Wikipedia:Reference desk/Archives/Science/2008 September 15

Welcome to the Wikipedia Science Reference Desk Archives
The page you are currently viewing is an archive page. While you can leave answers for any questions shown below, please ask new questions on one of the current reference desk pages.

September 15

Somewhere theres music

Somewhere thers music, how faint the tune
Somewhere theres heaven, how high the moon?
But exactly where is Heaven? —Preceding unsigned comment added by 79.76.225.5 (talk) 00:22, 15 September 2008 (UTC)

It's underneath Charing Cross station. DuncanHill (talk) 00:24, 15 September 2008 (UTC)

It is a place [1] where Fred dances forever with Ginger, and Louis sings forever with Ella [2]. Note that Ginger does everything Fred does, except backwards and wearing high heels. Edison (talk) 02:28, 15 September 2008 (UTC)

For Ginger's Heaven, wouldn't it be Fred who was in high heels and dancing backwards? Clarityfiend (talk) 04:11, 15 September 2008 (UTC)

And maybe having top billing. Julia Rossi (talk) 10:10, 15 September 2008 (UTC)

Ummm, top billing? In Heaven? Saintrain (talk) 14:24, 15 September 2008 (UTC)

The mind is its own place, and in itself, can make heaven of Hell, and a hell of Heaven. --- OtherDave (talk) 12:22, 15 September 2008 (UTC), channelling Milton.

Triple point

Do all compounds have a triple point? Nadando (talk) 04:42, 15 September 2008 (UTC)

I don't believe helium does. Does that count as a compound or would that be just an element? --Bennybp (talk) 11:27, 15 September 2008 (UTC)

(conflict) Helium is a weird one, but according to this diagram I found it still has a triple point. --Shaggorama (talk) 20:03, 15 September 2008 (UTC)

Yes, sort of. To clarify, it doesn't have a triple point between solid, liquid, and gas. The triple point is between solid, liquid, and superfluid. Because superfluid is also "liquid", helium doesn't really have a triple point like all other elements. ~Amatulić (talk) 20:29, 15 September 2008 (UTC)

Copper Sulfate decomposes by losing water at 110° and 150° and decomposing at 650°

No, many compounds decompose when heated and do not form a liquid, instead they turn into a other compounds. Graeme Bartlett (talk) 21:32, 15 September 2008 (UTC)

CO2 goes from gas to solid and back to gas again. I don't think you can ever have liquid CO2 (certainly not at normal atmospheric pressures). SteveBaker (talk) 01:23, 16 September 2008 (UTC)

But at elevated pressure, liquid CO2 is quite common at room temperature. A triple point need not be "at 1 ATM". See Image:Carbon dioxide pressure-temperature phase diagram.svg DMacks (talk) 01:29, 16 September 2008 (UTC)

How do we know there will be no problems with the LHC?

If it created a black hole, wasn't it supposed to grow gradually? It's only been a week, so it might be only the size of a pinhole or so. I'm afraid in about a month or two from now one of those scientists might be saying "Hey, what's that black spot over there? Charlie, you've got to be more careful with that marker, this machine is expensive! Here, I'll wipe it off.....uh,oh." Dr. Carefree (talk) 05:03, 15 September 2008 (UTC)

First, the LHC hasn't collided anything yet, and it shouldn't operate with energies larger than of existing accelerators until next year.

Now, our current theories predict that black holes evaporate through Hawking radiation, and small black holes should evaporate faster. We have no experimental evidence of this, but we're pretty confident on it. See the article for more details. If we are correct, the black holes created by the LHC, if any, should evaporate almost instantly.

If the LHC created a stable black hole, it would be extremely tiny (in fact, smaller than Planck length), and it would have a terribly small - practically nil - gravitational pull. The only chance it could accrete more mass would be direct contact with other particles which, given its size and the fact atoms are 99,99% vacuum, it wouldn't happen very often. There are probably some quantum effects that need to be taken into account here, but I haven't read anything about it yet.

Such a black hole would fall immediately towards the center of the Earth and orbit it. It would probably take millions of years for it to be of any concern. But again, chances are no such black holes will be produced.

Also, we can be pretty confident the LHC is safe because we have natural phenomena (cosmic rays) that replicates the experiments we're doing, even with much larger energies. There's a counter argument you'll see around that these natural collisions could produce "fast black holes", and the LHC would be creating "stationary black holes". This is sort of true, but it wouldn't matter. Neutron stars have huge densities and gravitational fields, and such particles or black holes wouldn't get through them. The fact that there are ancient Neutron stars everywhere in the Universe strongly suggests that there is no threat from such collisions. — Kieff | Talk 05:28, 15 September 2008 (UTC)

Unless I got my math wrong, a black hole of 100kg evaporates in about 8e-11 seconds. A black hole on the order of 1000 proton masses would evaporate in 4e-79 seconds. In that time it could move less than 1.2e-68cm. Now that time is shorter than the Planck time, and the distance is shorter than the Planck length. Things don't get more "instanteneous" than that. --Stephan Schulz (talk) 10:31, 15 September 2008 (UTC)

