They said the same things about the thirteenth amendment, you know. But in places where the freedom to keep slaves is artificially restricted, the mean freedom per capita is still greater than the mean freedom per capita in places where slavery is permitted. If you restrict the sample to the poorest 50% of the population, even more so.
And talking of slavery, who do you think was really responsible for its abolition? Not William Wilberforce, or William Lloyd Garrison. It was James Watt and Michael Faraday who really abolished slavery. Of course they were assisted by a whole host of others, so let's just say for the sake of fairness that IR1 was the event that made slavery unviable.
Now in a few years' time, there will be software available which will take a compiled binary as its input, and generate as its output a piece of source code which, when compiled, will produce exactly the same binary as the original input. It won't necessarily be the original source code {there are several many-to-one mappings in compilation and no guarantee that variable and function names will be preserved} but nonetheless, a sufficiently smart person ought to be able to work with it.
For the first time, it will be possible for users of even closed-source software to exercise Freedoms One and Three {they've been doing Zero and Two for years without the source code}. Once there is no longer be any way of concealing source code, there will cease to be any artificial benefit in attempting to do so. In years to come, IR2 will be seen as the event that made shafting your customer unviable.
Less with the tired old arguments about what comes out of a diesel engine already! The fuel contains carbon, hydrogen and oxygen; the products are water vapour and carbon in various states of oxidation {pure, CO and CO2}. Clue: if you can see it with the naked eye, it's not going to do you any harm.
Bollocks. Organic farming would be as cheap as, or even cheaper than chemical farming if it weren't for the huge subsidies paid out to chemical farmers {who already overproduce anyway}, and the fact that any farmer seeking organic certification must wait several years for the tainted land to purify itself -- during which period their produce can't legally be labelled organic.
Not necessarily. Governments never have to pay patent or copyright royalties; they can just change the law to exempt themselves. Some countries' governments have granted drug manufacturers a special licence to manufacture certain pharmaceutical products for use in their National Health Service. As they already have a statutory right to manufacture the drugs, they do not need to pay for a licence from the patent holder.
Cellulosic ethanol is ethanol derived from cellulose.
The idea is simple. You take any plant matter containing cellulose {a long chain polysaccharide which is fairly immune to yeast}, and hydrolyse the cellulose into mono-, di- and short-chain polysaccharides. Then you have something that will undergo fermentation.
Any dilute acid will hydrolyse cellulose, but then you have the problem to get rid of the acid {which will harm the yeast} without creating a salt which also will harm the yeast. {Might it be possible to use a base whose salt with the chosen acid is insoluble in water, and filter out the precipitate? Since solubility is affected by temperature, it should be possible to refrigerate the mixture in the neutralisation tank to help it precipitate, and dump the waste heat into the hydrolysis tank to speed up the reaction. Further Work Required.} Alternatively, there may exist enzymes which will decompose cellulose into sugars and starches. If these are found to be compatible with yeast it may be possible to work a single-stage conversion, otherwise it will be necessary to do a multi-stage process, neutralising the first enzyme before fermentation..... this does not seem to offer any advantage over the use of a dilute acid.
"Disney's America" - Graham Parker
I knew Virginia once
She was a pretty thing
She walked in the wide fields
And swam in the wide stream
I took her out one day
To the civil war battlefield
Way down in Manassas
where I told her my dreams
But now it's Disney's America
A long way from anywhere
I've got a brand new coat to wear
With a brand name on the seams
Just Disney's America
Virginia she chose to stay
And we drifted apart like runoff
Into the chesapeake bay
Then I had a family
Virginia, I guess she forgot about me
She lives near the concrete sea
Or so people say
I don't remember much
About her gentle touch
My skin just turned so hard
And my feet turned to clay
It was in Disney's America
A long way from anywhere
You get what you pay for there
Man, you get it in spades
Just Disney's America
Virginia she chose to stay
And we drifted apart like runoff
Into the Chesapeake bay
Bridge You can't get too excited
You can't get too enthused
From Dismal Land to the Tragic Mountain
We are not amused
Yeah I knew Virginia once
She was a pretty thing
She walked in the wild fields
And swam in the wild streams
But now it's Disney's America
That's where we collect our pay
Where we drifted apart like runoff
Into the chesapeake bay
Yeah we drifted apart like runoff
Into the chesapeake bay
Sounds to me like something you could do very easily with Kate. You just have to enter
ftp://mylogin@myisp.co.uk/ in the file requester, give your password when asked, and you should be in! If you have a shell account, try fish:// instead -- this uses an SSH connection and so everything is encrypted. If you accidentally muck up the password entry, try entering
ftp://mylogin:password@myisp.co.uk/ to fix it, but obviously don't do this if anyone is watching. The password will disappear when the URL is redisplayed.
This is not just confined to Kate -- it works with all the other KDE applications, too. You can even open an ftp or fish directory in Konqueror, and just drag-and-drop files into it.
The UK alone could feed the whole of the rest of the world {assuming a vegan diet and unrealistic logistics capabilities}. The EU pays for farmers to destroy much of what they grow.
Worrying about having to turn over land to energy crops is a bit like worrying about polluting outer space with radioactive junk.
Does this imply that the fuse/breaker box for a two-floor house would only have two 30A/240V breakers to serve the general-purpose power sockets on each floor?
Yes. In fact, a single ring main can serve anything up to 100m2 of floor area. It's not uncommon in older houses to see every power point wired through a single 30A wire fuse. We still have plenty of wire fuses, simply because they rarely fail -- every plug has a fuse in it, and the chances are that the lower-rated of two fuses in series will fail first.
Additional protection is provided by an ELCB {Earth Leakage Circuit Breaker}, also known as an RCD {Residual Current Detector}. This monitors the difference between the current flowing out of the live and the current returning into the neutral, and if the difference is more than a set threshhold {30mA is common} the device trips off -- the idea is that if more current is going out of the live than coming back via the neutral, some of it must be taking an unsanctioned path back to earth. ELCB protection is required for showers {bear in mind that 70 watts will make water flowing at one litre per minute just one degree hotter, so a 7kW shower gives you a choice between a lukewarm trickle or a cold gush} and outdoor circuits; but not on lighting circuits, since we use two-contact, push-and-twist fitting lampholders so the metal part of the bulb is isolated and anyway it requires a deliberate act to mess with a lighting circuit. Having the lights unsafe to touch is considered less hazardous than having them fail suddenly under fault conditions.
UK power distribution is three-phase; domestic wiring is 230V single-phase. At the substation {which usually serves a whole estate}, the three phases are star-wired with the neutral point earthed. Adjacent houses are on adjacent phases; with the co-operation of both neighbours you could run a three-phase appliance. Power enters on a two-core, steel armoured cable, with the neutral and earth joined together just upstream of the electricity meter and then a 60A or 100A fuse. New consumer units {fuseboxes} usually have space for 12 circuit breakers {6 with ELCB protection and 6 without}, but you will still find older ones with just six wire fuses.
