Wikipedia:Reference desk/Archives/Science/2009 April 20

= April 20 =

Amino acid properties
are amino acids always solids? It's a question on our daughter's nursing homework and we can't find the answer. thanks. —Preceding unsigned comment added by Marksilv242 (talk • contribs) 01:31, 20 April 2009 (UTC)


 * Given the context, I assume you mean the "standard" one most commonly found in proteins? You can look at our Proteinogenic amino acid table and check the melting-point of each. In the more general sense, our amino acid article talks about their structure, and in particular the idea of a zwitterionic form and its physical properties. Is there a "rule of thumb" regarding being an ionic structure and having either generally high or generally low melting point? DMacks (talk) 01:52, 20 April 2009 (UTC)


 * Well, they decompose before they melt. But any covalent molecule is going to become a gas at sufficiently low pressure, isn't it? Looie496 (talk) 02:43, 20 April 2009 (UTC)
 * True 'nuf. I guess I was thinking "solid at room temp" and ignoring what actual other state or physical change occurs when they the solid falls apart:) A real question is whether they really behave as "covalent molecules" (implying neutral, not covalent ions) or as ions. For example, Glycine is fairly low molecular weight, but it listed as a decomposition (rather than melting) temperature of 233 °C: there are clearly important ionic interactions and intermolecular stabilization. DMacks (talk) 04:24, 20 April 2009 (UTC)

Could we say that in (terrestrial) biological contexts, amino acids are always solids and almost always dissolved in water? ike9898 (talk) 20:00, 20 April 2009 (UTC)


 * The Zwitterion form makes amino acids more behaving like NH3+ caboxylates. Like most acid and amine salts they have higher melting points as the amine and carboxylic acid they consist of. So most of the amino acids I know are solids. --Stone (talk) 20:52, 20 April 2009 (UTC)
 * Solubility is interesting. Standard guide is "ionic stuff is water-soluble", but our amino acid article in the "Zwitterion" section notes: "Zwitterions have minimal solubility at their isolectric point and amino acids are often isolated by precipitation from water after adjusting the pH to their isolectric point." DMacks (talk) 21:01, 20 April 2009 (UTC)

Are there any (real) animals that can spit/spray corrosive substances as a defence mechanism?
Question as topic. When I say 'corrosive', I mean 'literally corrosive' (as in substances of the type mentioned in the linked article), not your run-of-the-mill toxic/foul-smelling compounds. I'm pretty sure that I once read about a species of spray beetle that has two specialized glands, producing two chemicals which are harmless alone but which when mixed together and concentrated into a high-pressure jet at the point of exit, form an incredibly nasty, highly-caustic substance (though I might be wrong). Thanks. --Kurt Shaped Box (talk) 01:45, 20 April 2009 (UTC)


 * Bombardier beetle? --98.217.14.211 (talk) 02:06, 20 April 2009 (UTC)


 * That sounds about right, thanks. I'm sure that I saw a similar species on a wildlife show that could mix up and spray a pyrophoric substance too... --Kurt Shaped Box (talk) 02:15, 20 April 2009 (UTC)


 * I'm guessing it depends what you mean by 'literally corrosive' and 'defence mechanism'. The formic acid a number of ants use for defense (and offense) is I presume somewhat corrosive as with most acids although I expect it may not be the only (or even primary) compound in some cases. Fire ants use an alkaloid venom Solenopsin which has "cytotoxic, hemolytic, necrotic" properties. I'm not sure what the pH is but I'm guessing it's high enough that the substance could be considered literally corrosive even if that isn't a key part of the stings action. It seems the same it's the same for bee stings  while wasp stings are usually alkali (not sure what they contain). But in both cases, the same as with fire ants it's not really the key mechanism of action of their stings. Formic acid is also used by a number of other insects it appears . As far as I'm aware, in none of these cases are the substances mixed outside the body, they're just (I presume) stored in specialised glands. Finally another example snake venom usually has digestive enzymes, would you consider them corrosive? Nil Einne (talk) 12:53, 20 April 2009 (UTC)


 * And vultures will occasionally vomit on things as a form of defense. I think most creatures have some form of acid for digestion.  —Preceding unsigned comment added by 65.121.141.34 (talk) 13:42, 20 April 2009 (UTC)


