Wikipedia:Reference desk/Archives/Science/2011 September 18

From Wikipedia, the free encyclopedia
Jump to: navigation, search
Science desk
< September 17 << Aug | September | Oct >> September 19 >
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 18[edit]

radar system in electric trains[edit]

can RADAR system be installed in ELECTRIC TRAINS to detect the presence of another train in that track and any faults in the track? The key idea is to detect the presence of another train in the same track and find any faults in the track such that the automatic brakes are applied and the train is stopped automatically preventing accidents and any collision! is it feasible?if so how the system should be designed? plz provide a detailed explanation with circuit diagram. — Preceding unsigned comment added by Ajayfame.47 (talkcontribs) 01:40, 18 September 2011 (UTC)

In the US at least, I think there is already some system that detects where trains are on tracks and relays the info to a central headquarters for each line. There is also system for checking safety on tracks I think; at least on Amtrak's Northeast Corridor I believe there is. Sir William Matthew Flinders Petrie | Say Shalom! 02:20, 18 September 2011 (UTC)
Here in Spain they let a little bit current flow through the track to know that it's still there. Otherwise, trains run in a dedicated track, so there's no big chance of collision. Trains are also tracked through GPS, so a central tracking office always know where they are. Other, less predictable problems, however, like a truck on the track or some repairing service train, could trigger an accident. 88.8.79.204 (talk) 02:40, 18 September 2011 (UTC)
Here in the USA, all of the most important railroads are protected by track circuits. This is a brief description of how they work: A small low-voltage current is passed through the rails, which activates a sensor that keeps the lights green as long as the track is intact and unoccupied; when a train enters the track, it temporarily sets up a crowbar circuit for as long as it remains on that track, causing the current to bypass the sensor, which detects the loss of current and changes the light to red; if the track is broken, this results in an open circuit, which also interrupts the current and changes the light to red. I believe that in most other civilized countries (Canada, Europe, Russia, etc.) a similar system is in use. And yes, it's perfectly feasible to integrate this system with an automatic brake for the train and/or with a lighted diagram for the dispatcher's use, and/or with a light display in the engineer's cab (as is done on many American mainlines). 67.169.177.176 (talk) 03:01, 18 September 2011 (UTC)
So, would RADAR be suitable ? No. Both oncoming trains and broken tracks could be behind a hill or bend, and thus hidden from RADAR (in the case of broken tracks, even a slight rise or bend would hide them). And, even if the broken tracks weren't hidden, seeing the break in a RADAR image at that shallow of an angle in time to stop might be impossible.
As for stopping trains based on automated systems, you could expect many false alarms with such a setup. And, if the braking is severe, that may also injure or kill people on board the train. StuRat (talk) 02:58, 18 September 2011 (UTC)
See Automatic train stop and Automatic train control for more info. Yes, it is perfectly feasible to stop trains based on automated systems, as you will see in these 2 articles. As for concerns about braking being too severe, this danger can be avoided with proper system design and calibration. 67.169.177.176 (talk) 03:04, 18 September 2011 (UTC)
Railways generally use(d) physical infrastructure (ie track circuits mentioned above, signals etc) for train safety. The idea that radar could be used to detect a train ahead on the line is not realistic and not used. eg just think of curves, tunnels, hills and other obstacles that would prevent it working properly.
RADAR is used in trains - eg one use is to detect speed of a locomotive ('super creep speed control') The article about the safety system European Train Control System also mentions the use of radar as one of a set of methods of calculating position. It isn't clear but I would guess this is to detect speed (as in a Radar gun)83.100.239.6 (talk) 12:44, 18 September 2011 (UTC)
An radar based experimental obstacle (ie fallen tree not train) detection system is described in New Scientist no.3783 , 24 Feb 1964, p 489 A similar system was trialled on the Tokaido Shikansen line in the 1960s [1] [2] [3] (none of these links are very informative about whether or not it every became a permanent fixture)
If it was tried back in 1964 and isn't in widespread use yet, you can pretty much assume it was a failure. They rarely follow-up such stories with news of a failure. StuRat (talk) 19:29, 18 September 2011 (UTC)
One source said the system couldn't distinguish a bird sitting on the radar transmittor from a fallen tree further away - probably fail - though these things can go unnoticed too if they work and are not glamorous.Imgaril (talk) 23:42, 18 September 2011 (UTC)
Trains often are moving or standing still on adjacent tracks, and when a train came around a curve it would be very hard for radar to tell a train on the same track up ahead from a train on the adjacent track, since both would be directly ahead of the train with radar. GPS sent to a central office with a computer monitoring for conflicts is a better bet, as is the old block system long ago installed on railroads in many countries. Edison (talk) 19:39, 18 September 2011 (UTC)

Rocking back and forth: symptom of...?[edit]

Social science statistics - weighting outliers?[edit]

Hello

I have an introductory course to statistics this semester, and as such my knowledge of the mathematical aspects are limited. Therefore, I must stress that the following is a question about method, sooner than how to do this mathematically. I want to know if someone has thought this thought before, and devised a scientifically sound method.

