experiment with a small amount of RF energy instead of DC, you will see the same result, sparks.
You can indeed use armoured bulldozers in complete safety as far as anti-personnel mines are concerned, and it is in fact done. (I have seen it on TV so it "must" be true, IIRC after the first war with Saddam.) But, apparently, some mines are at least semi-intelligent, the bulldozer may pass over it once, twice even, but after a while the next thing to be detected, quite possibly a person, will activate it. Some ignore heavy things altogether. The clever ones presumably have some kind of electronics, so the battery will run out eventually, but if they work mechanically with some kind of ratchet wheel or similar, they might be operational for a long time. It makes me sick that portions of industry in civilised countries were designing things like that, optimised to kill innocent people long after the wars are over.
To be sure of getting all of them you have to either detect them reliably (often no metal parts, which makes it a million times more difficult) or disturb all the soil, damaging crops, trees etc. Footpaths may be mined, they may run through trees, across hillsides, and so on, where it is impracticable to use a bulldozer.
In the case of the bigger mines, you could in theory build a very robust machine, and be willing to repair it quite often. It could run up and down all day automatically, using GPS. It would be much less stressful to its operators, I have heard that operating an armoured bulldozer leads to lots of nervous breakdowns because of the frequent random explosions. I think that might have been on TV some time after Mrs. Thatcher's Falklands war.
If people do have new ideas about this, they should perhaps communicate them to the proper authorities, it really does seem to still be a major problem.
We still occasionally find one of Hitler's bombs in the UK, I know it is a slightly different scenario, but they usually do appear to be still dangerous after 60 years. There was an evacuation somewhere a few weeks ago while the thing was made safe.
I also saw something else on TV a few years ago, again probably in Kuwait, which seemed to detonate an entire minefileld at once. IIRC it was a British invention, likely the US will have it also. Can't remember how it worked, whether it was EMP, or sonic, or what. All I remember is seeing a vast set of almost simultaneous explosions. I think it might not have been totally reliable, IIRC it missed a few.
I think that maybe you asked the question because it does not seem to be done, and I suspect that the answer mainly involves money rather than impossible technology.
Maybe! Have you ever read a book called "Willard". Don't remember the author, it was set in Edinburgh in maybe the 1930s. Worth a read!
Seriously though, I know that animal rights activists will be furious over this, but certain humanitarian issues do have to be tackled, and this may work.
You are absolutely right. Fortunately I live in the UK, it is not a problem here (yet?). But most domestic equipment,especially the older stuff, is junk, and has so many spurious responses.....
Our channel assignments were carefully worked out very many years ago, so that taking into account the standard IF frequency on TV sets (around 38MHz), there would be as few inompatabilities as possible, taking image frequency response and all that kind of stuff into account. In those days TV sets had front-end selectivity, a lovely little 4-gang variable capacitor tuning high-Q circuits, and valves (tubes) with good large-signal handling capability and not so prone to cross-modulation etc.
The junk of 10 years ago is not actually so good, front-end nonlinearity is a problem aggravated by varicap diodes in the tuners. I rather suspect that if they were to fill the locally unused channels here with what is, as far as the TV is concerned, simply broadband noise, most of the moderm sets would be struggling, the very oldest museum pieces with mechanical tuners might cope (if the IFs are still in correct alignment), and maybe some of the very latest might be OK also, but the vast majority would have problems.
Still, Tony B. Liar has a grand plan, he will end analogue TV transmission earlier than originally expected, so everyone has to upgrade to digital, with its problems. Those with the least money will be the most disadvantaged of course, as is always the way with socialism, or any other -ism for that matter.
Now the current digital systems might be better (arguably) than the NTSC system, which we in Europe have long suspected stands for "Never Twice the Same Colour", but they are vastly inferior to our analogue PAL system, especially on moving scenes. (To be fair to NTSC, it was first, and well-established, before the way of improving it to make PAL was thought about. The change would have broken compatability in the US, but was easy where colour did not already exist). So, it is backwards progress, and it looks as if we will never again have a clean, tidy and well-managed electromagnetic spectrum, as was once the case on both sides of the Atlantic.