 * Well, if you put it that way... Yeah, gulls and fulmars can also spray their vomit at will when threatened. I guess that I was thinking more along the lines of species with defence mechanisms that work *primarily* by inflicting corrosive chemical burns upon a potential predator. --Kurt Shaped Box (talk) 00:06, 21 April 2009 (UTC)#


 * Camponotus saundersi this ant has an explosive gland that covers attackers in formic acid. Quite a neat defence and shows that it's not only humans who are suicide bombers. Smartse (talk) 12:40, 21 April 2009 (UTC)


 * Either of the blister beetles (Cantharidae) excrete Cantharin from a gland at thier "knee" when threatened and yes IT BURNS. 67.193.179.241 (talk) 04:40, 22 April 2009 (UTC)Rana sylvatica

Kinetic energy of gases
Hi, I am doing a physics problem, I am given values of $$P_1$$, $$V_1$$ and T for some gas (assumed ideal) (note: it is done by web interface, values change every time) and asked to find the total translational kinetic energy of the atoms. The answer works out to be $$\frac{3PV}{16}$$ every time, and I'm not sure where that extra factor of 1/8 is coming from, could someone please help me? --130.216.30.234 (talk) 08:23, 20 April 2009 (UTC)
 * Note: I have looked at the article on kinetic theory, and am still not getting it. --130.216.30.234 (talk) 08:25, 20 April 2009 (UTC)
 * Also: related question (the one immediately following) is I am told that the balloon in the first part of the question (the one with $$V_1$$ of gas at $$P_1$$ and T) has been filled from some container containing $$V_2$$ of the same gas at $$P_2$$ and T, and asked how many of those same balloons could be filled using all the gas. The correct answer is always $$\frac{8 P_2 V_2}{P_1 V_1}$$, or $$\frac{P_2 V_2}{\frac{1}{8} P_1 V_1}$$. There's that 1/8 again!! I have no idea where it's coming from! --130.216.30.234 (talk) 08:46, 20 April 2009 (UTC)
 * For the first part, I'd say that the answer is $$\frac{3PV}{2}$$. I don't think any such factor of 8 exists. Martlet1215 (talk) 10:02, 20 April 2009 (UTC)
 * But that's not the answer. That's my point. --130.216.30.234 (talk) 10:04, 20 April 2009 (UTC)
 * It is the correct answer. Either the answer you've been given is wrong, or there's some relevant feature of the question you're not telling us. Can you give us the exact wording, or just a link to the problem? Algebraist 10:32, 20 April 2009 (UTC)
 * Okay, I was seconds away from posting it here when I just realised that I'm an idiot. In the original question, you had to calculate the volume of the spherical container using diameter, and I was not halving it before calculating volume, which of course would add an extra $$2^3$$ to my answers. Problem solved. Thanks for your time, everybody! --130.216.30.234 (talk) 11:34, 20 April 2009 (UTC)

Azimuth sweep angle?
In the game Call of Duty 4, in one mission you control the guns on a AC-130 Spectre Gunship. A guy on the plane says a couple of times: "Recalibrate azimuth sweep angle, adjust elevation scan." What does it mean? And yes I'm the guy who asked the Coriolis effect and sniping question. Just a big fan of COD4. And thanks for your answers. —Preceding unsigned comment added by 116.71.54.217 (talk) 08:52, 20 April 2009 (UTC)


 * There's a handy article on azimuth that might be of help. --superioridad (discusión) 09:01, 20 April 2009 (UTC)


 * Azimuth and elevation angles are the lateral and vertical aiming angles respectively. There are pictures of separate antennas that scan (sweep) these angles at Microwave Landing System. Cuddlyable3 (talk) 10:59, 20 April 2009 (UTC)

Infinite speed and perception
I first thought of posting this on the Language ref desk, then changed my mind. Isn't something grammatically wrong with this line from Speed of light: "Francis Bacon argued that the speed of light was not necessarily infinite, since something can travel too fast to be perceived."Jay (talk) 10:43, 20 April 2009 (UTC)
 * I can't see anything wrong with it. What do you have in mind? Algebraist 11:03, 20 April 2009 (UTC)
 * The grammar is ok but the logic is questionable because the underlying hypothesis about "something" is not explicit. The OP could bring this up on the article discussion page to see what others think. Would the following sentence be better? "Francis Bacon argued that the speed of light was not necessarily infinite, since it could just be travelling too fast to be perceived."Cuddlyable3 (talk) 11:12, 20 April 2009 (UTC)
 * Yes, sorry about that. I should have said logically wrong instead of grammatically wrong. Cuddlyable3's version is making sense to me now. The meanings I had in mind were: "Francis Bacon argued that the speed of light was not necessarily infinite, since nothing can travel too fast to be perceived.", and "Francis Bacon argued that the speed of light was not necessarily infinite, since something cannot travel too fast to be perceived.", both meanings are opposite of the original line. Jay (talk) 11:29, 20 April 2009 (UTC)