Black holes would only be produced at the LHC if there are large extra dimensions. If there are large extra dimensions then the value of G that we measure is just a long-distance effective value and the real G is much larger. The Planck units and the black hole evaporation time depend on G, so they change also. -- BenRG (talk) 14:22, 15 September 2008 (UTC)

I DONT BELIEVE ANY OF YOU- WERE ALL GONNA DIE! AAAAHHHH!!! I kid (only a little), but really- should scientists be messing around with this sorta thing? Is it that important to recreate the conditions of the Big Bang? What practical knowledge could we get from this anyways? Everytime science starts pokin around tryin to learn stuff that we have no real need to know, and has great potential to be dangerous, I'm reminded of my favorite quote by HP Lovecraft that goes ""The most merciful thing in the world, I think, is the inability of the human mind to correlate all its contents... some day the piecing together of dissociated knowledge will open up such terrifying vistas of reality, and of our frightful position therein, that we shall either go mad from the revelation or flee from the light into the peace and safety of a new Dark Age." --Dr. Carefree (talk) 17:48, 15 September 2008 (UTC)

Yes, it is that important. The question of the origin of everything has occupied mankind since humans learned to record their thoughts. We already have a multitude of incompatible answers from various religions. If it is a valid question to religion, it is certainly also a valid question for science. We do have a real need to know. Your comment reminds me of when horse-drawn carriages still outnumbered automobiles, when people feared that a person couldn't breathe if he traveled over 50 mph. Ignorance is no excuse for remaining willfully ignorant. ~Amatulić (talk) 20:24, 15 September 2008 (UTC)

We have an article on this. Among things, strangelet production will be less likely as collider energies increase apparently (ie. the LHC will be safer in this respect), there's a number of papers documenting why the blackhole production is not going to happen, and there's enormously powerful cosmic rays that smack into the Earth and other bodies in the solar system without collapsing the Universe.

You know we also can't say for certain that your posting this on the Wikipedia won't perturb the exact right part of the vacuum in the exact right way to trigger a transition to a new vacuum state or completely collapse the whole deal, right? 24.76.161.28 (talk) 00:00, 16 September 2008 (UTC)

To re-phrase your question: "Isn't it enough to know that gravity works? Why do we have to understand it?" Plasticup ^T/C 02:23, 16 September 2008 (UTC)

Dont get me wrong. I love science. But I'm talking practical science. Let's cure cancer science. Let's let math be the light that shines the way to truth science. We can study gravity, basic physics, and all that without too much trouble. But when we start messing with forces that's way beyond our grasp, especially when we don't need to, I think we're asking for trouble. I even like theoritical physics. Such as mathematical dimensions, black holes, strangelets, etc. as long as we dont create one. I'm not sure what a strangelet is (even after reading your article) but it doesn't sound friendly. This whole "Well it's pretty unlikely we'll create a black hole that wont go away, or some other wierd force that'll eat the planet. No guarantee mind you, but yeah, I wouldn't worry about it" attitude scares me. This is not child's play here. We're talking forces that we don't completely understand. Some of the math backs those Chicken Little scientists up, and I wouldn't dismiss them too easily. But they didn't build a multi billion dollar paperweight here, so I'm sure they're gonna use it. But before they start smashing atoms up(at 99.999% of the speed of light, yet! Yeah, that doesn't sound dangerous. Isn't something that moves at the speed of light supposed to have infinite mass? There's not much wiggle room here.) that were minding thier own business (being "atomy") they should consider the following:

• What if they create a God knows what particle that eats us from the inside out? And it drives us mad as it lays its particle "eggs" inside our helpless bodies and uses our nervous system for it's own agenda to propogate it's DNAless "species"? Plants, ameobas, and animals all have thier own agendas and are all radically different, so why couldn't this be the next stage of "life"?....if we let it.

• Or simply transforms matter into something mathematically indefinable that changes us into, uh....something we wouldn't like.

• What if it brings to this planet some creature, now freed from his atomic prison, something that's many levels beyond our worst nightmare!?

• Or God Himself! Think about that one. Now, wouldn't He be pissed? A:Because we yanked Him out of his bubble bath for an unscheduled visit. And B:Man outsmarted Him!

Sweet dreams.Dr. Carefree (talk) 06:15, 16 September 2008 (UTC)

This is basic science, not applied research, engineering, or "let's make a better mousetrap" work. I direct your attention to several previous science ref-desk discussions about what the goals are (in lay terms). As to your concerns, scientists who understand the possibilities have already analyzed the risks and it's been through a jillion reviews by experts. You can "what if" yourself into staying in bed with the drapes closed all day if you like, but you'll wind up pretty miserable. Every day you (every "you") do things that work for reasons you don't understand based on principles you haven't studied and you take for granted that a possible alternative "very bad thing" isn't going to happen:) I'd say most people may not completely (or even at all) understand what the LHC science is all about, but that's why it was reviewed by people who do. It's fine to be scared of the unknown, but let's not let ignorance (in the "don't know" not "you idiot!" sense) and fear drive society. Get concerned, but then do research enough to decide if there really is an actual problem, or ask someone who has. It's funny (in a sad way) that if popular press hadn't gotten ahold of "OMG they might make a black hole and swallow us all!" nobody would have cared...heck, even if they'd just used really complex scientific terms instead of scary Star Trek terms instead. Okay, I'll stop ranting now. DMacks (talk) 06:42, 16 September 2008 (UTC)