I don't keep all my old newspapers. I also don't know who does gather these statistics, but the most recent one in the UK happened exactly the way one might expect, the person with the phone had finished filling their car IIRC, and was passing the one being filled. The peole survived, their account is likely to be relaible. There have been moderately good reports published on a number of occasions, one in Holland a few years ago was said to be conclusive. Now it is a bit off topic, but there was a disaster in an oxygen-rich environment in a hospital in Italy a while back, again thought to have been a mobile, as no other source of ignition was present.
Out of curiosity, I will collect these reports from now on.
An eyewitness report certainly is not scientific proof, but if reliable eyewitnesses definitely contradict the science, then the science may well be in trouble. That is how some advances in science happen.
But, in this case, science most certainly says that the events can happen, what is uncertain is the range of possible positions of the phone, concentration of vapour, gap between nozzle and car, etc. It can be said that these events are possible, certainly the energy is thare, and can get to the wrong place, there is insufficient data to say how often a fire will result, or whether it is going to happen in any specific place or time. But the probability of it happening is not zero, whichis why phones should be switched off, no exceptions.
Safety works by analysing what can happen in a worst-case scenario, and then showing that the particular scenario is sufficiently improbable to be ignored. In this case that can not be shown. Maybe if there were no fires at all for 10 years, while everyone used their phone at the pump, it might be established, but still only on a statistical basis.
Yes, you make a good point about 7 sigma events. None of the scenarios I have ever postulated about this are 1 sigma, most likely in the 5 to 7 sigma region, based on the number of reported fires. It takes a lot of things to be within certain bands of tolerance to get a spark of sufficient energy to meet an ignitable mixture, the fact that an ignitable mixture and a spark separately occur on a daily basis means that on occasions they will meet......
I am glad you survived your habit, you are right about issues of vapour concentration, but I was thinking of the situation where a piece of red-hot ash dropped from the cigarette. Somewhere near the tank filer there would probably be an ignitable mixture.
It seems to me to be a bit like standing on a mountain peak in a thunderstorm....
Slashdot Mirror
But there is going to be lots of prior art anyway, no need to sell Intel shares over this.....
I usually use AMD, no doubt they will be the next target.
In fact the US system is very wrong, as has been said on /. before, because someone else can come along some time later and patent your invention, which would not be acceptable anywhere else.
If I weer Intel, I would have not offered them 1 cent, and billed them for my lawyer's time in answering them. They are just trying it on because of other recent cases, I think that this one, as we seen to agree, has no merit whatsoever.
They have worded the patent very badly, so that it clearly applies only to a plug-in accelerator. Intel don't make these! Also, it looks to me as if it is easily worked around by soldering the processor in, instead of using a socket, and in any case the replacement processor seems to have to be part of an "accelerator board", like those useful things (Evergreen?) that had a K6/II-350 on board and would go into a P100/P133 socket. It does not seem to cover the case where there is one design of motherboard, and one processor, not on its own board.
This sort of nuisance show why there is all the more reason for urgently reforming the USPTO, which I hope Dubya's elected successor will do as a matter of urgency. This sort of stupid and vexatious claim is becoming far too common. No simple arrangement of a small number of commonplace hardware building blocks should be capable of being patented, there is inevitably prior art, even if it is hard to find, because most things are never published.
Just remembered, did the old Intel Multibus not do something like this, there was the capability for multiprocessing, and they could run at different speeds than the main bus, so of course does VME, and there is nothing to say that you can't dervive all the clocks of all the diverse processors from the VME Bus Clock.
I tend not to like large corporations, especially the greedy sort (Intel is nowhere near as bad as the Criminal Monopoly of course), but Intel deserve to win this one, and I hope it happens quickly, to deter others. Sadly, all of Intel's customers will end up paying, one way or the other, but somehow I doubt that the legal bill will be very big.
Quick, everyone, patent your little (or big) bit of Linux now, while you can......
Only joking of course, I doubt that what Sir Bill has between his ears is capable of grasping how extensive and powerful the facilities provided by Linux really are.
If they want to make a *nix-like system, they will face serious sompetition, from IBM, Sun (now remember how quickly Bill fell out with IBM, the same will happen there...), the SCOundrel even, all of whom do it better than the Monopolist would ever manage. I have not forgotten Xenix, it was truly pathetic. But, by the time they ever get anything out of the door, SuSE, Red Hat, Mandrake and others I can't be bothered to remember will be selling so well that a new entrant will stand no chance.