As to the UK ring main system -- what happens if one of the two hot paths should open? You wouldn't necessarily know it, because all the outlets would still be getting power from the other path. Since the wire gauge is thinner than would be needed in a star-pattern, wouldn't an overheat and fire be more likely? Or is there some device in the breaker to detect this out-of-balance condition?
Not much would happen, really. Even 2.5mm2 cable doesn't cause enough voltage drop at 30 amps to be a fire hazard {and you'd have to be drawing the full 30A from sockets on the same one of the two now-separate paths}. The condition would go unnoticed for years; there isn't any obvious way to detect it. It's difficult to see how it would be caused in the first place, though, unless someone was doing something deliberate that they shouldn't. The three terminals on the back of the socket faceplate {which carries one or two switched 13A sockets} are each designed to accept 7.5mm2 of wire; this allows for the feed from the fusebox, the return to the fusebox, and an additional "spur" connection for another one faceplate, close to the first {so minimising power loss in the single cable} in case the builder did not put in enough power points.
If the chip is in a socket, the best way to get it out is always to cut up the socket in situ, remove the pins from the motherboard one-at-a-time and fit a new socket.
The easiest way to get a multi-pin, non-socketed, through-leaded IC out is to apply heat to all the connections at the same time. There are soldering iron tips shaped for this purpose, also hot-air blowing equipment. Obviously you will need a powerful, thermostatic soldering iron; cheap ones which just rely on atmospheric cooling can't put in enough energy. Failing this, use a proper vacuum desoldering system to work on one pin at a time -- and be extremely careful not to take out the through-hole plating. You don't want to melt the IC, so start with diametrically-opposed corners and work your way up and down opposite sides in turn.
If the chip is not in a socket, and things are really desperate, simply cut through each lead, right on the bend. Make a temporary adaptor for testing, reading, reprogramming or whatever it is you want to do {at the level of desperation you're at, this shouldn't pose any problems}. Then solder the stumps of the IC back to the stumps on the motherboard. Start at one corner, then do the diametrically opposed corner, then the remaining two corners; then the rest, one from each side in turn.
With any kind of surface-mount work, you need to be certain that the solder is well and truly molten; otherwise you risk pulling the tracks off the board. Blast the leads with hot air -- don't even think about doing it any other way -- and try to lift the chip, squarely and very, very, very gently.
BTW, the procedure of removing a BIOS chip from a "live" motherboard is nothing new. Anything more sophisticated than a '386 copies the BIOS ROM contents {which are stored in an 8-bit-wide chip} to RAM {which is 32-bits or wider} first, and runs the BIOS routines {which are still used by bootstrap loaders and superannuated operating systems} from RAM..... this obviously requires some fancy logic on the address decoding. The memory chip {it's usually a flash PROM} can be swapped out, as long as you don't do anything stupid like shorting the power lines or delivering a charge of static into the address and data lines. I seem to remember that some similar technique was used for making the earliest XBox mod chips.
Circuit breakers are there to protect the building's permanent wiring. Plug fuses are there to protect the appliance's flexible cable. Another good thing about our mains plugs is the way the cable exits at 90 degrees to the pins, meaning you have to hold the plug when withdrawing it -- pulling on the cable just tightens it! On the downside, it does mean that they can lie on their backs, pins up in the air, just waiting to spike an unwary bare foot. The purpose of the switch on the socket is to ensure that no current is flowing through the pins of the plug at the moment when they are withdrawn from the socket, otherwise there is a chance of arcing occurring. The live and neutral pins are half-sleeved, and the socket is designed so that the plug will have disengaged completely from the socket before any metal is exposed -- even if you try to use some kind of blade to prise the plug out of the socket, you will only ever catch the protective insulation until the pins are out of contact.
Every socket outlet in the UK can supply up to 13 amperes at 230V, giving a nominal 3kW. All the socket outlets on each floor of a house are connected together in what is called a "ring main" -- parallel connection with two connections to the consumer unit {fuse box}, one from each end of the daisy chain -- and served by a 30 amp fuse or trip switch. There are two parallel pathways to each socket outlet, so the permanent cable can be a lighter grade {2.5mm2 as opposed to 4.0mm2} since it is effectively doubled-up.
The fact of there being a 30A fuse at the consumer unit means that if any appliance develops a short-circuit internally, its flexible supply cable might potentially have to withstand enough current to blow a 30A fuse {and such a beast won't even flinch at anything shy of 60 amperes}. That's a lot of energy and there's a good chance of the 30A fuse not being the weakest link in that circuit. So rather than fit an inordinately thick power lead to each appliance, an extra fuse is fitted in the plugtop. Now, if the appliance develops a fault, its flex only has to withstand enough energy to blow the lower-rated plug fuse {and when the filament bulb in my flatmate's table lamp shines its last and takes out the plug fuse, it won't affect the DVD I'm burning}.
Also, every UK socket is earthed, and every plug has three pins. The earth pin is longer than the live {phase} and neutral, so as to mate first, and is required to be present because it operates a safety shutter covering the live and neutral apertures on the socket when no plug is present. The only power outlets permitted in bathrooms are shaver sockets, and these must be supplied via an isolating transformer {so either one terminal is safe to touch} rated no more than 20VA, fitted with protective devices and mounted as close as possible to the outlet {means less chance of parasitic coupling undoing the effects of the transformer}.
I'll respond to your points in order. They do seem to be mainly nitpicking; there are already established procedures in place to deal with any scenario you could name with the time-honoured pencil-and-paper system used in the UK which would adapt readily to a semi-automated scheme.
You would probably have to conduct elections for separate posts on separate days. In the UK there are generally only one or two posts being contested at a time; rarely three {Local, National and EU}.
This is almost how the system already works in the UK -- voters are issued with a card by post. But you're right, there is a backup system in place. The name and address on the card {or, if you don't have the card and the Presiding Officer does not actually know you personally, any piece of ID with them on it} are crossed off a list by the Presiding Officer's assistant, and the Presiding Officer -- who only sees a name being crossed off, not whose name -- marks an un-numbered ballot paper with a special punch and gives it to you. The pattern of the punch is unique to that polling station on that date; but in any case, you can't falsify a ballot paper because the Presiding Officer would not allow you into the voting booth without first issuing a ballot paper. {With my electromechanical system, charging the VBU capacitor represents issuing a ballot paper.} Even if you do manage to steal someone else's vote by giving a false name and address, you will automatically lose your own vote when the Presiding Officer recognises your face. And if, at the end of the election, there is a mismatch between votes cast and voters accounted for, the Presiding Officer will be aware and will follow the established procedure.
In some organisations where membership cards are used, the membership cards have a row of numbers which are punched out to signify voting in elections at meetings.
How many candidates do you actually have in your elections? 20 positions ought to be more than plenty. Note also that there needs to a core in the cable for each selection, plus two for the capacitor terminals and one more for the moving contact {which gets disconnected from the capacitor by the charge/vote switch} so this will impose limitations of its own.