Let us say we have 500 respondents in eg. the European Social Survey. A range of questions have been asked, but we're concerned with the dependent variable Happiness ratings, and have found a strong correlation to Income. In this example, we've checked for other variables and have found the correlation remains. However, there are statistical outliers: Perhaps 30-50 of our respondents are extremely un/happy, despite their income. Did they just make a joke of their survey? They are ruining what would otherwise be a perfectly linear relationship.

The question is: Is there a method for removing or weighing these respondents differently? In the ESS, if we look at these 50 respondents and the many many other questions they answered, does there exist a method whereby we assert their 'outlierness'? Meaning their propensity to answer on the far ends of the scales. I'd imagine one can look at a range of scales and judge how far they lie from the average on each - and assign a value accordingly. By doing so, in our example we may find that 10-20 individuals answered in jest; the rest, however, are 'genuine' outliers.

Thank you in advance for any help!

88.91.84.136 (talk) 12:29, 18 September 2011 (UTC)

I believe the usual standard is to discount outliers that lie outside of two standard deviations from the mean. I believe this is the 95% confidence interval. But it has been decades since I did this stuff... --Jayron32 13:00, 18 September 2011 (UTC)
If you want to automate outlier detection, you first need a model of your expected distribution. In my previous employment, I wrote complicated modeling programs to estimate state vectors for certain systems, and then optimized a distribution fit that minimzed the error of the curve-fit, ignoring error due to outliers. The theory of measuring error for multidimensional problems is very complicated, but most people in the social sciences will recognize at least the least squares method, which minimizes the sum of the squared error for each element. What this means in practice - for outlier detection - is that your distribution implicitly results in "soft edges" - a smooth transition from "normal" to "outlier." If you want to define other types of distributions with more sophisticated outlier rejection, you need to redefine your error function or your error measure. Nimur (talk) 15:00, 18 September 2011 (UTC)
The problem with just ignoring anyone more than two SDs from the mean is that you are ignoring the possibility that your distribution isn't actually normal at all (that's where the 95% comes from - it's the proportion of a normal distribution that is within 2 SDs of the mean - you need to go higher than 2 SDs, really, since you expect 1 in 20 people to be outside 2 SDs, so out of 500 people there should be 25 outside 2 SDs and they aren't outliers, they are just regular extreme values). If you have a very strong reason to believe that your distribution is normal (or approximately normal, perhaps by the central limit theorem), then sure, remove people that are a long way from the mean. If you don't have a very strong reason to believe it's normal, though, then you need to consider the possibility that they aren't outliers and that your distribution genuinely has a long tail. --Tango (talk) 15:28, 18 September 2011 (UTC)
But central limit theorem only applies for uniformly-random distributions! I can name dozens of scenarios in the physical sciences, and probably thousands of scenarios in the social sciences, where the data is not uniformly random. The answer to a political poll question, for example, shouldn't be uniformly random. As Tango points out, this means your distributions are complicated - possibly with multiple peaks - and so you need a more sophisticated classifier. If you treat these situations as bell-curves, you will (1) poorly fit your bell curve, (2) discard "outliers" that contained valuable data, and (3) include data points that fit the bell-curve, but do not fit to your desired model distribution. Nimur (talk) 16:38, 18 September 2011 (UTC)
The CLT applies to any distribution, not just uniform ones. If you are taking the average (or sum) of enough identically distributed, independant random variables, then it will be approximately normal, regardless of what the distribution is. That's why normal distributions come up so often. There are plenty of situations where normal distributions don't apply, of course, which is why you can't just blindly assume that anyone more than 3 SDs from the mean is a mistake. --Tango (talk) 13:11, 19 September 2011 (UTC)

This is a delicate issue: as soon as you start removing data, you raise the possibility of forcing the data to fit your hypothesis by removing the points you don't like. The two main precepts for avoiding that are (1) decide on your criteria before you collect the data, and (2) make sure that the criteria you use do not specifically select against data that don't fit your hypothesis. Even then it's always nicer if you can find a method of analysis that allows you to use all of the data, even if it weakens your findings a bit. Looie496 (talk) 16:40, 18 September 2011 (UTC)