"Ugh not smash rocks together. Rocks may make fire. Fire burns. Maybe rock fire different from wood fire, rock fire burns rocks, burns everything! Ugh, Orgh and Ruma all die in quantum singularity! Grunt! Grunt!" --Stephan Schulz (talk) 07:10, 16 September 2008 (UTC)

Okay, that argument has been discussed to death elsewhere as well. [3] for instance (a CERN scientist arguing it). -- Consumed Crustacean (talk) 14:29, 16 September 2008 (UTC)

TURN IT AROUND. WHY would there be a problem with the LHC? Do you understand nuclear physics better than the guys running the LHC? What is so special about the LHC to you? --mboverload@ 06:57, 16 September 2008 (UTC)

You know, guys, feeding trolls just makes them hungrier. Saintrain (talk)

views of Thuria and Cluros

Moons of Barsoom. NASA/JPL

Astronomy on Mars doesn't say: Other than transits, have the hurtling moons of Barsoom ever been photographed from the ground? —Tamfang (talk) 07:11, 15 September 2008 (UTC)

That question sounds suspiciously ER Burroughs'ish. Anyway, the answer is yes: [4]. Franamax (talk) 08:02, 15 September 2008 (UTC)

Yeah, someone freecycled a big bag of ERB books my way; I'm on #4. —Tamfang (talk) 04:04, 16 September 2008 (UTC)

Hydrogen gas - analytical test???

Is there any analytical method to detect hydrogen gas? The pop test just tells qualitatively. can hydrogen gas be converted through cheical reaction so that it can be detected easily in lab scale.....? ....Venki —Preceding unsigned comment added by 218.248.46.116 (talk) 08:22, 15 September 2008 (UTC)

You could burn it like in the pop test, and find a way to get the mass of the water. From there, it's stoichiometry. It would be tricky, since the burning would vaporize the water, and would tend to shoot it all over the place. --Bennybp (talk) 11:30, 15 September 2008 (UTC)

de:Wasserstoff says that NMR spectroscopy is frequently used for that purpose in laboratories. --Ayacop (talk) 14:01, 15 September 2008 (UTC)

Burn a known amount it in a closed container, then all the water is still in there. That's the basis for elemental analysis, a standard way of determining "how much mass of each element is in this sample of stuff?". Also could use a mass spectrometer: add a known amount of some other gas and you will find the ratio of amounts of hydrogen vs that other gas. DMacks (talk) 18:41, 15 September 2008 (UTC)

One more - Use a calorimeter (something completely sealed like a bomb calorimeter). Given a heat of reaction, you should be able to tell how much burned by looking at the temperature. You have to make sure you don't have anything else flammable - just pure hydrogen (and oxygen, in excess). --Bennybp (talk) 00:06, 16 September 2008 (UTC)

Why Is Electromagnetic Wave Sine Wave?

There are many kind of waves such as saw wave, square wave, triangle wave etc. But why is electromagnetic wave always depicted as sine wave? How to prove that? - Justin545 (talk) 11:34, 15 September 2008 (UTC)

Good question. The reason is that circle functions like sinus are the solutions of differential equations that describe damped movement. I'm not able to say which equations and why but I remember that was the reason. --Ayacop (talk) 14:05, 15 September 2008 (UTC)

Undamped surely? Damped motion decays exponentially IIRC, but undamped motion, such as Simple harmonic motion, results in nice sinusoidal motion.

I would guess it's also related to the fact that, through Fourier Analysis, all of those other waves can be represented by adding sinusoidal waves of different frequencies. AlmostReadytoFly (talk) 14:16, 15 September 2008 (UTC)

PS: See Electromagnetic wave_equation#Solutions to the homogeneous electromagnetic wave equation. AlmostReadytoFly (talk) 14:22, 15 September 2008 (UTC)

A non-sine wave (a squarewave say) is the sum of a number of different sine waves at different frequencies. A square wave is the sum of a wave at some frequency ('f'), plus another at 3f, another at 5f, 7f, 9f...and so on. So if (for example) you had a light that emitted square waves, it might be emitting mostly red light - with some at three times that frequency, some more at five times and so on. Well, three times and five times the frequency of red light is ultraviolet light - and seven and nine times times red light is in the X-ray spectrum and so on. So you could in principle make square-wave "light" by carefully arranging an exact set of visible light, UV light and X-ray emitters. It wouldn't look much different from red light...except that it would irradiate your retinas pretty nastily!

It's certainly possible - but it doesn't tend to happen in nature because the processes that produce waves of such radically different frequencies are very different and their absorption in atmosphere is wildly different too.

SteveBaker (talk) 17:39, 15 September 2008 (UTC)

anything where the force is proportional to the negative of the displacement will give you a sine wave for motion, and since the derivatives and integrals of a sine wave are also sine waves, it ends up being sine waves all the way down. since that arrangement of force, or an approximation, is very common in a lot of physical systems, everything from a weight on a spring to an L-C oscillator ends up giving you sine waves; a pendulum is approximately close enough you get an almost sine wave, etc. Gzuckier (talk) 15:13, 19 September 2008 (UTC)

Hurricanes over warm water, meeting with freezing air over land, making snow?