In such a case, as is done with pencil-and-paper ballots, the voter would have to be allowed to take an able-bodied carer whom they trust into the voting booth to help them.
Hello? Neither does a pencil-and-paper ballot, or any kind of voting machine. Once the paper goes through the slot, or you have pressed the voting button, there's no changing your mind. What my system does have is a two-stage, "choosing and confirming" process. You get plenty of time to make your mind up. If you were taking so long that the meter fell out of the green zone, you would have to call to the Presiding Officer {who can see whether or not the master counter has advanced} for a recharge.
As currently happens in pencil-and-paper ballots, the Presiding Officer would notice any attempt to tamper with the election process and inform the police that a crime was being committed against the Representation of the People Acts.
The system is not designed to replace common sense -- ultimately, nothing can. A Presiding Officer is still required. However, the counting of the votes is done instantaneously {it has actually been going on the whole time}. But most crucially, the system is absolutely open to public scrutiny at every stage -- and uses science that should be capable of being understood by anyone with a high school education.
Software developer my backside. You're probably just some snot-nosed whining kid who fails it. Do you honestly imagine your "competitors" are any less smart than you are? Or that you could gain nothing of worth from a look at their code, which they would be obliged to show you? If you really are a software developer, you're going nowhere with that attitude. It's not the 1980s anymore, pal.
The Government is there exactly to tell people how to do business. We -- well, those of us who actually have jobs, that is -- pay taxes to the government precisely to stop unscrupulous people exploiting unfair advantages. Software developers keeping the source code from users are misusing an unfair advantage, by preventing users from exercising their natural rights to study the software they use and adapt it to their needs. Nobody ever tried to stop you from taking a clock to pieces to see how it worked, or shortening the legs on a table to make it more comfortable -- why should it be any different with computer software?
Well, it may not make sense to most users, but that is not the point. Nobody who installs a fire extinguisher ever expects to have to use that either. They're just glad of it when they need it.
terrific ..... not
on
IE7 Leaked
·
· Score: -1, Troll
The world does not need another closed-source browser. In fact it does not need any more closed-source software at all. What the world needs is for some government somewhere to decide that every computer owner has the right to view the source code of any program running on a computer that they own, and if the software vendor doesn't like that then bollocks -- users outnumber developers. Think about it..... Food has to be labelled with its ingredients list and mean nutritional analysis. Clothing has to be labelled with its fibre composition. Cosmetics have to be labelled with their ingredients. Beer has to be labelled with its alcohol content, fags have to be labelled with tar and nicotine content..... it makes absolutely no sense that software should be supplied without the corresponding source code.
That is what I said. The losses on both the primary and secondary sides -- due partly to resistance heating and mainly to badly-designed cores -- can all be lumped back to the primary side. Diodes that don't rectify properly {even the difference between 50Hz and 60Hz is noticeable with some of the two-terminal devices laughingly described as "diodes" by some Far Eastern manufacturers} and capacitors that leak both contribute to inefficiency. OTOH, running an appliance from even the least-efficient plug-in PSU you can find is still a hell of a lot greener than running it on disposable batteries.
Also as I said before, part of the problem is that most appliances are not plugged directly into the wall, where each socket always has its own separate on-off switch, but run from an extension lead; and extension leads do not usually have a separate switch for each socket. I don't know why this should be the case, apart from sheer cheapness.
As for the 7805, that is just a series regulator: a zener diode acting as a voltage reference, and a transistor in emitter-follower mode amplifying the current. The excess energy gets turned into heat. If you feed it with 13.5V and try to draw 150mA, it will drop 8.5V at 150mA, giving over a watt of heat dissipation. In general, the efficiency of such devices = (output voltage / supply voltage) * 100%. There are better devices available. These use a slightly different form of SMPS, in which a capacitor is partly-charged through an inductor to limit the current; this time, the transistor is switched hard off and on, so the only resistive loss is due to the parasitic resistance of the inductor. Unfortunately, they aren't simple drop-in replacements, because the inductor isn't something that can be built on-chip.
..... when vacuum-tube valves were all there were, "Standby" was a power mode on a well-built stage amplifier. This provided power only to the heater filaments; the HT was disconnected. The idea was simply to allow the amps to warm up without any danger to the speakers {or an unlucky roadie replacing a defective valve}.
Modern appliances with a remote control need to draw just enough power to run the sensor and decoding circuitry -- otherwise they would invariably have to be switched on at the wall socket. And since nobody has enough wall sockets, everyone uses multi-way extension leads; but extension leads rarely have individual switches like wall sockets.
Also, even modern appliances without remote controls tend to have the on-off switch in the secondary circuit of the power transformer. This seems to be something to do with a general reluctance of manufacturers to touch mains. In an ideal world, this would not matter; but in this real world, copper is not a perfect conductor and steel loses its magnetism. So a transformer is not 100% efficient {although it's about the closest anyone's ever built}. By the Principle of Equivalence, we can represent all the losses as a simple resistor in parallel with the primary winding.
Switched-mode power supplies are becoming more common {laptops, and even most telephones except Nokia, use them for their rechargers}. The basic principle of a switched mode supply is that you rectify the incoming mains to DC; then use a high-power oscillator to turn it back into AC but at a much higher frequency, typically in the 25-50kHz range {you could go even higher but LW radio starts around 150kHz and you have to be careful to avoid interference} which can then be passed through a physically smaller transformer, as the current spends less time flowing in any one direction so the core can be smaller {every iron atom is like a miniature compass needle which tries to align itself with an externally-applied magnetic field, which in the case of a transformer is imposed by the current applied to the primary and opposed by the current drawn from the secondary; when all the "needles" are aligned in the same direction, the core is said to be saturated and the winding is now behaving as a resistor which is not what you want. At 50Hz the current flows for 10ms in one direction, then 10ms in the other; at 25kHz it flows for only 20us in each direction}. In an SMPS it is mainly the leakage current of the input smoothing capacitor which determines the wastefulness.
If Standby mode is not going away {and it isn't} then the best we can hope for is to ensure through regulation that new appliances are wasting as little power as possible in Standby. And while we're making legislation for new appliances, this would be an ideal time to consider a few other factors; such as minimum standards for product lifetime {I still feel inclined to say "forever" would be an acceptable minimum}, field maintainability, manufacturer responsibility for end-of-life recycling, and whether working standards in the country of manufacture would be acceptable in the destination country.
It sounds as though they are basically saying that a fake dog is a poor substitute for a real dog. Not much surprise there.
There is something about the relationship between human and dog that cannot quite be replicated artificially. It is a real two-way relationship: we are accepted into our dogs' homes just as much as they are accepted into ours. It's simply not possible to feel the same way about a piece of machinery, however pretty you try to make it. Artificial intelligence to date still looks very artificial; it's not so much that smart programs are passing the Turing test, but dumb humans are failing it.