Specifically to answer the OP's original question about social-science surveys: the standard practice is to discreetly hide questions amongst the surveys that should have "known answers" - or, questions that are intended to duplicate earlier questions. If someone is answering the survey questions in a joking way, or by selecting "choice A" for each question, these "easy questions" end up with improbable answers, and that entire survey can be discarded. This is similar to a captcha. I'm sure there are variants of this technique, and they probably have been widely researched. Nimur (talk) 16:47, 18 September 2011 (UTC)
Thanks, Nimur, for bringing it back to the OP's question. As far as I know, there are three ways to deal with outliers, one tactical (the one given by Nimur, to design the survey questions to expose them) and two mathematical. The first of these is the one given, to remove figures that fall more than 3 standard deviations away from the mean. This is hard if you are dealing with multivariate data, but not impossible. Also, if you know the data are normally distributed, there is no justification for removing outliers - contrary to Tango's comment, it is exactly when you don't know the distribution that you must consider removing outliers, which happens frequently. The other mathematical method is to use Robust statistics which are ways of removing outliers or downweighting them according to some predetermined algorithm. Perhaps one could call the "3 standard deviations" rule a kind of robust statistic, but it is usually done along with other tests, and requires a lot of inspection by the naked eye of graphical representations of data. The discipline of Robust statistics is really an attempt to automate this process, and achieve similar results. Hope this was useful. It's been emotional (talk) 06:49, 20 September 2011 (UTC)
Could you elaborate on your comment about normal distributions? If I know something is normally distributed, but I see points more than 3 SD from the mean (on a sample size that means there shouldn't be such extremes - obviously if you have a million data points, you expect a few to be more than 3 SDs from the mean and would need to either look at 4 or 5 or whatever SDs or count how many extreme values you have an compare that to the number you expect, although if you did the latter I'm not sure how you would deal with the surplus) then I know something has gone wrong with those points. If I don't know the distribution, then I don't know if something has gone wrong with those points or if they genuinely are that extreme. There are some distributions where you fully expect some points to be very extreme. It's those distributions where you shouldn't be removing points that look like outliers, since they aren't really outliers - they are consistent with the distribution. Consider the discrete distribution with 5% chance of being 0, 90% chance of being 50 and 5% chance of being 100. The mean is obviously 50. The standard deviation is just over 15. That means 10% of the distribution is more than 3 SD from the mean. It would obviously be wrong to remove 10% of your data points as errors in that situation. --Tango (talk) 18:00, 20 September 2011 (UTC)
Yes, you are correct from the point of view you describe - I wasn't clear about your terminology. To clarify what I said, the data come from some distribution or other, and if you repeatedly get extreme scores, more than you would with a normal distribution, then your distribution is by definition not normal, since the distribution is the thing that the actual data (not the "ideal" data) are taken from. At least that was the terminology in robust statistics, as I understood it, and I was under the impression it was orthodox across all of statistics. I may have been wrong. You are clearly assuming there is an underlying distribution which is the thing you want, with some measurement error superimposed upon it, creating the extreme scores. Then you are correct to reason as you do (as far as I understand it) but you must be quite certain you know the shape of the true distribution if you just label something measurement error. Admittedly, that happens often, or at least we often know a distribution is close enough to normal to call something an error. The situation I was discussing was effectively the same, but hypothetically if your actual data are normal, you will not have anything called an outlier, so removal of outliers assumes the actual data are not normal, but nearly so. If you have a totally different distribution, things change, but often the trick is to transform the data so you get a normal or nearly normal distribution, say by taking a logarithmic plot. As far as the OP is concerned, the techniques I gave above are the three in use that I am aware of, and the rest is a more technical discussion. Hope I've been clear, and thanks for your comment, It's been emotional (talk) 14:23, 21 September 2011 (UTC)

──────────────────────────────────────────────────────────────────────────────────────────────────── It's not clear to me why you need to remove the outliers. If you have scatter plot then for each value of the income you a probability distribution for the happiness, and unlike the mean, the maximum of this function is not strongly influenced by the outliers. By sampling the data and using extrapolation techniques, you should be able to quite accurately construct the curve that traces the maximum of the probability distribution. Count Iblis (talk) 19:07, 20 September 2011 (UTC)