Some of the rain from Hurricane Ike reached the middle of the U.S., and I recall one in Boston in 1991. So, I got to thinking during the windy remnants last night - suppose a hurricane or tropical storm hit in early November, while waters in the Atlantic Ocean were still warm, but while there was an unusually early cold snap and lows were below freezing over land. Would the foot of rain fall as maybe 10 feet or so of snow?

I'm guessing not, because after reading the article here, it seems like the water would have to be warm enough to warm the air around it to above freezintg - but that the interaction of warm and cold air would be such as to produce some *very* violent thunderstorms. However, I wanted to post the question here, too, to learn more and in case anyone else wondered.Somebody or his brother (talk) 12:17, 15 September 2008 (UTC)

See Halloween Blizzard ("the perfect storm"). Saintrain (talk) 14:35, 15 September 2008 (UTC)

LEDs or LASERs penetrating the body

Is there any LED or LASER that penetrates the body? —Preceding unsigned comment added by 117.196.163.244 (talk) 13:03, 15 September 2008 (UTC)

A pulse oximeter functions by shining a LED through the body (e.g. the fingertip). -- Coneslayer (talk) 13:10, 15 September 2008 (UTC)

Ophthalmologists use LASERs all the time to treat conditions of the lens and retina, both of which are truly internal structures. Dermatologists use LASERs for a variety of purposes, though the penetration is not very deep. Cardiologists use LASERs to remove endocardial plaque, though they do this using a catheter so most of the penetration is mechanical rather than optical. My guess is that neither of these is what you were looking for, though. ;-) --Scray (talk) 15:34, 15 September 2008 (UTC)

Carbon dioxide lasers can penetrate the body, but not in a good way. — DanielLC 16:46, 15 September 2008 (UTC)

Well actually:

They are also very useful in surgical procedures because water (which makes up most biological tissue) absorbs this frequency of light very well. Some examples of medical uses are laser surgery, skin resurfacing ("laser facelifts") (which essentially consist of burning the skin to promote collagen formation), and dermabrasion. Also, it could be used to treat certain skin conditions such as hirsuties papillaris genitalis by removing embarrassing or annoying bumps, podules, etc.

Nil Einne (talk) 17:06, 15 September 2008 (UTC)

Semiconductor diode

What happens to the current when you apply 0.7 volt to a semiconductor diode?

http://static.howstuffworks.com/gif/diode-graph.gif

The current reaches infinity or what?-Abhishek (talk) 13:24, 15 September 2008 (UTC)

The picture you provide is an ideal approximation. The actual curve of diode is non-linear and should be more or less like this

When 0.7 volt is applied, the current should be large but not infinity. And the overcurrent could break the diode. See Semiconductor diodes for more - 219.70.173.218 (talk) 13:52, 15 September 2008 (UTC)

Trains and Trucks running on gasoline instead of diesel

Would they run properly if they ran on gasoline engines instead of diesel? How would their performances compare? 98.221.85.188 (talk) 14:05, 15 September 2008 (UTC)

Would they run as-is? No. Diesel engines are not compatible with gasoline and vice versa. Could you put a gasoline engine in a train? You could, but diesels are significantly more fuel efficient than their gasoline counterparts. See our comparison of diesel to spark-ignition engines for more. — Lomn 14:49, 15 September 2008 (UTC)

As I stated in my original question, I was asking about trains and trucks running on gasoline engines, not gasoline in diesel engines. 98.221.85.188 (talk) 16:40, 15 September 2008 (UTC)

My mistake, but I hope you'll note that I answered that question as well. — Lomn 16:57, 15 September 2008 (UTC)

Gasoline locomotives would probably have more catastrophic fires and explosions due to fuel leaks or accidents than diesel locomotives. Such an event would not constitute "running properly," and would be a foreseeable consequence of a shift to gasoline. Edison (talk) 19:56, 15 September 2008 (UTC)

Nobody seems to consider that a gasoline powered car, equally or more prone to fire and explosion, 'doesn't run properly'. DJ Clayworth (talk) 03:45, 16 September 2008 (UTC)

Diesel has a flash point of >62°C (143°F), but gasoline has a flash point of <−40°C (−40°F). Most power boats under 35 feet have gas engines and most over 35 feet have diesel engines. Gasoline engines burn more fuel per hour than diesel engines of the same horsepower. Diesel engines are cheaper to operate, gasoline engines are cheaper to build, for the same horsepower. Economics of operation favor diesel, initial cost favors gasoline, safety favors diesel. [5]. In "Safety preparations for cruising" p71 it says "there is much less risk of fire and explosion" with diesel boat engines than gas engines. [6] It is hard to find that bald statement comparing gasoline and diesel locomotives, because no one uses gasoline engines in locomotives of railway size, in modern times, although it was considered in the 1920's [7] as an alternative to coal burning steam locomotives. I would expect that a diesel car is in fact less likely to burn up than a gasoline car, factoring in fuel spills while fueling or when fuel is spilled from a portable container, and fuel leaks from fuel lines, or fumes accumulating from spills/leaks in the garage.[8] Edison (talk) 15:52, 16 September 2008 (UTC)