Just the other night, I was lying in bed with my pit..... er..... Staffordshire Bull Terrier Crossbreed lying alongside me. Her powerful, muscular jaws, packed with razor sharp teeth, were just millimetres from my face. Yet at no time did I feel any cause to be afraid. Not because she was programmed not to harm me, like some robot, but because she had chosen not to harm me. That was a great feeling, but it also made me aware of my own responsibilities to her. As she drifted off to sleep, her legs began to twitch and she gave out a few little high-pitched barks; no doubt she was dreaming her wolf-dreams, running with the pack. And with the light of dawn, she would lick me awake.
In almost any review of software, failure to mention the licence terms {GPL, BSD, Other Free, Non-Free} is tantamount to a criminal omission. The only exception would be in a publication concerning itself only with one particular licence, or in a jurisdiction where the Four Freedoms are enshrined in law.
Funny as the mods mod'ed it, this was actually meant as a serious suggestion. But they are unlikely to come back and explain their reasons now, since to do so would undo their moderation..... guess I'll have to write it off as an example of/. weirdness.
The Indian machines are actually a lot more sophisticated than my description -- they're fully electronic, no moving parts. But that in its own turn makes them harder for a lay-person to scrutinise. OTOH, the Indian implementation of the "unique token which is exchanged for an anonymous token" is wonderfully simple and next to abuse-proof. Token One is the voter's own skin, and is "spoiled" by applying a dye which will take longer to remove than the polls stay open for -- unless you use some technique such as bleach and a scrubbing brush that will leave a mark of its own.
I came up with the "solenoids in series" idea as a way to get around the need for mechanical interlocks between the counters -- my first idea was for each candidate counter solenoid to have a peg that caught in a slotted strip running along the row and linked to the master counter, so any one would move on its own and the master counter but no others {the slot would slide straight past the peg of any non-energised counter}. But that would require specially-made parts; you can buy non-interlocked counters off the shelf and the coil impedance isn't exactly a secret from anyone with an AVOmeter. Doing the interlocking electrically should work just as well, as long as the coil resistances are within a certain tolerance of one another and the counter mechanisms are not too worn {another good reason to use non-resetting counters and record the before and after readings, rather than using resetting counters: there is an instantaneous reading of how worn the counters are}.
This system is still second-best to pencil and paper and counting by hand though, IMHO. As another reader points out, you would need to have separate machines for each question being asked; it doesn't do away with most of the traditional requirements of the Presiding Officer; and it's only good for a "first past the post" system where votes can be counted as you go along and still be meaningful. Most of the world actually uses Single Transferrable Votes, and I'm not sure that there is any way to count these on-the-fly. Each vote probably would have to be stored on a machine-readable {but also, for transparency's sake, human-readable} ballot paper or something like it, and these run through the machine several times {so the votes were counted from the actual human-readable ballot papers and not some auxiliary memory susceptible to tampering; and also to minimise the total parts count, which makes the scrutineers' job easier}. This raises the question, how do we ensure that the votes are not kept in chronological order {which might give some traceability} whilst only allowing them to be counted once each? Separate cards for each vote, maybe; or a continuously-circulating loop of paper tape which would be punched in essentially random positions.
This is my idea for a voting machine. It depends for its operation on the idea that when a current is passed through two solenoids in series, both armatures will pull in. The machine itself has two units: the voting booth unit and the presiding officer's unit, linked by a cable. When not being used for an election, the machines would be made available for public scrutiny.
The voting booth unit {VBU} has a large rotary switch, a pushbutton and a meter with a green zone. The Presiding Officer's unit {POU} contains a power supply, and a column of non-resettable electromechanical counters, all but one of which are covered by a metal plate. This plate is fastened in place with a wire with an aluminium seal bearing the Returning Officer's mark. The counter readings before the start of the election are recorded on a paper label affixed to the underside of the cover plate. There is also a switch labelled "CHARGE" and "VOTE".
Each voter is issued with a unique, identifiable token -- a postcard with their name and address on it. The voter shows the token {Token One} to the Presiding Officer, who first spoils Token One and then moves the switch on the POU to "CHARGE" as the voter steps into the booth. The Presiding Officer then moves the switch to "VOTE". The voter has now traded Token One for a second token, all of which are absolutely anonymous, identical and indistinguible from one another: Token Two is an electrical charge stored in a capacitor contained within the VBU.
The voter spins the rotary switch to their preferred candidate, checks that the meter is in the green zone and depresses the voting button. The VBU capacitor is discharged through the coil of one of the concealed counters in the POU. One terminal of each of these counters is commonned together; the current through any one of the candidate counters also flows through the master counter, and returns to the other plate of the capacitor. The charge in the capacitor is soon exhausted, and cannot be replenished unless the Presiding Officer moves the POU switch to CHARGE. The voter then has the option to move the rotary switch to a different position so as to conceal their preference -- or to leave it there to advertise their preference.
Every voter has a receipt to show that they have voted {the spoiled Token One} but once a vote has been cast, the only record of that vote is the fact that the master counter and one of the candidate counters have advanced by one place. There is thus no way to link a voter with their vote. The master counter is in view of {and the counting mechanism is within earshot of} the PO, who can thus confirm visually and aurally that a vote has been cast {or separately, manually record a "no vote" if the voter leaves the booth without voting for any candidate}. All the candidate counters are concealed until the close of polling, when a few minutes' worth of mental arithmetic will reveal the true count. By virtue of its simplicity, and the fact that it has been subjected to public scrutiny, we can take for granted that the mechanism is behaving as it is supposed to; the Returning Officer need only inspect the tamper-evident seals to determine whether the result is valid or compromised.
{In case the above constitutes a patent claim, I hereby licence it for use royalty-free in all applicable jurisdictions, in the hope that it will be of service to Humankind}.
I don't think it would make a blind bit of difference if they announced, tomorrow, that there was absolutely no causal link between phones and ill health, and described an experiment that could be carried out using commonly-available materials to demonstrate this, people would still believe that there was a link. Why? Because people seem to prefer the idea that things are bad for them.
We're better fed now than we've ever been -- but there are still people with eating disorders, and there isn't a single foodstuff that is agreed by all experts as safe to eat. Crime is at the lowest level it's ever been at for years -- but the Authorities are stoking up the fear of crime as an excuse to invade our civil liberties. I'm reminded of a lyric from a song by Del Amitri:
And I won't pretend that I'm the saviour of the innocent and bad,
But put two withered old blooms in a couple of rooms
And they'll behave like lunatics and crave what makes them sad
It seems that Doom and Gloom just make for better news than boring nice stuff. Nobody is interested in fluffy kittens unless they're being brutally hacked to pieces, or trees and flowers unless they're deadly poisonous. And never mind about all the lives that have been saved just because someone had their mobile with them.....
What else is there to say? It's a bit of legacy code left over from the days when it was safe to assume that any code on a computer had been put their with the owner's knowledge and consent. That assumption has since been invalidated by subsequent events. A backdoor it may be -- but when it was put there, there used to be a fence around the back garden.