The first rule of marketing, so I am told, is timing. Longhorn is late and getting later, if they hack it about and take bits out to speed up the development, it will get later still. So they will fall short on the first rule.
Now, two bits of metal, the nozzle and the car, moving relative to one another, make a spark gap potentially much finer than anything Marconi could achieve by adjustment, as they break contact momentarily, the gap is as near zero as possible, a very small voltage will jump the gap.
As for your useless and ill-conceived experiment, you have totally failed to see the point. If you lightly brush the antenna against the plate, which this source of ignition does depend on, you will in fact see sparks, sams as if you use DC and a small battery. it is the intermittent contact caused by the movement of the nozzle into the tank that creates the conditions for the spark. Yes, Marconi could get the right conditions with careful micrometer adjustment, it is far easier to get it momentarily by random movement. But, I see that all this is well beyond your intellectual capacity.
And, you are wrong about the metal in the hose, at least in the UK. It is deliberately arranged to have a high, but not infinite, electrical resistance so that in the other high-risk case of someone who is charged picking up the nozzle, the discharge current is limited to a milliamp or so, not the 100 amps peak that can, and does, happen when you touch bare metal. And you will not believe that current either, as you clearly have no comprehension of anything above DC, but it is roughly what an ESD gun, which mimics the human body, will produce. About 100 amps peak, for a few nanoseconds.
And, furthermore, if you had been anywhere near any serious RF work, or EMC testing (which I have BTW), you would know that the side of the car crudely resembles a ground plane, while the long, flexible hose, even if it was copper, is utterly useless as a ground at anything above a few MHz. I never suggested tha car was a "perfect" ground plane,of course it is not, but the length of the hose is many wavelengths, it is nowhere near a ground conductor. In fact, a metal hose would, depending on how it was arranged, likely have a resonance on at least one of the mobile channels, it could make the problem very much worse.
But as you wish to remain ignorant, and are certain that without a single scrap of knowledge or experience to back up what you choose to believe, I suspect that you will blow yourself up at a petrol pump quite soon. Just don't do it near me.
If you can get an infinitely narrow beam (you can't of course, but in general the larger the dish, the narrower the beam) you can go as far as you like, in theory. The practice is not always so easy. Nevertheless these guys have done very well indeed, maybe what you might expect from O'Reilly.
A word of warning though, in most countries (don't know about the FCC in particular) it is your "effective radiated power" (ERP) that is licensed, not yout transmitter power. By using a high-gain antenna, the ERP is increased (you don't get power from nowhere, it is mostly concentrated in the beam instead of being isotropic), so you will in many cases be illegal, even with the modest, and relatively harmless, actual power of the transmitter. You can put as high a gain antenna on a receiver as you want, of course, but when it gets too big, planning laws and other considerations might apply.
Apart from that, I can see there being a good market for dish antennae..... But, if too many people do this, it will upset the big boys of the telecom industry, there might be changes to the regulations. Enjoy while you can!
It just occured to me (my memory is a bit slow sometimes) that about 15 years ago, laser links were in fashion, for example to go between buildings where you could not run cables. Now a laser is, in clear weather, an extreme example of a directional antenna, there is no inverse square law loss, if the spread is sufficiently small that all the beam hits the detector, so it will go a long way.But, you would need very stable alignment, maybe not across the San Andreas fault, for example, or from a tall building that sways in the wind. IIRC the equipment they used to sell, and likely still do, cost about $3000, which is not cheap, but not too silly either. But, not good in a location where adverse weather is expected!
It is unfortunate that the IETF, or W3C, or some other democractic body who understands the issues can not be delegated the task of setting internet law. It should not be left to local or state level either, it is an international thing, it needs an international solution. Nevertheless, this is a good start.
In some places this is arguably covered by existing laws, for example in the UK the Data Protection Act might have some effect, but it also might criminalise the unfortunate person whose PC got infected with spyware and so disclosed someone else's personal data. That is why a uniform approach is vital.