If there are so many advantages to diesel, including overall cost in the long run, then why do most normal cars use gasoline (I'm aware of a few diesel cars). 98.221.85.188 (talk) 22:26, 16 September 2008 (UTC)

(unindent) As mentioned at Diesel engine#Diesel applications, historically diesel engines in the United States have not been popular as they "have been traditionally perceived as heavier, noisier, having performance characteristics which make them slower to accelerate, sootier, smellier, and of being more expensive than equivalent gasoline vehicles." Additionally until recently diesel fuel in the US was of a rather poor grade, with a high sulfur content. Old diesels, while easier to maintain, were also viewed as more finicky - some diesels had a problem with stopping/starting, especially in cold weather, so you had to leave them running - not a problem for industrial applications, but a nuisance for someone just stopping to get milk. And since there weren't a lot of diesels around, there weren't a lot of stations selling diesel fuel, so it was less convenient than gasoline. Of course, modern diesels escape many of those problems, driven mostly by improvements for use in European automobiles, where diesels have been historically popular. -- 128.104.112.147 (talk) 16:05, 17 September 2008 (UTC)

Statistics in research

Let's say you're using an instrument to detect the concentration of something, and you want the average and standard deviation of the measured concentration. You measure 5 times. Four of the times you measure a very low concentration, the fifth time you don't detect any. One the fifth time, it is not exactly as if the instrument is telling there is none, it's just that the concentration is below the level that the instrument can detect. When calculating average and standard deviation, should you A) treat this measurement as zero, B) exclude this measurement from the calculation, C) something else.

I realize this might make little practical difference in many cases, but I like to know what the right thing to do is, when being rigorous with the statistics and experimental method. Thx ike9898 (talk) 14:50, 15 September 2008 (UTC)

The most rigorous thing to do is to start again with an instrument that can detect lower concentrations. If that's not possible, the next best thing to do is to take further readings. Note that if you only have five readings, then there is little point in calculating the average and SD. You can and probably should just list all the readings. If you are taking a greater number of readings and you have the best possible instrument, then the answer to your question depends on whether you are sure you always know the difference between a "true" zero and a "below detection" zero. Itsmejudith (talk) 15:16, 15 September 2008 (UTC)

One common convention is to determine the lower limit of quantitation (LLQ), or the lower limit of detection (LLD), and then when none is quantitated or detected (depending on which you use) then the measurement is represented as a value half of the LLQ/LLD (on a graph the values below the LLQ/LLD can easily be distinguished). Because these values are not normally distributed, you should use a nonparametric test, like a Mann-Whitney or similar, to compare medians (rather than means and standard deviations). --Scray (talk) 15:40, 15 September 2008 (UTC)

You might find Detection limit useful as well. --Scray (talk) 15:45, 15 September 2008 (UTC)

I would not object to someone calculating the SD and AVG of 5 measurements. Consider that they might be making these 5 measurements for hundreds of different samples or conditions. There are ways of comparing the random variation within samples to the difference between conditions, such as Analysis of variance, and there are ways of designing the experiment to determine how many samples per condition are advisable. If I measured 2,3,1,2, and 0, I would see no basis for excluding the zero, unless you noticed the battery had failed at the time of the zero measurement or some other experimental blunder. The correct average of 5 measurements in my example would be 1.6, but if the fifth measurement of zero was excluded it would be 2.0, which would overstate the mean. The zero is not necessarily "artificial" or "incorrect." Let the data speak for itself.Edison (talk) 19:53, 15 September 2008 (UTC)

My concern is that inclusion of the zero artificially depresses the average. For example, the case where the lower limit of detection is 0.9 and the true value for the fifth measurement was 0.8.

I totally agree with your comment that there is no problem in reporting an average and standard deviation for 5 observations. This is common practice in some fields. ike9898 (talk) 21:07, 15 September 2008 (UTC)

Whereas an earlier response did make a generalization questioning the validity of calculating mean and SD for 5 observations, my point specifically was that the mean should not be used if the data are not known to be normally distributed. This latter point is an often-overlooked, very important assumption of such analysis. --Scray (talk) 23:50, 15 September 2008 (UTC)

I understand that many methods have underlying assumptions such as normal distribution, but that is a good point to bring up -- yes, many people forget this. ike9898 (talk) 01:41, 16 September 2008 (UTC)

To expand your hypothetical scenario a little, if your sample size was 100 and you got 99 positive results high above the detection threshold and one result of zero (i.e. below the detection threshold) then you could discard the datapoint as an outlier. Plasticup ^T/C 04:11, 16 September 2008 (UTC)

Sometimes experimenter doing ESP research have done selective retention of data and artificially produced significant results. But in careful lab work, such as at the US Bureau of Standards doing thousands of repetitive comparisons with official standards, there are some measurements found which are so far outside the distribution that they should not have been seen in millions of years. The trick is that a methodology of excluding outliers should be established in advanced and not adopted on the spot, since that might reflect experimenter bias and favor one hypothesis or another. Bogus positive results from areas such as ESP research have resulted from the experimenter excluding data he doesn't like. It is a poor lab practice for any experimenter to look at the data and decide on the spot which data points to exclude, without an a priori rule for excluding data, such as it being a certain number of standard deviations away from the mean. Otherwise leaving out certain measurements could be construed as fraud. Edison (talk) 15:32, 16 September 2008 (UTC)