And this is just one example of a whole class of things that are really, seriously, terribly wrong with Windows {and for that matter, closed-source software in general}. A lot of benign application software has come to depend on behaviour in the operating system that ought never to have been allowed in the first place -- behaviour that makes the propagation of viruses and worms so much simpler. Now, if Microsoft change the way Windows works so as not to just hand out permission for any process to interfere with any other process, then the worms and viruses that depend on this behaviour will die off -- but so will all those applications that depend on this broken behaviour. Then what used to be a choice between "Stay with Microsoft, and all your old software will still work like it did before" and "Leave Microsoft, and none of your old software will work anymore" becomes one between "Continue to splash out good money after bad to Microsoft, but none of your old software will work anymore" and "Wave goodbye to Microsoft, none of your old software will work anymore but there are better-than-adequate replacements for all of it".
And my prediction? A company that still makes extensive use of an obsolete software product will find themselves -- and their precious data -- orphaned as a consequence of the switch to Vista. They will have to obtain a pirated copy or copies of an earlier Microsoft OS {because there is no way to obtain a legitimate one} just in order to read their own files. This will only be discovered in a Licencing Gestapo bust.
Slashdot Mirror
They said the same things about the thirteenth amendment, you know. But in places where the freedom to keep slaves is artificially restricted, the mean freedom per capita is still greater than the mean freedom per capita in places where slavery is permitted. If you restrict the sample to the poorest 50% of the population, even more so.
And talking of slavery, who do you think was really responsible for its abolition? Not William Wilberforce, or William Lloyd Garrison. It was James Watt and Michael Faraday who really abolished slavery. Of course they were assisted by a whole host of others, so let's just say for the sake of fairness that IR1 was the event that made slavery unviable.
Now in a few years' time, there will be software available which will take a compiled binary as its input, and generate as its output a piece of source code which, when compiled, will produce exactly the same binary as the original input. It won't necessarily be the original source code {there are several many-to-one mappings in compilation and no guarantee that variable and function names will be preserved} but nonetheless, a sufficiently smart person ought to be able to work with it.
For the first time, it will be possible for users of even closed-source software to exercise Freedoms One and Three {they've been doing Zero and Two for years without the source code}. Once there is no longer be any way of concealing source code, there will cease to be any artificial benefit in attempting to do so. In years to come, IR2 will be seen as the event that made shafting your customer unviable.
Less with the tired old arguments about what comes out of a diesel engine already! The fuel contains carbon, hydrogen and oxygen; the products are water vapour and carbon in various states of oxidation {pure, CO and CO2}. Clue: if you can see it with the naked eye, it's not going to do you any harm.
Bollocks. Organic farming would be as cheap as, or even cheaper than chemical farming if it weren't for the huge subsidies paid out to chemical farmers {who already overproduce anyway}, and the fact that any farmer seeking organic certification must wait several years for the tainted land to purify itself -- during which period their produce can't legally be labelled organic.
Not necessarily. Governments never have to pay patent or copyright royalties; they can just change the law to exempt themselves. Some countries' governments have granted drug manufacturers a special licence to manufacture certain pharmaceutical products for use in their National Health Service. As they already have a statutory right to manufacture the drugs, they do not need to pay for a licence from the patent holder.
Cellulosic ethanol is ethanol derived from cellulose.
..... this does not seem to offer any advantage over the use of a dilute acid.
The idea is simple. You take any plant matter containing cellulose {a long chain polysaccharide which is fairly immune to yeast}, and hydrolyse the cellulose into mono-, di- and short-chain polysaccharides. Then you have something that will undergo fermentation.
Any dilute acid will hydrolyse cellulose, but then you have the problem to get rid of the acid {which will harm the yeast} without creating a salt which also will harm the yeast. {Might it be possible to use a base whose salt with the chosen acid is insoluble in water, and filter out the precipitate? Since solubility is affected by temperature, it should be possible to refrigerate the mixture in the neutralisation tank to help it precipitate, and dump the waste heat into the hydrolysis tank to speed up the reaction. Further Work Required.} Alternatively, there may exist enzymes which will decompose cellulose into sugars and starches. If these are found to be compatible with yeast it may be possible to work a single-stage conversion, otherwise it will be necessary to do a multi-stage process, neutralising the first enzyme before fermentation
I knew Virginia once
She was a pretty thing
She walked in the wide fields
And swam in the wide stream
I took her out one day
To the civil war battlefield
Way down in Manassas
where I told her my dreams Then I had a family
Virginia, I guess she forgot about me
She lives near the concrete sea
Or so people say
I don't remember much
About her gentle touch
My skin just turned so hard
And my feet turned to clay Yeah I knew Virginia once
She was a pretty thing
She walked in the wild fields
And swam in the wild streams
Sounds to me like something you could do very easily with Kate. You just have to enter
ftp://mylogin@myisp.co.uk/
in the file requester, give your password when asked, and you should be in! If you have a shell account, try fish:// instead -- this uses an SSH connection and so everything is encrypted. If you accidentally muck up the password entry, try entering
ftp://mylogin:password@myisp.co.uk/
to fix it, but obviously don't do this if anyone is watching. The password will disappear when the URL is redisplayed.
This is not just confined to Kate -- it works with all the other KDE applications, too. You can even open an ftp or fish directory in Konqueror, and just drag-and-drop files into it.
The UK alone could feed the whole of the rest of the world {assuming a vegan diet and unrealistic logistics capabilities}. The EU pays for farmers to destroy much of what they grow.
Worrying about having to turn over land to energy crops is a bit like worrying about polluting outer space with radioactive junk.
Additional protection is provided by an ELCB {Earth Leakage Circuit Breaker}, also known as an RCD {Residual Current Detector}. This monitors the difference between the current flowing out of the live and the current returning into the neutral, and if the difference is more than a set threshhold {30mA is common} the device trips off -- the idea is that if more current is going out of the live than coming back via the neutral, some of it must be taking an unsanctioned path back to earth. ELCB protection is required for showers {bear in mind that 70 watts will make water flowing at one litre per minute just one degree hotter, so a 7kW shower gives you a choice between a lukewarm trickle or a cold gush} and outdoor circuits; but not on lighting circuits, since we use two-contact, push-and-twist fitting lampholders so the metal part of the bulb is isolated and anyway it requires a deliberate act to mess with a lighting circuit. Having the lights unsafe to touch is considered less hazardous than having them fail suddenly under fault conditions.
UK power distribution is three-phase; domestic wiring is 230V single-phase. At the substation {which usually serves a whole estate}, the three phases are star-wired with the neutral point earthed. Adjacent houses are on adjacent phases; with the co-operation of both neighbours you could run a three-phase appliance. Power enters on a two-core, steel armoured cable, with the neutral and earth joined together just upstream of the electricity meter and then a 60A or 100A fuse. New consumer units {fuseboxes} usually have space for 12 circuit breakers {6 with ELCB protection and 6 without}, but you will still find older ones with just six wire fuses. Not much would happen, really. Even 2.5mm2 cable doesn't cause enough voltage drop at 30 amps to be a fire hazard {and you'd have to be drawing the full 30A from sockets on the same one of the two now-separate paths}. The condition would go unnoticed for years; there isn't any obvious way to detect it. It's difficult to see how it would be caused in the first place, though, unless someone was doing something deliberate that they shouldn't. The three terminals on the back of the socket faceplate {which carries one or two switched 13A sockets} are each designed to accept 7.5mm2 of wire; this allows for the feed from the fusebox, the return to the fusebox, and an additional "spur" connection for another one faceplate, close to the first {so minimising power loss in the single cable} in case the builder did not put in enough power points.