 * I think it's vague about what is being perceived; I mean, light is definitely perceived so he can't be referring to that. I assume he's referring to the finite speed being perceived i.e it's really fast and we can't tell the difference between that and infinite speed.  How about
 * "Francis Bacon argued that the speed of light was not necessarily infinite, since it could be travelling at a very large but finite speed which was beyond the scope of contemporary experimental accuracy."
 * Then again, I could have misinterpreted his point. Martlet1215 (talk) 11:37, 20 April 2009 (UTC)
 * I think you nailed it. Light is perceived as something instantaneous in all "normal" situations, but that is just an artifact of imperfect observation - and that was the point Bacon was making. --Stephan Schulz (talk) 11:49, 20 April 2009 (UTC)
 * User:Wile E. Heresiarch who added the line in 2004, has since left Wikipedia, and it is hard to confirm the source. This reference however says that Francis Bacon believed the speed of light to be infinite! Jay (talk) 12:25, 20 April 2009 (UTC)
 * It says that Roger Bacon believed it to be finite, though. Perhaps Heresiarch confused the two. Algebraist 12:29, 20 April 2009 (UTC)
 * Guys you should take your content discussion to this talk page Cuddlyable3 (talk) 13:30, 20 April 2009 (UTC)


 * It seems like the original quote implies that Bacon argued that it's possible to imagine a thing that travels imperceptibly fast but not infinitely fast, and that light might be one of those things.  User:cuddlyable3's version could be less clear depending on what Bacon actually said. Cuddlyable3's version implies that he specifically theorized that light could travel very fast, but doesn't specify whether Bacon thought this was a special case or whether other items could also attain imperceptible speeds. I have no idea which is correct, but they don't mean exactly the same thing.
 * This distinction might not matter at all in context. I dunno. APL (talk) 13:46, 20 April 2009 (UTC)
 * I'd can "can" and replace it with "could". Clarityfiend (talk) 15:53, 20 April 2009 (UTC)
 * This guy at least included a direct quotation from (Francis) Bacon and a citation: "'Even in sight, whereof the action is most rapid, it appears that there are required certain moments of time for its accomplishment...things which by reason of the velocity of their motion cannot be seen—as when a ball is discharged from a musket' (Philosophical Works of Francis Bacon; J.M. Robertson, Ed.; London, 1905; p. 363)." The ellipsis is awkwardly placed, though, and I don't happen to have a copy of the Philosophical Works lying around in which to examine Bacon's full argument. Deor (talk) 17:01, 20 April 2009 (UTC)
 * I should have guessed that Google Books would have snapped this up. If one reads the whole chapter, one can say only that Bacon seems to be rather unclear whether he's discussing physics or physiology. Sometimes he seems to be maintaining that light has a finite speed; at other times, he seems to argue that it's our visual impressions that require time for their formation. And I admit to not being able to make much sense of his discussion of the idea that starlight takes time to reach us: he seems to say that he has discarded his initial suspicions that this is so, and now believes that our perception of stars is instantaneous. Deor (talk) 18:00, 20 April 2009 (UTC)
 * I don't think it's the light that's being not-perceived it's the speed of light. If you have a friend standing on a far distant hilltop with a mirror and you light a candle - the time it takes for the light from the candle to travel to the mirror and back again is too fast to perceive.  Bacon argues that the time might not be zero - but merely so short that you can't perceive the delay...smart guy...he turned out to be correct!  SteveBaker (talk) 16:58, 20 April 2009 (UTC)