For servers, if you do this kind of thing often, it would pay to accumulate obsolete PCs (a lot of people seem to be disposing of 733MHz Pentiums thse days, they don't cost very much at all), for use as servers, they are perfectly adequate for Linux or FreeBSD. That way you could keep them all properly configured, just need to change the gateway/router config each time, and it is a good use for what might otherwise become landfill. I am collecting some for a different purpose, large offices are regularly disposing of them, no doubt because Sir Bill's latest bloatware is too slow and needs 2 or 3 GHz. What a waste! If it has only a 10MHz network, putting in a 100MHz card will not exactly break the bank, in fact if you put in several cards, filling all the PCI slots, you have an instant router/bridge/firewall (depending which you need). You sometimes also see 100MHz ethernet switches being disposed of, maybe they have moved to GHz or need more ports, or whatever. They are definitely worth acquiring.
I would advise allowing plenty of setup time, something will go wrong. Best to be ready at least a day before people arrive, if possible.
Why not try negotiating with the site owner about having the system installed permanently, that way the costs can be recovered from users over a long time, so an expensive solution might become viable?
If you can get an ADSL line to the site, it gets easier and you only need standard stuff, firewall, router, WiFi access points and so on. It would be best to have a caching proxy server, it can take a lot of load off the line as many people might be looking at the same sites. All of this is straightforward Linux/BSD/Windoze (if you really must) territory, the only thing difficult is the telecom infrastructure. Oh, and watch out for electrical safety regulations, there are lots of potential difficulties, and you reall do need to know your local regulations, and comply with them, or use low-voltage batteries for the lot.
I would ask the local telecom provider how much a decently fast line would cost before messing with satellite systems. The people who supply networking, cable TV, or voice comms in that geographic area are the people who know. In the UK, which does not help you very much, BT can and will provide a voice line or lines (you could aggregate the bandwidth with multiple modems) anywhere, or an ISDN line in most places, ADSL gets slightly harder because of the distance limit. All this is at a cost, of course......
You can walk about in fumes all day, or use the phone all day (assuming no sparks inside the phone, batery contacts, switches, etc) without anything happening. It is the fact that two bits of metal, the pump nozzle and the tank neck of the car, are in very light (i.e. imperfect) contact electrically that creates a place where sparks can be generated.
If you took a small battery, say 6 volts, and connected one terminal to the pump, and the other to the car body, with a resistor to limit the current to a sensible level, say 10 to 100mA, and then poked the nozzle about, in and out of the tank, how long would it be before there was an explosion?
Probably not the first time, but 10, 100, 1000, 1000000 attempts? Would you try such a thing, while filling the tank?
The fact that RF from the phone is energising the thing instead of DC from a small battery, is not of much importance to what would happen, except that the connection to the phone is less obvious, and most people seem to be unable to see how the voltage arises. They seem to think that I have suggested direct ignition by the RF field, well no, that will not happen, and I never suggested it would.
If you are charged with static, and walk up to a car being filled, and poke your finger at the tank, you might well cause an explosion, but if the same person gets out of the car, removes the filler cap, picks up the nozzle, and puts it in the tank, the chances of a static spark are minimal because everything will by then be at the same potential.
At UHF frequencies, the car body is almost ground, by virtue of its capacitance. Yes, at 50Hz, it may well be insulated by the tyres.
The pump nozzle is on the end of a rubber hose, with some built-in conductivity (high resistance) to reduce the effects of static discharge. It is nowhere near ground at RF, even if the hose was metal, it is long and has a significant inductance.
Your assumption sabout free space impedance are also utter junk, the thing that matters is the apparent impedance at the end of the bit of metal, which can go a lot higher. What is the impedance at the end of a dipole antenna, at resonance, for example? What is the voltage at the tip of a resonant CB antenna, with 4 watts input? By your reasoning, it is 38.8 volts, it can actually go over 200 quite easily. The free space impedance is a mathematical fiction to enable wave propagation to be worked out, it does not have physical reality in terms of current-carrying conductors, but it does give the right answers in free space.
And furthermore, as I have said all along, the sparks you get are because of making monentary contact between the nozzle and the car, if you were at all competent, even at DC, you would know that you can make a spark with a single cell battery in these conditions.
When people like you resort to bad language, it is because they either have no coherent argument, have lost the argument, or are just plain incompetent. Any proper electrical engineer would be aware of capacitance and inductance, resonance and transmission line effects.
Furthermore there have been explosions thought to be caused by police radios, in many places they are banned from using them in petrol stations for that reason.
It is interesting that none of the abusive comments have come from RF or even EMC engineers, but only from the ignorant, ill-informed and in your case, just plain rude.