Google for "IEE regulations 16th edition" if you're interested to know more.
If the chip is in a socket, the best way to get it out is always to cut up the socket in situ, remove the pins from the motherboard one-at-a-time and fit a new socket.
..... this obviously requires some fancy logic on the address decoding. The memory chip {it's usually a flash PROM} can be swapped out, as long as you don't do anything stupid like shorting the power lines or delivering a charge of static into the address and data lines. I seem to remember that some similar technique was used for making the earliest XBox mod chips.
The easiest way to get a multi-pin, non-socketed, through-leaded IC out is to apply heat to all the connections at the same time. There are soldering iron tips shaped for this purpose, also hot-air blowing equipment. Obviously you will need a powerful, thermostatic soldering iron; cheap ones which just rely on atmospheric cooling can't put in enough energy. Failing this, use a proper vacuum desoldering system to work on one pin at a time -- and be extremely careful not to take out the through-hole plating. You don't want to melt the IC, so start with diametrically-opposed corners and work your way up and down opposite sides in turn.
If the chip is not in a socket, and things are really desperate, simply cut through each lead, right on the bend. Make a temporary adaptor for testing, reading, reprogramming or whatever it is you want to do {at the level of desperation you're at, this shouldn't pose any problems}. Then solder the stumps of the IC back to the stumps on the motherboard. Start at one corner, then do the diametrically opposed corner, then the remaining two corners; then the rest, one from each side in turn.
With any kind of surface-mount work, you need to be certain that the solder is well and truly molten; otherwise you risk pulling the tracks off the board. Blast the leads with hot air -- don't even think about doing it any other way -- and try to lift the chip, squarely and very, very, very gently.
BTW, the procedure of removing a BIOS chip from a "live" motherboard is nothing new. Anything more sophisticated than a '386 copies the BIOS ROM contents {which are stored in an 8-bit-wide chip} to RAM {which is 32-bits or wider} first, and runs the BIOS routines {which are still used by bootstrap loaders and superannuated operating systems} from RAM
Circuit breakers are there to protect the building's permanent wiring. Plug fuses are there to protect the appliance's flexible cable. Another good thing about our mains plugs is the way the cable exits at 90 degrees to the pins, meaning you have to hold the plug when withdrawing it -- pulling on the cable just tightens it! On the downside, it does mean that they can lie on their backs, pins up in the air, just waiting to spike an unwary bare foot. The purpose of the switch on the socket is to ensure that no current is flowing through the pins of the plug at the moment when they are withdrawn from the socket, otherwise there is a chance of arcing occurring. The live and neutral pins are half-sleeved, and the socket is designed so that the plug will have disengaged completely from the socket before any metal is exposed -- even if you try to use some kind of blade to prise the plug out of the socket, you will only ever catch the protective insulation until the pins are out of contact.
Every socket outlet in the UK can supply up to 13 amperes at 230V, giving a nominal 3kW. All the socket outlets on each floor of a house are connected together in what is called a "ring main" -- parallel connection with two connections to the consumer unit {fuse box}, one from each end of the daisy chain -- and served by a 30 amp fuse or trip switch. There are two parallel pathways to each socket outlet, so the permanent cable can be a lighter grade {2.5mm2 as opposed to 4.0mm2} since it is effectively doubled-up.
The fact of there being a 30A fuse at the consumer unit means that if any appliance develops a short-circuit internally, its flexible supply cable might potentially have to withstand enough current to blow a 30A fuse {and such a beast won't even flinch at anything shy of 60 amperes}. That's a lot of energy and there's a good chance of the 30A fuse not being the weakest link in that circuit. So rather than fit an inordinately thick power lead to each appliance, an extra fuse is fitted in the plugtop. Now, if the appliance develops a fault, its flex only has to withstand enough energy to blow the lower-rated plug fuse {and when the filament bulb in my flatmate's table lamp shines its last and takes out the plug fuse, it won't affect the DVD I'm burning}.
Also, every UK socket is earthed, and every plug has three pins. The earth pin is longer than the live {phase} and neutral, so as to mate first, and is required to be present because it operates a safety shutter covering the live and neutral apertures on the socket when no plug is present. The only power outlets permitted in bathrooms are shaver sockets, and these must be supplied via an isolating transformer {so either one terminal is safe to touch} rated no more than 20VA, fitted with protective devices and mounted as close as possible to the outlet {means less chance of parasitic coupling undoing the effects of the transformer}.
- You would probably have to conduct elections for separate posts on separate days. In the UK there are generally only one or two posts being contested at a time; rarely three {Local, National and EU}.
- This is almost how the system already works in the UK -- voters are issued with a card by post. But you're right, there is a backup system in place. The name and address on the card {or, if you don't have the card and the Presiding Officer does not actually know you personally, any piece of ID with them on it} are crossed off a list by the Presiding Officer's assistant, and the Presiding Officer -- who only sees a name being crossed off, not whose name -- marks an un-numbered ballot paper with a special punch and gives it to you. The pattern of the punch is unique to that polling station on that date; but in any case, you can't falsify a ballot paper because the Presiding Officer would not allow you into the voting booth without first issuing a ballot paper. {With my electromechanical system, charging the VBU capacitor represents issuing a ballot paper.} Even if you do manage to steal someone else's vote by giving a false name and address, you will automatically lose your own vote when the Presiding Officer recognises your face. And if, at the end of the election, there is a mismatch between votes cast and voters accounted for, the Presiding Officer will be aware and will follow the established procedure.
- How many candidates do you actually have in your elections? 20 positions ought to be more than plenty. Note also that there needs to a core in the cable for each selection, plus two for the capacitor terminals and one more for the moving contact {which gets disconnected from the capacitor by the charge/vote switch} so this will impose limitations of its own.
- In such a case, as is done with pencil-and-paper ballots, the voter would have to be allowed to take an able-bodied carer whom they trust into the voting booth to help them.
- Hello? Neither does a pencil-and-paper ballot, or any kind of voting machine. Once the paper goes through the slot, or you have pressed the voting button, there's no changing your mind. What my system does have is a two-stage, "choosing and confirming" process. You get plenty of time to make your mind up. If you were taking so long that the meter fell out of the green zone, you would have to call to the Presiding Officer {who can see whether or not the master counter has advanced} for a recharge.
- As currently happens in pencil-and-paper ballots, the Presiding Officer would notice any attempt to tamper with the election process and inform the police that a crime was being committed against the Representation of the People Acts.