Scooter electric motors
I know how electric motors work (I've had to build them by coiling wires by hand). I was discussing scooters the other day and I thought about coasting. A motor with high torque is hard to turn by hand if there is no power. So, assuming scooters use high torque motors to move a lot of weight with one small motor, are those scooters able to coast (free-spinning axle with no power applied to the motor)? If so, is there a special type of electric motor that allows for a free-spinning axle or is my memory of electric motors completely wrong? -- k a i n a w &trade; 14:14, 20 April 2009 (UTC)
 * Do they have a clutch? Rmhermen (talk) 16:04, 20 April 2009 (UTC)


 * I would do it with a clutch - somewhat backwards than a slipper clutch. I was just wondering how it is done in real scooters (and I'd rather not hunt one down and take it apart). --  k a i n a w &trade; 18:39, 20 April 2009 (UTC)


 * Unless there is some load on the engine, I don't see why it should be hard to turn. Where would the energy be going? Electric motors can be tough to start spinning, because the rotor settles in a low-energy state. But once you're out of the bottom of the potential well, you get back the energy if you slip into the next. It's hard to notice if you turn the axle by hand, but obvious if you install e.g. a flywheel. You can see a similar effect with hub generators in bicycles - if you do not attach a load, a free spinning wheel will go a long time. --Stephan Schulz (talk) 08:54, 21 April 2009 (UTC)


 * When you spin an electric motor, it acts like a generator - so if you have a load (like the battery of the scooter) connected to the motor - then it may be harder to turn. For coasting, all that should be necessary is to install a switch that disconnects one of the wires leading to the motor - and it should free-run without problems. SteveBaker (talk) 19:42, 21 April 2009 (UTC)


 * How is "coasting" on an electric scooter different from cruising at constant speed? Is the questioner referring to an electric bicycle, where the harder you pedal the faster the motor makes it go to supplement your efforts, or an electric motorcycle where the battery provides all the propulsion? Regenerative braking uses the motor like a generator to put energy back into the battery (way less than 100% efficiency) rather than slowing by friction in a brake which turns the energy into waste heat. I agree that an electric motor should turn freely unless the bearings are shot or a winding is connected to something to brake electrically. Edison (talk) 22:52, 21 April 2009 (UTC)
 * When 'coasting' (at least on level ground) your speed doesn't stay constant. It slowly decreases due to friction & drag.  In order to cruise at constant speed, you have to supply energy to the motor.  SteveBaker (talk) 01:42, 22 April 2009 (UTC)


 * If you live in flatland, there is little need for coasting. With hills, you use battery power by turning the accelerator to get up one side of the hill.  Then, you coast down the other without any accelerator at all.  Similar to a car - you press on the gas to go up a hill and take your foot off the gas to coast down the other side.
 * Steve mentioned something that caused my initial concern. An electric motor is also a generator.  If you force the axle to spin, it creates electricity.  Part of the energy is used to create the electricity.  My understanding is that the electrical potential is there with or without a battery connected.  As reference - try spinning a starter motor for a car by hand.  It is very hard.  So, if that was the motor for your scooter, when you let off the accelerator, the scooter would quickly come to halt.  That is assuming that the motor is directly connected to the wheel with no clutch or slip gear of any kind. --  k a i n a w &trade; 16:55, 22 April 2009 (UTC)
 * No, you're missing the point. It's only hard to turn a car starter if it's connected to something.  If you unhook all of the wires, it spins quite easily (unless it's broken or something!). SteveBaker (talk) 19:52, 22 April 2009 (UTC)


 * Yes - that is exactly my question... I thought it was always hard to turn. Being easy to turn when disconnected means that it is possible (though may not be the case) that you could connect an electric motor directly to a wheel and when power is disconnected it will allow the vehicle to coast along.  Perhaps a fancy circuit could be in place to recharge a battery while it coasts, as long as the load of the battery doesn't kill the ability to coast.  That, however, is a completely different topic.  I was only wondering about the motor's ability to allow for free spin. --  k a i n a w &trade; 00:32, 23 April 2009 (UTC)