Some types of accident are easy to eliminate, this being one. But where statistics and reality diverge is that some things you can control, others not. You, as an individual, can choose to drive more carefully, and I hope you do, but you can't control the other guy, who has a fairly good probability (in terms of the sort of numbers we are interested in here) of being drunk, or asleep, or having faulty brakes. So, you have reasonable control over one subset (your actions) of the inputs into the accident scenario, but no control over the others. At the filling station, you have full control over one input which can totally avoid the risk.
Now, curiously, people usually have not the slightest clue about managing risks, it is well established that in the UK, rail is the safest form of transport, with air second, by a factor of about 2 to 5. Worldwide, air probably wins as you suggest, in sparsely populated regions the rail infrastructure can't be maintained to the same standards for economic reasons, in a very densely populated country like Japan, rail does even better relatively, because it economically viable to achieve a very high standard). Now, in the UK, which I know best, following a rail crash (typically there are only 3 or 4 fatalities, and then less than one per year, so the picture is easily distorted by one major incident), huge numbers of people take to the roads, when all the statistics show that it is far more dangerous, as a result about 100 extra fatalities might occur (3500 per year on UK roads, 1000 due to drink alone). All because people do not understand the risks. they are willing to accept a much higher risk, because thay imagine that they are in control, than the obvious but much very lower one, where thay are not in control. They don't see the difference between avoidable and unavoidable risks, and they assume that the ones where they have some semblance of control will not happen.
All I am saying really is that if one risk can be totally eleimiated, why not do it? It is easy. Our Health and Safety Executive makes laws and rules which use the term ALARP (as low as reasonably practicable) quite a lot. That means that you are without excuse if you did not get rid of the easy ones, some more difficult risks cannot be changed very much. Allowing mobiles in filling stations does not comply with ALARP, because it is "reasonably practicable" to not allow them. It costs nothing, and needs no new technology.
And yes, looked at objectively, we should ban cars, for many reasons, but I hope we do not. But, there is no argument that I could advance to justify their existence, except for emergency services etc. Maybe I would love to be able to get rid of my car, if public transport could get me to work efficiently. Approximately 1/4000 of a life would be saved each year if my car was to go. But, my employer is not located near public transport, so that is one risk reduction that is not "reasonably practicable".
My opinion is that there will eventually be shown to be a small health risk, minimal to those who use the things sensibly, as is thought to be the same for all RF fields. The Russians, not known for concern about public safety, used to, and presumably still do, have microwave safety limits 30dB (or was it more?) below those of the western world, that in itself is what makes me think there is a small risk, but the issue is now so badly obscured by vested interests on all sides that it will be a long time before the truth is known. Meanwhile, I use mine (not 3G) occasionally, maybe once a month, for a few minutes, because that is all I need, it is unlikely, I hope, to present a serious hazard. But, it is a dual-band Nokia, it does not tell me which band it is using at any time. It would be interesting to know.
What I do know, is that using an earpiece is bad news, the extra coupling of RF from the phone to the head, as a result of the wire, was fairly convincingly demonstrated some time ago. My colleagues, especially the EMC engineers, were not surprised, we were well used to the phenomenon.
You can safely sit in your car in the worst thunderstorm, it might even be fun, but you might get burned paintwork, possibly even dented metal, and damaged tyres, if there is a strike. And beware of damage to the electronics, you do not want a protruding antenna of any sort, if you must do that sort of thing.
Given proper precautions, observing thunderstorms at close quarters is a very safe thing indeed, otherwise very unsafe. There is no middle ground here.
The evidence of ignition of fuel vapour as a result of RF energy causing sparking has been embodied in safety legislation for many years, the research would have happened from the early 1900s onwards, as soon as the explosions started happening. I don't have access to a comprehensive library of past research papers, but some people reading this probably do. There are data published about ignition energies, flash point temperatures, and so on in reference books, available in good libraries. There are even things like the statistical distribution of the ignition energy, it is a probability distribution, on rare occasions an event of well above the ignition threshold might not cause ignition, on other occasions something well below the threshold will. The measured value is a mean, the limits statistically may be quite broad. I don't have access to a university library nowadays otherwise I would look all thi sup for you. What I do know is that in most countries (not, maybe, the US, where they even control oil refineries with Windoze NT, according to a previous /. post some time ago) there are laws, or rules established by the industry themselves, or both, which have evolved as a result of laborious research, and sound theoretical calculation. These laws apply to things like oil refineries, chemical plants and so on.