The system is not designed to replace common sense -- ultimately, nothing can. A Presiding Officer is still required. However, the counting of the votes is done instantaneously {it has actually been going on the whole time}. But most crucially, the system is absolutely open to public scrutiny at every stage -- and uses science that should be capable of being understood by anyone with a high school education.In some organisations where membership cards are used, the membership cards have a row of numbers which are punched out to signify voting in elections at meetings.
Software developer my backside. You're probably just some snot-nosed whining kid who fails it. Do you honestly imagine your "competitors" are any less smart than you are? Or that you could gain nothing of worth from a look at their code, which they would be obliged to show you? If you really are a software developer, you're going nowhere with that attitude. It's not the 1980s anymore, pal.
The Government is there exactly to tell people how to do business. We -- well, those of us who actually have jobs, that is -- pay taxes to the government precisely to stop unscrupulous people exploiting unfair advantages. Software developers keeping the source code from users are misusing an unfair advantage, by preventing users from exercising their natural rights to study the software they use and adapt it to their needs. Nobody ever tried to stop you from taking a clock to pieces to see how it worked, or shortening the legs on a table to make it more comfortable -- why should it be any different with computer software?
Well, it may not make sense to most users, but that is not the point. Nobody who installs a fire extinguisher ever expects to have to use that either. They're just glad of it when they need it.
The world does not need another closed-source browser. In fact it does not need any more closed-source software at all. What the world needs is for some government somewhere to decide that every computer owner has the right to view the source code of any program running on a computer that they own, and if the software vendor doesn't like that then bollocks -- users outnumber developers. Think about it ..... Food has to be labelled with its ingredients list and mean nutritional analysis. Clothing has to be labelled with its fibre composition. Cosmetics have to be labelled with their ingredients. Beer has to be labelled with its alcohol content, fags have to be labelled with tar and nicotine content ..... it makes absolutely no sense that software should be supplied without the corresponding source code.
That is what I said. The losses on both the primary and secondary sides -- due partly to resistance heating and mainly to badly-designed cores -- can all be lumped back to the primary side. Diodes that don't rectify properly {even the difference between 50Hz and 60Hz is noticeable with some of the two-terminal devices laughingly described as "diodes" by some Far Eastern manufacturers} and capacitors that leak both contribute to inefficiency. OTOH, running an appliance from even the least-efficient plug-in PSU you can find is still a hell of a lot greener than running it on disposable batteries.
Also as I said before, part of the problem is that most appliances are not plugged directly into the wall, where each socket always has its own separate on-off switch, but run from an extension lead; and extension leads do not usually have a separate switch for each socket. I don't know why this should be the case, apart from sheer cheapness.
As for the 7805, that is just a series regulator: a zener diode acting as a voltage reference, and a transistor in emitter-follower mode amplifying the current. The excess energy gets turned into heat. If you feed it with 13.5V and try to draw 150mA, it will drop 8.5V at 150mA, giving over a watt of heat dissipation. In general, the efficiency of such devices = (output voltage / supply voltage) * 100%. There are better devices available. These use a slightly different form of SMPS, in which a capacitor is partly-charged through an inductor to limit the current; this time, the transistor is switched hard off and on, so the only resistive loss is due to the parasitic resistance of the inductor. Unfortunately, they aren't simple drop-in replacements, because the inductor isn't something that can be built on-chip.
..... when vacuum-tube valves were all there were, "Standby" was a power mode on a well-built stage amplifier. This provided power only to the heater filaments; the HT was disconnected. The idea was simply to allow the amps to warm up without any danger to the speakers {or an unlucky roadie replacing a defective valve}.
Modern appliances with a remote control need to draw just enough power to run the sensor and decoding circuitry -- otherwise they would invariably have to be switched on at the wall socket. And since nobody has enough wall sockets, everyone uses multi-way extension leads; but extension leads rarely have individual switches like wall sockets.
Also, even modern appliances without remote controls tend to have the on-off switch in the secondary circuit of the power transformer. This seems to be something to do with a general reluctance of manufacturers to touch mains. In an ideal world, this would not matter; but in this real world, copper is not a perfect conductor and steel loses its magnetism. So a transformer is not 100% efficient {although it's about the closest anyone's ever built}. By the Principle of Equivalence, we can represent all the losses as a simple resistor in parallel with the primary winding.
Switched-mode power supplies are becoming more common {laptops, and even most telephones except Nokia, use them for their rechargers}. The basic principle of a switched mode supply is that you rectify the incoming mains to DC; then use a high-power oscillator to turn it back into AC but at a much higher frequency, typically in the 25-50kHz range {you could go even higher but LW radio starts around 150kHz and you have to be careful to avoid interference} which can then be passed through a physically smaller transformer, as the current spends less time flowing in any one direction so the core can be smaller {every iron atom is like a miniature compass needle which tries to align itself with an externally-applied magnetic field, which in the case of a transformer is imposed by the current applied to the primary and opposed by the current drawn from the secondary; when all the "needles" are aligned in the same direction, the core is said to be saturated and the winding is now behaving as a resistor which is not what you want. At 50Hz the current flows for 10ms in one direction, then 10ms in the other; at 25kHz it flows for only 20us in each direction}. In an SMPS it is mainly the leakage current of the input smoothing capacitor which determines the wastefulness.
If Standby mode is not going away {and it isn't} then the best we can hope for is to ensure through regulation that new appliances are wasting as little power as possible in Standby. And while we're making legislation for new appliances, this would be an ideal time to consider a few other factors; such as minimum standards for product lifetime {I still feel inclined to say "forever" would be an acceptable minimum}, field maintainability, manufacturer responsibility for end-of-life recycling, and whether working standards in the country of manufacture would be acceptable in the destination country.
It sounds as though they are basically saying that a fake dog is a poor substitute for a real dog. Not much surprise there.
..... er ..... Staffordshire Bull Terrier Crossbreed lying alongside me. Her powerful, muscular jaws, packed with razor sharp teeth, were just millimetres from my face. Yet at no time did I feel any cause to be afraid. Not because she was programmed not to harm me, like some robot, but because she had chosen not to harm me. That was a great feeling, but it also made me aware of my own responsibilities to her. As she drifted off to sleep, her legs began to twitch and she gave out a few little high-pitched barks; no doubt she was dreaming her wolf-dreams, running with the pack. And with the light of dawn, she would lick me awake.
There is something about the relationship between human and dog that cannot quite be replicated artificially. It is a real two-way relationship: we are accepted into our dogs' homes just as much as they are accepted into ours. It's simply not possible to feel the same way about a piece of machinery, however pretty you try to make it. Artificial intelligence to date still looks very artificial; it's not so much that smart programs are passing the Turing test, but dumb humans are failing it.
Just the other night, I was lying in bed with my pit
No machine is ever going to replace that.
In almost any review of software, failure to mention the licence terms {GPL, BSD, Other Free, Non-Free} is tantamount to a criminal omission. The only exception would be in a publication concerning itself only with one particular licence, or in a jurisdiction where the Four Freedoms are enshrined in law.