 * My wife has owned a number of electric scooters - they don't have a clutch and they are not able to coast. In fact, they don't have any brakes either - cutting the power to the motor is the only way to stop, as the motor works as a brake. They do have a lever at the back which disengages the motor to allow you to manoeuvre the scooter into a small space by pushing it, but this is not in a such position that it could be used whilst riding on the scooter as you would have no way of braking. I suppose some people with limited mobility may enjoy the adrenalin rush of careering down a hill at speed in an unbraked vehicle (the thought of it takes me back a few years!) but the majority probably wouldn't :) Richerman (talk) 12:09, 27 April 2009 (UTC)

ctrl+f If You Can't Find It
What does " ... compression molded ... " " ... compression molded ... " click on the Technology TECHNOLOGY link (bolded and underlined) mean?68.148.145.190 (talk) 16:28, 20 April 2009 (UTC)
 * We have an article about it: Compression molding. Does this help? -- Kateshort forbob  18:06, 20 April 2009 (UTC)

disappearance of gills in higher vertebrates and evolution of lungs (comparison with embryonic tadpole -> frog transition)
The whole process of an evolutionary mechanism in which the gills could be lost while the lungs developed seems kind of complex, because IIRC the gills are not evolutionarily homologous (or embryonically homologous) to lungs. So how would this sort of thing come about? Especially with the development of the circulatory system, since this would suggest a sort of fantastical rewiring that seems amazing to me -- suddenly you have to rerout the systemic and gill circulations back to the heart (whereas the gill circulation used to flow into the systemic circulation)? Can someone just clarify the whole process with me? Thanks. John Riemann Soong (talk) 17:45, 20 April 2009 (UTC)


 * (uh, What are you asking for clarification on? The maturation of tadpoles? Or the evolution of lungs? My reply here assumes that you're asking about evolution.)
 * I think it's assumed that animals with gills developed lungs in addition (possibly mutated swim bladders), and then long afterwards dropped the gills. Lung Fish are still in that transition state.  Check out Lung. APL (talk) 18:29, 20 April 2009 (UTC)


 * Lungs and swim bladders did indeed evolve in addition to gills, probably in response to an environment featuring frequent oxygen-depleted water (such as lungfish inhabit today) to enable the fish involved to literally swallow air and absorb additional oxygen. However, lungs did not develop from earlier swim bladders: either the two homologous structures evolved as near parallel alternatives in two separate lineages, or (so I have seen suggested) lungs evolved first and then in some lineages that no longer needed so much supplementary oxygen (having returned to deeper waters?) they became swim bladders, still sometimes with oxygen absorbing capabilities but primarily now employed for buoyancy control. I love how often evolution repurposes organs for new tasks. 87.81.230.195 (talk) 22:59, 20 April 2009 (UTC)

Also, do tadpoles have 2-chambered hearts like fishes, or 3-chambered hearts like adult frogs? John Riemann Soong (talk) 17:53, 20 April 2009 (UTC)


 * Heart formation in a tadpole is a pretty damn complicated process. As far as I remember, it starts with one chamber and ends with three. For details, you can see, for example, Mohun et.al. "The Morphology of Heart Development in Xenopus laevis", Developmental Biology 218, 74–88 (2000). --Dr Dima (talk) 04:58, 21 April 2009 (UTC)


 * Now, regarding your original question, the appearance of lungs and disappearance of gills occurred naturally with the transition from aquatic to terrestrial lifestyle. Some transitional, semiaquatic, or mud-dwelling species possess both lungs and gills. Axolotl - neotenic salamander - has both lungs and gills, for example. By the way, vertebrates are not "special" in this respect. Coconut crab, too, has both gills and lungs, although gills are largely useless in an adult. --Dr Dima (talk) 05:14, 21 April 2009 (UTC)
 * As an aside - Coconut crabs are VERY cool. They often grow to three feet across and there are reports of six-foot coconut crabs - which are the kinds of thing you'd expect to see in old sci-fi movies!  Kids in the area where they live keep them as pets and they can live for 30 years!  Check out the article. SteveBaker (talk) 19:39, 21 April 2009 (UTC)

Witnessing evolution
I'm very interested in evolution, and I basically want to ask whether it's possible for someone who is not a scientist (I'm a university student though) to see evolution take place, and if so what would be the best way? Keep in mind that my knowledge of biology is rather limited but you would probably want to growing something like bacteria (possibly in a chemostat?), flies or fungi? I've heard of a fly experiment where a fly was given a different diet and in some 8 generations it had evolved. --BiT (talk) 18:30, 20 April 2009 (UTC)