Now in a refinery, there are strict rules about electrical bonding between sections of pipe etc, to prevent the possibility of sparking at the interface, and exposure levels to RF are set accordingly.
Some years ago, a small local radio station in Scotland, which had occupied their site for a number of years, was forced to move by an oil company which wanted to build a refinery nearby, because of the RF threat to a fully bonded installation. The field strength at the oil company's boundary was similar to that of a mobile phone at 10 to 15 feet, yet it was perceived as a hazard to a fully bonded installation. You do not have electrical bonding between the pump nozzle and the car, which makes ignition much more probable, and is precisely the reason for my opinion. Ask yourself, are you allowed to use a mobile in an oil refinery, and if not, why not? They don't allow it even although teh structure provides good electrical conductivity, because they know and understand the risks, and so does their insurance company, and the legislators. A gap between two bits of metal adjacent to the flow of petrol vapour out of the tank as you fill it is a very bad scenario, compared to any regulated and controlled continuous structure of pipework, the riska sre far higher. Analysing risk involves understanding what can happen, even if the risks are statistically improbable, but not zero. It does not mean dismissing a risk because you "think" it can not happen. The risk here is fairly improbable, but not zero. Lots of equally improbable things can, and do, happen, and cause accidents.
I drive by the laws and advisory guidelines which apply in the UK, it so happens that per head of population our road accident rate is a fraction of that in the US, in fact the lowest of any developed country (it could and should be much better of course, there are still 1000 people killed by drunken scumbags evey year for a start), despite our speed limits being higher, so we must be doing something right. Yet we have had oil refinery explosions, which might suggest that the existing legislation is too lax.......
I am glad we seem to agree, and even more glad that some of us remember the days when they tried hard to do things properly, even though the technology was not quite ready. Now they have the basic technology, but use it to make junk, and more junk, and I mean program material as well as hardware......
It does need an adverse set of circumstances, another poster has referred to 7 sigma. Statistically, it does not happen very often, but one preventable accident is one too many, which sums up my perspective on this matter. 100 or more accidents worldwide is too many, and you can't tell when it might happen, therefore don't do it, simple as that, end of controversy, and end of long, boring, loud, irritating conversation from the person in the next car. Unfortunately there is no technical reason for banning phones in most places where we might like them to be banned.
And as to your other questions, I have always worm my seatbelt when driving since 1971, when I first had a car with belts. No exceptions, ever. I do not measure the typr pressures every day, they do get a visual check for signs of severe loss of pressure. I try hard not to drive when I am tired, and I use my turn signals, as you call them, in accordance with the rules which prevail in this country, always if traffic is around, probably not if I know that the road is entirely clear. Yes, all of these things have caused deaths, which is why I started with the one which can be dealt with most easily, the seatbelts. It involves neither opinion nor speculation, and was known since maybe 1950 or earlier to be a worthwhile safety measure, therefore I adopted it absolutely at the earliest possible opportunity.
Likewise there is no controversy among informed experts, not those in the pay of one vested interest or another, about using mobile phones near petrol pumps, so I do not do it, ever.
The only time I ever splled petrol over a hot manifold was in a very old car, 1968 vintage IIRC, the fuel then was leaded and possibly not so easily ignitable, but I have heard of it happening to other people also. The sight of a small quantity of furiously boiling petrol is quite frightening.....
I am a bit surprised that one of the several possible sources of ignition did not get it, either the distributor or the dynamo brushes, although the engine did stop when the fule pipe came off, maybe it was stationary with no sparking before the vapour got to either place.
Now concerining sparks with apparently low energy circuits, you could try shorting a fresh 3 volt battery with a piece of wire (just lightly brush it against the contact.) You will see sparks. If you hold the wire on the terminals, it may get hot (depending on the gauge of wire), the battery itself may get hot, and if you persist, in some cases, may eventually explode. This is provable fact. Some cellphone batteries if shorted might momentarily supply a current 0f 50 or 100 amps, some maybe only 10 or 20, but still a lot of current, also provable fact. (Don't blame me if you ruin your battery, or worse). Ask an airline about their regulations for shipping batteries, and why..... (they are relatively safe in your laptop or whatever, but no way, in the cargo hold.)