Understood, but there are plenty of legitimate reasons to disregard outliers. Some estimators give undue weight to extreme values, possibly amplifying measurement error. Plasticup ^T/C 22:02, 16 September 2008 (UTC)

As Itsmejudith pointed out, you need different equipment. Common calibration numbers are for the middle of the measurement range. It would be extremely difficult to calibrate a machine for the conditions you describe. The values you are getting are probably inaccurate.

NIST has a very nice intro for Exploratory Data Analysis. Lots of graphical techniques. Saintrain (talk) 18:09, 16 September 2008 (UTC)

off the top of my head you have two ways to treat this; you can in fact treat the zero as a measurement, just as the other measurements. they all include measurement error, so does this. secondly, you can consider the measurement truncated at zero; i.e., if your instrument can't read below zero, then obviously that affects the relationship between the "true value" and the measurement error. (note that it's not guaranteed that the instrument will not read below zero, in which case you don't need to go with this option). there are nonparametric statistical tests which allow for this case, where there is either a maximum or minimum reading you can get, no matter if the true value is higher or lower than that. but to get the "average" in such a case, you have to get rid of the mean, and use the median, which is unaffected by this kind of truncation. Gzuckier (talk) 15:20, 19 September 2008 (UTC)

the flight of projectiles

If you were to depress the trigger of a standard modern day automatic weapon, an M16 on burst on an MP5 or something, I know Ican look up firing rates per minute but what would be the physical distance between the tip of one bullet and the tail of the one infront, assuming they all flew dead straight? And how does this vary compared to something extremely rrrrrapido like the gattling-minigun-type-things you can get on this thing found on the M167 VADSs or UH-60 helicopters? SGGH ^speak! 16:36, 15 September 2008 (UTC)

You need to know the rate of fire in rounds per minute (r), the muzzle velocity in m/s (v) and the length of each bullet in metres (l). The separation in metres (s) would then be $s={\frac {60v}{r}}-l$. Those numbers should all be easy enough to find - I expect the Wikipedia article for each weapon will give you them. --Tango (talk) 16:44, 15 September 2008 (UTC)

The gattling fires 100 a second, so thats 1/100 of a second between one bullet and the next, so the distance between the tip of one and the tip of the other would be the distance travelled in 1/100 of a second according to the muzzle velocity. Take the length of the bullet from that and you have the answer, right? SGGH ^speak! 16:58, 15 September 2008 (UTC)

For the M16 case, from AR-15 and 5.56 NATO I get r = 800 and v = 838. I can't find a WP:RS telling its length, but let's say l = 20mm = .02 m. So, $s={\frac {60*838}{800}}-.02=62.83m$ --Sean 17:24, 15 September 2008 (UTC)

That gives me, if the 20mm projectile is 125mm long (the longest of this typoe according to article) then that is 10.275 meters between each weapon. Does this sound reasonably? SGGH ^speak! 17:02, 15 September 2008 (UTC)

Actually that has to be correct. 100 in a second, a one second burst at 1km/s means 100 bullets spread across 1km, which is one every ten meters, so I guess that must be right. Sounds big though. SGGH ^speak! 17:07, 15 September 2008 (UTC)

Looks like you've calculated it correctly. It does look like a very large distance, but then 1 km/s is extremely fast! --Tango (talk) 17:18, 15 September 2008 (UTC)

If you ever saw one of those things firing tracer rounds - and remembering that typically one round in four or maybe one in six is a tracer round, I have no trouble at all believing the tracers are 40m to 60m apart - so 10m spacing for the bullets seems entirely reasonable. SteveBaker (talk) 01:17, 16 September 2008 (UTC)

New "unusual object" discovered

So:

"Hubble began seeing something brighten. It continued brightening for about 100 days and peaked at 21st magnitude in two near-infrared colors. It then faded away over a similar timescale, until nothing was left in view down to 26th magnitude. The object brightened and faded by a factor of at least 120, maybe more.

The mystery object did not behave like any known kind of supernova. It is not even in any detectable galaxy. "The shape of the light curve is inconsistent with microlensing," say the researchers. They recorded three spectra of it — and its spectrum, they write, "in addition to being inconsistent with all known supernova types, is not matched to any spectrum in the Sloan Digital Sky Survey database" of vast numbers of objects. "We suggest that the transient may be one of a new class."[9][10]

I, for one, am very excited. They've narrowed it down to being 120 light years to 11 billion light years away. Any thoughts? -- MacAddct  1984 ^{(talk &#149; contribs)} 16:56, 15 September 2008 (UTC)

Doesn't sound like the know much about it if they've 'narrowed' it down to between 120-11,000,000,000 light years away Nil Einne (talk) 17:01, 15 September 2008 (UTC)