Funny as the mods mod'ed it, this was actually meant as a serious suggestion. But they are unlikely to come back and explain their reasons now, since to do so would undo their moderation ..... guess I'll have to write it off as an example of /. weirdness.
The Indian machines are actually a lot more sophisticated than my description -- they're fully electronic, no moving parts. But that in its own turn makes them harder for a lay-person to scrutinise. OTOH, the Indian implementation of the "unique token which is exchanged for an anonymous token" is wonderfully simple and next to abuse-proof. Token One is the voter's own skin, and is "spoiled" by applying a dye which will take longer to remove than the polls stay open for -- unless you use some technique such as bleach and a scrubbing brush that will leave a mark of its own.
I came up with the "solenoids in series" idea as a way to get around the need for mechanical interlocks between the counters -- my first idea was for each candidate counter solenoid to have a peg that caught in a slotted strip running along the row and linked to the master counter, so any one would move on its own and the master counter but no others {the slot would slide straight past the peg of any non-energised counter}. But that would require specially-made parts; you can buy non-interlocked counters off the shelf and the coil impedance isn't exactly a secret from anyone with an AVOmeter. Doing the interlocking electrically should work just as well, as long as the coil resistances are within a certain tolerance of one another and the counter mechanisms are not too worn {another good reason to use non-resetting counters and record the before and after readings, rather than using resetting counters: there is an instantaneous reading of how worn the counters are}.
This system is still second-best to pencil and paper and counting by hand though, IMHO. As another reader points out, you would need to have separate machines for each question being asked; it doesn't do away with most of the traditional requirements of the Presiding Officer; and it's only good for a "first past the post" system where votes can be counted as you go along and still be meaningful. Most of the world actually uses Single Transferrable Votes, and I'm not sure that there is any way to count these on-the-fly. Each vote probably would have to be stored on a machine-readable {but also, for transparency's sake, human-readable} ballot paper or something like it, and these run through the machine several times {so the votes were counted from the actual human-readable ballot papers and not some auxiliary memory susceptible to tampering; and also to minimise the total parts count, which makes the scrutineers' job easier}. This raises the question, how do we ensure that the votes are not kept in chronological order {which might give some traceability} whilst only allowing them to be counted once each? Separate cards for each vote, maybe; or a continuously-circulating loop of paper tape which would be punched in essentially random positions.
This is my idea for a voting machine. It depends for its operation on the idea that when a current is passed through two solenoids in series, both armatures will pull in. The machine itself has two units: the voting booth unit and the presiding officer's unit, linked by a cable. When not being used for an election, the machines would be made available for public scrutiny.
The voting booth unit {VBU} has a large rotary switch, a pushbutton and a meter with a green zone. The Presiding Officer's unit {POU} contains a power supply, and a column of non-resettable electromechanical counters, all but one of which are covered by a metal plate. This plate is fastened in place with a wire with an aluminium seal bearing the Returning Officer's mark. The counter readings before the start of the election are recorded on a paper label affixed to the underside of the cover plate. There is also a switch labelled "CHARGE" and "VOTE".
Each voter is issued with a unique, identifiable token -- a postcard with their name and address on it. The voter shows the token {Token One} to the Presiding Officer, who first spoils Token One and then moves the switch on the POU to "CHARGE" as the voter steps into the booth. The Presiding Officer then moves the switch to "VOTE". The voter has now traded Token One for a second token, all of which are absolutely anonymous, identical and indistinguible from one another: Token Two is an electrical charge stored in a capacitor contained within the VBU.
The voter spins the rotary switch to their preferred candidate, checks that the meter is in the green zone and depresses the voting button. The VBU capacitor is discharged through the coil of one of the concealed counters in the POU. One terminal of each of these counters is commonned together; the current through any one of the candidate counters also flows through the master counter, and returns to the other plate of the capacitor. The charge in the capacitor is soon exhausted, and cannot be replenished unless the Presiding Officer moves the POU switch to CHARGE. The voter then has the option to move the rotary switch to a different position so as to conceal their preference -- or to leave it there to advertise their preference.
Every voter has a receipt to show that they have voted {the spoiled Token One} but once a vote has been cast, the only record of that vote is the fact that the master counter and one of the candidate counters have advanced by one place. There is thus no way to link a voter with their vote. The master counter is in view of {and the counting mechanism is within earshot of} the PO, who can thus confirm visually and aurally that a vote has been cast {or separately, manually record a "no vote" if the voter leaves the booth without voting for any candidate}. All the candidate counters are concealed until the close of polling, when a few minutes' worth of mental arithmetic will reveal the true count. By virtue of its simplicity, and the fact that it has been subjected to public scrutiny, we can take for granted that the mechanism is behaving as it is supposed to; the Returning Officer need only inspect the tamper-evident seals to determine whether the result is valid or compromised.
{In case the above constitutes a patent claim, I hereby licence it for use royalty-free in all applicable jurisdictions, in the hope that it will be of service to Humankind}.
Well, I'm not a paying customer; but I don't get advertisements either since I set up Squid to block any site serving adverts.
We're better fed now than we've ever been -- but there are still people with eating disorders, and there isn't a single foodstuff that is agreed by all experts as safe to eat. Crime is at the lowest level it's ever been at for years -- but the Authorities are stoking up the fear of crime as an excuse to invade our civil liberties. I'm reminded of a lyric from a song by Del Amitri: It seems that Doom and Gloom just make for better news than boring nice stuff. Nobody is interested in fluffy kittens unless they're being brutally hacked to pieces, or trees and flowers unless they're deadly poisonous. And never mind about all the lives that have been saved just because someone had their mobile with them
What else is there to say? It's a bit of legacy code left over from the days when it was safe to assume that any code on a computer had been put their with the owner's knowledge and consent. That assumption has since been invalidated by subsequent events. A backdoor it may be -- but when it was put there, there used to be a fence around the back garden.
And this is just one example of a whole class of things that are really, seriously, terribly wrong with Windows {and for that matter, closed-source software in general}. A lot of benign application software has come to depend on behaviour in the operating system that ought never to have been allowed in the first place -- behaviour that makes the propagation of viruses and worms so much simpler. Now, if Microsoft change the way Windows works so as not to just hand out permission for any process to interfere with any other process, then the worms and viruses that depend on this behaviour will die off -- but so will all those applications that depend on this broken behaviour. Then what used to be a choice between "Stay with Microsoft, and all your old software will still work like it did before" and "Leave Microsoft, and none of your old software will work anymore" becomes one between "Continue to splash out good money after bad to Microsoft, but none of your old software will work anymore" and "Wave goodbye to Microsoft, none of your old software will work anymore but there are better-than-adequate replacements for all of it".
And my prediction? A company that still makes extensive use of an obsolete software product will find themselves -- and their precious data -- orphaned as a consequence of the switch to Vista. They will have to obtain a pirated copy or copies of an earlier Microsoft OS {because there is no way to obtain a legitimate one} just in order to read their own files. This will only be discovered in a Licencing Gestapo bust.