If the outliers are extreme enough, they could influence the best-fit curve quite a bit. If you are trying to find the best straight line to fit your data points and there are few data points that clearly don't fit on the line, including them in your least-squares (or whatever method you are using) doesn't make much sense. You can draw one of two conclusions - either a straight line isn't appropriate or the data points are erroneous. You shouldn't try and draw a straight line through points that clearly aren't linearly correlated, since that doesn't work for either conclusion (if a straight line isn't appropriate, why are you drawing one? if the points are erroneous, why are you using them?). --Tango (talk) 19:33, 20 September 2011 (UTC)

physics of shrinking and growing physical objects and people[edit]

I don't know why my question was removed, I specifically marked it as non-troll. I would like to know the physics of shrinking and growing physical objects and people, i.e. whether it will ever be possible or already is, etc. 82.234.207.120 (talk) 13:05, 18 September 2011 (UTC)

A big fat NO, it is certainly not possible! At least not in the sense that you are talking about - the only way things can shrink is by becoming denser. The space between particles can decrease, but the particles themselves cannot. Plasmic Physics (talk) 13:34, 18 September 2011 (UTC)

I've warned this user about posting wasteful and nonsense messages here. --RA (talk) 13:45, 18 September 2011 (UTC)

Such blue-sky questions need not be wasteful. Why can't we shrink or expand an object? Because it's made up of atoms. Conservation of mass dictates we can't just make some of the atoms go away, even if there were some way to choose a "representative sample" of atoms in a person to make go away without harming them (which we also can't do). The actual size of the atoms is dictated by the atomic radius and such considerations, based ultimately on the effective size of the electron orbitals that form the basis of chemistry. The atoms need to contact one another by Van der Waals forces in order to stay in the same overall state (liquid, solid...) as they were, and overlap is strongly prohibited by the Pauli exclusion principle. Now getting down to brass tacks, what would we need to shrink or expand something? Well, barring those other possibilities, we'd have to change the size of the atoms themselves, i.e. change the Compton wavelength of the electron - the electron mass, or the base speed of light in vacuum, I think. Nobody has the slightest clue how to do this in a person-sized area, or how you'd keep such a fundamental change confined there. Wnt (talk) 14:31, 18 September 2011 (UTC)
It's not immeadiately obvious that you couldn't shrink a person by removing some of the atoms and leaving a representative sample behind. You would probably need to do it by removing entire cells, rather than trying to shrink cells. Within a cell, you can be at the point where individual molecules play vital roles. It's not usually the case that an individual cell plays a vital role within the body (the one exception I can think of is the brain - we don't really know how memory is stored, but there is a good chance individual cells are involved). So, apart from the brain, there is nothing at the fundamental physical level to stop you shrinking a perosn (although we obvious don't have anything close to the technology to do it). The big problem you would have if you tried to shrink someone by removing cells is that different things scale in different ways. If you halve someone's height, you quarter their surface area and you eighth their volume. Take bones, for instance. Their strength is roughly proportional to their cross sectional area. The weight they need to support is proportional to the volume of the body. That means that if you halve someone's height, their bones will end up twice as strong as they need to be to support the body. That may not sound like a problem (although it's clearly a problem for growing people - if you double someone's height, their bones will be half the strength they need to be and the person will collapse under their own weight), but it means the body is now massively inefficient. The body is a very delicate balance with everything doing just the amount it needs to and no more. If you mess with that, everything will go horribly wrong. Obviously, you can have small animals, but they aren't just scaled down versions of larger animals. --Tango (talk) 15:12, 18 September 2011 (UTC)
Hi, OP here. I'm not so much interested in the theoretics (obviously it is in some sense "possible", as there are animals that live on that scale...) I am interested technologically in what the state of the art is in this matter. I do not even know what word to search Wikipedia for, I can just describe the technology as I have right now... 82.234.207.120 (talk) 16:42, 18 September 2011 (UTC)
Animals live at many size scales because of their different evolutionary pasts, and ecological niches. We can't shrink individuals, but we can shrink species, given the desire and time. See e.g. the chihuahua or shetland pony. 108.15.146.137 (talk) 17:27, 18 September 2011 (UTC)
I'm sorry if I'm not making myself clear. I've trolled elsewhere, but this question is serious: whether it's fantasy or reality, what is the NAME of the technology in question? Should I just resort to guessing Greek and Latin roots? Surely, since this possibility has appeared at least in fiction, there is a name for what we're talking about. What is the technology in question, whether it exists, is theoretical, or is fantasy? I would like to look up more about it... Thank you! 82.234.207.120 (talk) 17:34, 18 September 2011 (UTC)
Shrink ray pretty much covers it. Franamax (talk) 18:24, 18 September 2011 (UTC)
Thanks. I never thought to check such a pedestrian name, which is as far from Greek or Latin as the article is from having any scientific content. I was hoping for something more... 82.234.207.120 (talk) 18:49, 18 September 2011 (UTC)
There is no actual or theoretical process for shrinking objects. It is purely an invention of fiction writers, and hence there is no science to speak of. Dragons flight (talk) 18:56, 18 September 2011 (UTC)
It is possible in classical physics, allowing you build computers that can solve non-computable problems see e.g. here. Count Iblis (talk) 18:59, 18 September 2011 (UTC)
Ah yes, the old-time ball-bearing computers, that takes me back. ;) Franamax (talk) 19:29, 18 September 2011 (UTC)
Objects grow and shrink all the fnording time: Thermal expansion. Hcobb (talk) 19:04, 18 September 2011 (UTC)
And, if objects are allowed to be mostly gas, then they can expand and contract quite a bit more. (Also, if they are allowed to take in or expel a fluid, like the puffer fish with water or air.) StuRat (talk) 19:25, 18 September 2011 (UTC)
That is not the same though, the OP is talking about shrinking the dimensions of an object on all scales. Essentially creating a situation where the object shares properties with a fractal structure, the dimensional proportionallity is constant at every magnification. This would require changing universal constants, which is unonceivable in this universe. Plasmic Physics (talk) 23:42, 18 September 2011 (UTC)
A Pantograph can be used to make smaller or larger copies of drawings and objects. Cuddlyable3 (talk) 17:06, 20 September 2011 (UTC)