If they successfully got a decent spectrum for it - they should be able to use redshift information to narrow the distance range down more tightly than that. If the spectrum is a crappy one for some reason - then how would they know that it didn't match anything? This is a bit suspect. SteveBaker (talk) 17:28, 15 September 2008 (UTC)

Plots of the spectrum are shown in the arXiv link above. The problem with getting a redshift is that you have to be able to identify the spectral features—they don't come with nametags saying "Lyα 1216" or whatever. In this case, the absorption lines are few (4–5) and broad (which gives your poor S/N in observed wavelength). Two hypotheses are tried in Section 3 of the paper; a z=0 hypothesis and a high-redshift hypothesis. In each case, there are extra unidentified lines and/or missing lines that you would expect to be there. If the source is at high redshift, you could also have absorption lines due to intervening systems at lower redshift, which further complicates the identification of spectral features. I can assure you that it's not amateur day on the author list or acknowledgments, so if they can't come up with a solid redshift, it's not due to either lack of effort or lack of competence. -- Coneslayer (talk) 17:45, 15 September 2008 (UTC)

At this point, this is an unexplained observation, not an unknown object. Which of these seems more likely, then: (i) a previously unknown object, or (ii) a known type of object with observations distorted by an unusual superposition of known phenomena? My nickel is on the latter because there are so many possibilities. Interesting either way, of course. --Scray (talk) 17:58, 15 September 2008 (UTC)

Redshift only gives you distance information if the object is moving with the rest of the universe. --Carnildo (talk) 21:31, 15 September 2008 (UTC)

Over large distances the fabric of the Universe is receding at more than the speed of light, which is a obviously much greater velocity than that of the object, so the object's motion becomes irrelevant. Plasticup ^T/C 22:44, 15 September 2008 (UTC)

If it's the exhaust plume of a nearby Bussard ramjet moving away from us at half the speed of light, the object's motion is quite relevant. --Carnildo (talk) 20:35, 16 September 2008 (UTC)

I know. I was just offering a counterexample to your claim that "redshift only gives you distance information if the object is moving with the rest of the universe" Plasticup ^T/C 21:59, 16 September 2008 (UTC)

Sounds like a lighthouse to me. Plasticup ^T/C 22:38, 15 September 2008 (UTC)

Are these comparisons acurate?

"The explanation for this [biological] diversity, the theory of evo- lution by natural selection, will form the backbone of your study of biological science, just as the theory of the covalent bond is the backbone of chemistry, or the theory of quan- tum mechanics is that of physics."

At first sight not all three "backbones" are really the backbones of these three sciences. What about other bonds - like the ionic bond - in the case of chemistry? And other physical theories like the theory or relativity in the case of physics? I also wouldn't say that the backbone of biology is the evolution theory. Isn't it the cell biochemistry? Mr.K. (talk) 18:23, 15 September 2008 (UTC)

Yeah, I think all 3 of those statements are rubbish. I'm not sure any of those fields have backbones, really, that's just not how science is structured. --Tango (talk) 18:42, 15 September 2008 (UTC)

Note that it is asking for the backbone of study, not the backbone of the discipline. For example, your study of ionic bonds could be based on an understanding of covalent bonds. In physics, quantum mechanics is rarely the backbone of study. Newton's laws are. But, the backbone of study changes from school to school and professor to professor. -- kainaw™ 18:47, 15 September 2008 (UTC)

Nothing in Biology Makes Sense Except in the Light of Evolution -- MacAddct  1984 ^{(talk &#149; contribs)} 18:54, 15 September 2008 (UTC)

I think the key phrase in that quote is "just as". It's probably fair to say that evolution is just as important to biology as quantum theory is to physics. That is to say - that it's right up there in the top five most important theories. But to pick a particular theory and call it number one is kinda silly. It depends on what you're trying to do. SteveBaker (talk) 01:12, 16 September 2008 (UTC)

I might rephrase it, "The theory of evolution is the underlying presupposition behind your study of biological sciences." Kristamaranatha (talk) 02:37, 16 September 2008 (UTC)

It's not a presupposition - there is plenty of evidence for it. SteveBaker (talk) 03:40, 16 September 2008 (UTC)

It is also hardly the underlying theory. Plenty of biology exists independent of evolution. Plasticup ^T/C 03:54, 16 September 2008 (UTC)

My point entirely - there is plenty of physics that's independent of quantum theory. So the importance of evolutionary theory to biology is comparable to (or "just as") the importance of quantum theory to physics...that is to say: It's incredibly important - but it's not the only thing. SteveBaker (talk) 15:08, 16 September 2008 (UTC)

I agree. I was replying to Kristamaranatha's "summary". Plasticup ^T/C 21:57, 16 September 2008 (UTC)

Czech Fingerprint Scientist

I recently came across a short note in a biology textbook about a Czech scientist who discovered the individuality of fingerprints or something like that. What was his name? I can't find anything about it on Wikipedia. Vltava 68 (talk contribs) 21:47, 15 September 2008 (UTC)

It was the anatomist Jan Evangelista Purkyně (but our article doesn't mention it!). Fribbler (talk) 22:30, 15 September 2008 (UTC)

I was about to say that! Also, the Chinese and Assyrians may have had a clue. Plasticup ^T/C 22:33, 15 September 2008 (UTC)

Demon's Egg

I wanted to ask about the scientific "accuracy" of this fantasy story. A bunch of British alchemists lead by Sir Issac Newton has "made a pact with the devil" and perfected a secret alchemic procedure for producing "Demon's egg".