If you repeat the dark room experiment with a small amount of RF energy instead of DC, you will see the same result, sparks.
To be sure of getting all of them you have to either detect them reliably (often no metal parts, which makes it a million times more difficult) or disturb all the soil, damaging crops, trees etc. Footpaths may be mined, they may run through trees, across hillsides, and so on, where it is impracticable to use a bulldozer.
In the case of the bigger mines, you could in theory build a very robust machine, and be willing to repair it quite often. It could run up and down all day automatically, using GPS. It would be much less stressful to its operators, I have heard that operating an armoured bulldozer leads to lots of nervous breakdowns because of the frequent random explosions. I think that might have been on TV some time after Mrs. Thatcher's Falklands war.
If people do have new ideas about this, they should perhaps communicate them to the proper authorities, it really does seem to still be a major problem.
We still occasionally find one of Hitler's bombs in the UK, I know it is a slightly different scenario, but they usually do appear to be still dangerous after 60 years. There was an evacuation somewhere a few weeks ago while the thing was made safe.
I also saw something else on TV a few years ago, again probably in Kuwait, which seemed to detonate an entire minefileld at once. IIRC it was a British invention, likely the US will have it also. Can't remember how it worked, whether it was EMP, or sonic, or what. All I remember is seeing a vast set of almost simultaneous explosions. I think it might not have been totally reliable, IIRC it missed a few.
I think that maybe you asked the question because it does not seem to be done, and I suspect that the answer mainly involves money rather than impossible technology.
Seriously though, I know that animal rights activists will be furious over this, but certain humanitarian issues do have to be tackled, and this may work.
Our channel assignments were carefully worked out very many years ago, so that taking into account the standard IF frequency on TV sets (around 38MHz), there would be as few inompatabilities as possible, taking image frequency response and all that kind of stuff into account. In those days TV sets had front-end selectivity, a lovely little 4-gang variable capacitor tuning high-Q circuits, and valves (tubes) with good large-signal handling capability and not so prone to cross-modulation etc.
The junk of 10 years ago is not actually so good, front-end nonlinearity is a problem aggravated by varicap diodes in the tuners. I rather suspect that if they were to fill the locally unused channels here with what is, as far as the TV is concerned, simply broadband noise, most of the moderm sets would be struggling, the very oldest museum pieces with mechanical tuners might cope (if the IFs are still in correct alignment), and maybe some of the very latest might be OK also, but the vast majority would have problems.
Still, Tony B. Liar has a grand plan, he will end analogue TV transmission earlier than originally expected, so everyone has to upgrade to digital, with its problems. Those with the least money will be the most disadvantaged of course, as is always the way with socialism, or any other -ism for that matter.
Now the current digital systems might be better (arguably) than the NTSC system, which we in Europe have long suspected stands for "Never Twice the Same Colour", but they are vastly inferior to our analogue PAL system, especially on moving scenes. (To be fair to NTSC, it was first, and well-established, before the way of improving it to make PAL was thought about. The change would have broken compatability in the US, but was easy where colour did not already exist). So, it is backwards progress, and it looks as if we will never again have a clean, tidy and well-managed electromagnetic spectrum, as was once the case on both sides of the Atlantic.
Out of curiosity, I will collect these reports from now on.
An eyewitness report certainly is not scientific proof, but if reliable eyewitnesses definitely contradict the science, then the science may well be in trouble. That is how some advances in science happen.
But, in this case, science most certainly says that the events can happen, what is uncertain is the range of possible positions of the phone, concentration of vapour, gap between nozzle and car, etc. It can be said that these events are possible, certainly the energy is thare, and can get to the wrong place, there is insufficient data to say how often a fire will result, or whether it is going to happen in any specific place or time. But the probability of it happening is not zero, whichis why phones should be switched off, no exceptions.
Safety works by analysing what can happen in a worst-case scenario, and then showing that the particular scenario is sufficiently improbable to be ignored. In this case that can not be shown. Maybe if there were no fires at all for 10 years, while everyone used their phone at the pump, it might be established, but still only on a statistical basis.
I am glad you survived your habit, you are right about issues of vapour concentration, but I was thinking of the situation where a piece of red-hot ash dropped from the cigarette. Somewhere near the tank filer there would probably be an ignitable mixture.
It seems to me to be a bit like standing on a mountain peak in a thunderstorm....