 * Well, there are certainly experiments you can do with bacterial cultures that allow you to see evolution happening before your very eyes - but without some practical knowledge of lab techniques, I think they'd be hard for you to perform. 8 generations sounds a little short for the fly experiment - and in order for any evolution to happen at all, you'd need a very large number of flies to ensure a sufficiently diverse gene-pool.  There are computer simulations of evolution that work fairly well.  Richard Dawkin's "biomorphs" for example (see Weasel program).   Google for "biomorphs" and you'll find several online versions of the program - and a bunch more that you can download for free. SteveBaker (talk) 21:05, 20 April 2009 (UTC)

Not sure how you would get access to a lab strain of E. Coli, but I agree that it's probably the easiest/most visible experiment you could try (the wiki page for it is here), due to the small size of bacteria and the extremely quick rate of reproduction, as well as the relative ease of culturing and "harmlessness" of the E. Coli itself. There are some other bacteria listed at Model organism. 124.154.253.25 (talk) 02:35, 21 April 2009 (UTC)


 * You could take thousands of Arabidopsis thaliana seeds, germinate and grow them and then spray them with Roundup. If any survive then they must have mutation that lets them grow when roundup is present. Smartse (talk) 12:45, 21 April 2009 (UTC)


 * Yes, you could...theoretically. But one round of that test doesn't demonstrate evolution at all.  A skeptic would rightly say that while 'Roundup' only kills 99% of plants, it's perfectly possible that the 1% survive by pure luck for environmental rather than genetic reasons.  You'd have to go on and collect seeds from the survivors of that crop - grow them into plants and spray THEM with roundup in order to demonstrate a significant statistical improvement in the survival of the second generation - and thereby get any idea whether the roundup-resistance is inheritable.  In fact, since it's highly likely that some plants are indeed surviving for environmental rather than genetic reasons - you'd probably need to do this over many generations in order to show a really convincing effect where 100% of plants survive being sprayed.  You really need to prove a substantial increase in the percentage of plants that survive AND that this trait is inherited - or else all you've done is shown that some plants survive Roundup rather than that they EVOLVED a natural resistance to the herbicide.  You might also need a VAST number of plants in each generation.  If only one plant in a million carries some random mutation for roundup-resistance then you'll need to grow several million plants in order to have enough of them survive to make a million seeds for the second generation of the trials.  Since the plants might take a year to grow to maturity and produce seed - this experiment could easily require acres of land and dozens of years in order to produce results.


 * That's why we do these kinds of experiments with bacteria - where an experimenter can easily have a few million bacteria available - and they reproduce and mature over hours rather than years. That's not to say that the experiment wouldn't work - just that you'd need to dedicate a LOT of money and years of your life to doing it using plant seed. SteveBaker (talk) 13:46, 21 April 2009 (UTC)
 * You would also need to be careful where you got your seeds - if they all come from the same, or a small number of, plants then there might not be sufficient genetic variation for any of them to survive due to genetic reasons (if you want to rely on random mutations, you need far more generations and a far larger population). --Tango (talk) 14:23, 21 April 2009 (UTC)


 * Do you want to witness evolution in general, or just in living creatures? See evolutionary algorithm. I know people have evolved models for Soda Constructor. — DanielLC 21:13, 22 April 2009 (UTC)

Quantum bits at the macro level
I was reading this. It says that one bit of semiconductor memory contains about 60,000 electrons. It struck me that that's a finite number of qubits. A qubit can only be determined once, so a bit of memory can only be read a finite number of times. You can use classical error correction to copy the bit of memory a bunch of times, but that just uses up your number of reads. This makes me suspect: These are first impressions; what do you think? Can you only read a book a finite number of times before the information degrades? If you've already memorized it, is it really a "read" of the qubits? Is the repeated operation of a half adder (or a calculator, or a router) "reading" procedural information that will eventually degrade? .froth. (talk) 20:29, 20 April 2009 (UTC)
 * Any particular piece of information available on the Internet "degrades" (becomes less well-defined) over time as it's read. The more the information is distributed (Wikileaks :D) the less sure we can be of the integrity of the original and its copies.
 * Eventually we'll run out of information! Fortunately we have people typing on their keyboards and people uploading porn to 4chan, adding bits to the system. It's like Shannon entropy: my Linux box uses the unpredictable movements of my mouse cursor to increase the complexity of the parameters to its pseudorandom number generator. No outside input is a very bad idea. Without input the information on the hard drives and semiconductor memories of the Internet would degrade until they pass the manufacturer-guaranteed tolerances of their restoring logic and then you start losing information.