The "Demon's egg" is a black spherical artifact which has the unusual property where

• It heats very quickly up when cooled

• It cools down when heated

• It grows when it is heated

• It shrinks when it is cooled

For practical usage, the egg must be kept in a brazier to keep it from shrinking and periodically it must be dunked in water to keep it from growing too large. It main usage is as a source of "instantaneous" heat. For example, it is used to bring a cauldron of cool water to boil quickly, by dunking the "Demon's egg" in it.

So is this an accurate description of an evaporating black hole?

How much energy can a 1 kilogram of "Demon's egg" hold?

What temperature must the brazier be to keep the 1 kilogram black hole from shrinking?

Thank you 202.147.44.80 (talk) 22:08, 15 September 2008 (UTC)

This is not an accurate description of an evaporating black hole. Black holes have no temperature. A 1 kilogram black hole would be so small as to be unnoticeable except as a source of hawking radiation. A black hole of any size would be impossible to hold up conventionally (though you probably could hold it up with magnetic levitation). The egg itself makes no sense. If it heats up when cooled and cools down when heated, that would just make it nearly impossible to heat or cool. It would effectively just keep everything around it at its neutral temperature, so I suppose if that's boiling then you could use it to boil stuff. You wouldn't need to do anything to keep it from growing to large or shrinking to small, as it would regulate its own temperature, and therefore size. — DanielLC 22:25, 15 September 2008 (UTC)

That's nonsensical. It loses heat when it gains heat? What does that mean? 24.76.161.28 (talk) 23:55, 15 September 2008 (UTC)

One weird property of black holes is that they have a negative specific heat: tossing more mass-energy into the hole reduces the Hawking temperature. "Heating" a black hole could mean adding heat (reducing the temperature), or "making it hotter" (increasing the temperature). So maybe if you were writing a riddle to which "black hole" was the answer, you could say that it "heats when cooled and cools when heated." But I would call that a deliberately misleading description, not an accurate one. For what it's worth, the temperature of a one-kilogram black hole would be about 10²³ degrees (Kelvin, Celsius or Fahrenheit) and it would radiate about 10³³ watts (about a million times the total power output of the Sun). (edit: But not for long, since 10³³ watts is about 10¹⁶ kg/second.) -- BenRG (talk) 00:26, 16 September 2008 (UTC)

It's really impossible - something that gains heat when you put it into a cool place is a straight up violation of the laws of thermodynamics - so no. SteveBaker (talk) 01:03, 16 September 2008 (UTC)

Assuming it doesn't contain a fuel source, of course. Or suck it's power from somewhere else. There are plenty of space heaters that will get hot when you put them in a cool place ... so long as their cord will reach. APL (talk) 03:03, 16 September 2008 (UTC)

Yes - that's true - but then we can't explain why it would cool down when heated...but at any rate, to (seemingly) cheat thermodynamics requires energy - and whatever internal source it had would soon run down. SteveBaker (talk) 15:04, 16 September 2008 (UTC)

Darn clever though. If this is an idea for a story you are writing don't let a little thing like physical impossibility put you off. DJ Clayworth (talk) 03:43, 16 September 2008 (UTC)

Yes - I adhere to the principle that any work of fiction is entitled to explore the consequences of ONE gross violation of known science...but no more. SteveBaker (talk) 15:04, 16 September 2008 (UTC)

importance of one engineering to the other

what are the relevance of engineering geology to the civil engineering industry?80.78.17.58 (talk) 22:38, 15 September 2008 (UTC)

The St. Francis Dam article may illuminate you. -- Finlay McWalter | Talk 22:45, 15 September 2008 (UTC)

Electrical grounding chemicals

while preparing an electrical earth certain chemical compounds are used as back fill material in the earth pit around earthing electrode which in turn increases the water retaintion capacity of the earth pit & increases the electrical conductivity of the earthing earth system.

I am interested in knowing the chemical composition of this back fill material & its chemistery. —Preceding unsigned comment added by 59.161.66.131 (talk) 23:53, 15 September 2008 (UTC)

see this link [11] —Preceding unsigned comment added by 79.76.201.117 (talk) 02:14, 16 September 2008 (UTC)

I question whether such chemicals are widely used in utility grounding systems, since the ground rods are typically simply driven into the ground to an adequate depth to obtain the desired low resistivity. There is typically no excavation, so no opportunity to che,ically amend the composition of the soil. If the grounding electrode is in a location where there is shallow earth over bedrock, then there might be excavation and backfill with chemical additives, but I have not encountered that. Edison (talk) 02:44, 16 September 2008 (UTC)

If you are doing it yourself, you can add salt and ashes to the ground. However this will increase corrosion of the grounding rod, as well as the conductivity of the ground. These add ions to the water, but if there is no water around will not conduct. Graeme Bartlett (talk) 21:26, 16 September 2008 (UTC)