Peanuts used as ingredient of Dynamite?[edit]

I've read somewhere that peanuts are used as an ingredient in dynamite, and i was wondering if this was true. Thanks. — Preceding unsigned comment added by RocketMaster (talkcontribs) 15:48, 18 September 2011 (UTC)

Peanut oil (as with other vegetable oils) contains triglyceride from which one could extract glycerol which could be used to make nitroglycerin which is the exciting constituent of dynamite. But there are lots of other sources of industrial glycerol, and I don't think there's anything special about peanuts per se. -- Finlay McWalterTalk 16:00, 18 September 2011 (UTC)
Yes. At first glance, this would seem unlikely, but as our dynamite article states, the explosive is made from "sawdust (or any other type of absorbent material) soaked in nitroglycerin". The article lists "powdered shells" as one of the alternatives to sawdust, so it would seem possible that peanut shells have been used. On the other hand, our peanut article states that "Peanuts have a variety of industrial end uses. Paint, varnish, lubricating oil, leather dressings, furniture polish, insecticides, and nitroglycerin are made from peanut oil". There doesn't seem to be a direct source for this, but our glycerol article states that glycerol is a used to manufacture nitroglycerine, and it would seem that, via saponification, it will be possible to produce glycerol from peanut oil. So, to sum up, I think the answer may well be that you could make dynamite using peanuts as a raw material, though whether they are normally used is another question. AndyTheGrump (talk) 16:12, 18 September 2011 (UTC)
Indeed, any vegetable oil could be used, sunflower, corn etc... A possible source appears to be Peanuts: the illustrious history of the goober pea, Andrew F. Smith, p102 - reading it out loud to you it appears that the USA used vegetable oils sourced in the 'far east' ie the philipines prior to WW2. (also happend during WW1 too same book p.69) During WW2 alternative sources had to be found due to japanese control of these areas - this turned out to be peanut oil. See also How peanuts helped to save the free world.Imgaril (talk) 16:30, 18 September 2011 (UTC)
One possible lead could be the work of George Washington Carver, who was a botanist known for working with, among other things, peanuts. He had supposedly catalogued several hundred uses for peanuts. --Jayron32 04:58, 19 September 2011 (UTC)

Corvid intelligence[edit]

Why are corvids (e.g. crows and magpies) so smart? --75.33.217.234 (talk) 17:23, 18 September 2011 (UTC)