You're atempting to apply qubit theorems, like the No cloning theorem to classical bits. That's not really correct. There is really no limit to the number of times you can copy a classical bit. Dauto (talk) 20:44, 20 April 2009 (UTC)


 * And it's certainly the case that as a practical matter some bits in some computers are read trillions of times a day for years on end without problems and most of the bits on (say) Wikipedia are read maybe a few times a day at most...so the practical answer is also that this certainly isn't a problem. SteveBaker (talk) 20:57, 20 April 2009 (UTC)


 * There's no classical limit but classical bits are an idealization, right? The real world works in qubits, does it not? I know I'm mixing levels, that's the whole question: what do the qubit theorems mean for classical information theory? .froth. (talk) 01:25, 21 April 2009 (UTC)


 * A classical bit is consistent with many different quantum states. Quantum decoherence isn't a problem and the act of reading the classical bit will not change the bit (eventhough the quantum state will be changed), and you can read the bit as many times as you want. The qubit theorems have very little to do with classical information thory. Dauto (talk) 03:00, 21 April 2009 (UTC)


 * A qubit can be read any number of times and it will always give you the same result (in theory). Reading destroys the quantum properties of the qubit, but it doesn't destroy its classical value (0 or 1). Cloning and copying are different things. Copying means that if you started with a probability p of having a 0 and a probability q (=1−p) of having a 1, then you end up with a probability p of having two zeros and a probability q of having two ones. Cloning means that you end up with a probability p2 of having two zeros, a probability q2 of having two ones, and a probability 2pq of having a zero and a one. Both classical bits and qubits can be copied; neither classical bits nor qubits can be cloned. If classical cloning were possible then you could just read a single random bit from /dev/random and clone it to get as many random bits as you needed. In reality, if you need n random bits then you need to read at least n bits from /dev/random. The /dev/random driver has the same problem: it can't invent random bits to give you, it has to get them from somewhere. So there is a limited supply of randomness, and I think this is what you were talking about in your initial post. The randomness supply can run dangerously low on a standalone machine without good sources to replenish it (like a Winchester hard drive or a keyboard in active use). -- BenRG (talk) 13:25, 21 April 2009 (UTC)


 * Thanks for then clarification on the difference between cloning and copying. It is a relief to see that two completely different operations such as the reproduction of a qubit (cloning) and the reproduction of a classical bit (copying) receive different names. Dauto (talk) 14:49, 21 April 2009 (UTC)
 * I think you've misunderstood what BenRG was saying - he specifically talked about both quantum and classical bits being both cloned and copied, not one term for one and one term for the other. --Tango (talk) 22:17, 21 April 2009 (UTC)


 * I hear you. I heard him too. I just don't think what he said about cloning a classical bit makes any sense. A classical bit is either definitaly 0 or it is definitaly 1. It cannot be either one or the other with probabilities p and q=1-p. A quantum bit, on the other hand, cannot be copyed unless it's read first. And once it's read, the wavefunction colapses and it becomes a classical bit anyways. Dauto (talk) 22:43, 21 April 2009 (UTC)


 * "the Shannon entropy is a measure of the average information content one is missing when one does not know the value of the random variable" Yes classical bits too can have probabilities. And thanks BenRG that helps a lot. Basically you can copy bits, you just can't create new bits. Interesting.. algorithms can't create new information no matter how much procedural information the steps of the algorithm contain.. right? Can't map 5 bits of input into 6 bits of output. Fits in nicely with information entropy only increasing or standing still. .froth. (talk) 23:35, 21 April 2009 (UTC)
 * The inability to create information is well-known in information-theory - and mostly pre-dates all of this quantum stuff. If you think about it, that's not surprising because you can use some algorithm to take some set of source data and transform it into more bits of output data - but you could then compress redundant information out of the output by replacing it with the input - using whatever your algorithm is to uncompress it again on demand.  So the actual information content of the output is no larger than the input.  This is why computers can't generate proper random numbers without relying on some external input.  It's been argued that this is why computers can never become intelligent - but that misses the point that humans are equally unable to create new information...it just seems like we can. SteveBaker (talk) 01:37, 22 April 2009 (UTC)