Because they have smart brains. Dauto (talk) 17:26, 18 September 2011 (UTC)
What survival advantage do they gain that outweighs the extra energy that's required to maintain an improved brain? Why didn't other birds develop this? --75.33.217.234 (talk) 17:28, 18 September 2011 (UTC)
A better answer is because of their social structures, ecological niches and life history. Most corvids make their living extracting food from a wide variety of sources, and being able to survive in fairly wide range of environmental conditions. Having complex, adaptive cognitive abilities (learning, communication, tool making) is very beneficial for this life style. Post (EC): other birds have different lifestyles. Also, some other birds are quite 'smart', see e.g. african grey parrot.108.15.146.137 (talk) 17:36, 18 September 2011 (UTC)
Yes, that of course is the key issue. I suspect that it is probably related to their omnivorous feeding habits - they will exploit a wide range of food sources, and being able to locate and exploit resources that more specialist feeders can't may well require a higher level of intelligence (and probably, a better memory too - they may well exploit resources that are only available on a seasonal basis, and at limited locations, and would need to remember them over long periods). Many corvid species are strongly social too, which can often imply higher levels of intelligence. And yes, Parrots are likewise highly intelligent, social, and able to exploit a wide range of food sources. AndyTheGrump (talk) 17:40, 18 September 2011 (UTC)
By extension, would the push for vegetarianism perhaps be a conspiracy to "dumb down" the naturally omniverous human species? ←Baseball Bugs What's up, Doc? carrots→ 18:09, 18 September 2011 (UTC)
The pursuit of food is a pretty negligible portion of what most modern humans use their intellect on. I can go to the supermarket and get a billion choices with very little effort. In fact, arguably vegetarians are spending more time thinking about their food choices than most people, though I would still argue that it is a relatively small part of what they use intelligence for. Dragons flight (talk) 18:54, 18 September 2011 (UTC)
Funny, there's research on exactly this topic. Here's a small, on point sampling: [4] [5]. The primary indicators seem to be 1) highly social groups and 2) long developmental period before adulthood. They also seem to have vast spatial memories because they're hiding food when they're not dropping children off for their neighbors to raise. They're kind of the grifters of the bird world. Google scholar for "corvid intelligence" and you'll find a ton of research. Shadowjams (talk) 22:17, 18 September 2011 (UTC)
It's all about the High vocal center. (It seems Corvids are songbirds, by the way, or at least vocalistbirds.)  Card Zero  (talk) 08:53, 19 September 2011 (UTC)

Jurassic Park[edit]

Were any dinosaurs venomous (like the Dilophosaurus in Jurassic Park)? Whoop whoop pull up Bitching Betty | Averted crashes 18:35, 18 September 2011 (UTC)

Nobody really knows: venom glands do not fossilize. The living animals most closely related to dinosaurs are birds and crocodilians, and none of them are venomous. A group of scientists claimed to have fossil evidence that the dinosaur species Sinornithosaurus was venomous, but their arguments have been disputed (see the article). Looie496 (talk) 18:47, 18 September 2011 (UTC)
Yeah no one knows, but we can infer some things based on animals today. Smaller, more frail predatory animals tend to be venomous to help them assist in taking down prey. Larger animals have no need of such things as their sheer size, strength, fangs, claws are more than enough to take down prey. So it's pretty safe to assuming that a t.rex probably had no venom while smaller dinosaurs.... I don't know, it's possible but there's no reason to believe that they did. Birds are directly descended from theropod dinosaurs, and they can be quite small, however I don't believe any of them are venomous so it's prolly safe to assume that none of their ancestors did either. ScienceApe (talk) 23:17, 18 September 2011 (UTC)
Dilophosaurus wetherilli scale comparison.
Dilophosaurus, unlike the runty specimen in Jurassic Park, were also actually fairly large theropods. In contrast, the Velociraptor depicted there were waaay oversized.-- Obsidin Soul 08:32, 19 September 2011 (UTC)
The raptors in Jurassic Park were more like Deinonychus, weren't they? --Kurt Shaped Box (talk) 18:48, 19 September 2011 (UTC)
Naked ones with genius-level brains. LOL. Artistic license, I guess.-- Obsidin Soul 19:14, 19 September 2011 (UTC)
To be fair to Michael Crichton, it wasn't (widely?) known that raptoroid dinosaurs had feathers when the books were written. --Kurt Shaped Box (talk) 19:32, 19 September 2011 (UTC)
Our understanding of dinosaurs has changed a LOT since those days. The idea that birds are actually dinosaurs and that theropod dinosaurs were very similar to birds was virtually unknown in the public eye, and Jurassic Park helped changed that image in the eyes of the general populace. For that, you have to give Michael Crichton a lot of credit. ScienceApe (talk) 21:22, 19 September 2011 (UTC)
Oh don't get me wrong, I <3ed the book (s, if you include The Lost World). :P Though I covered my eyes through almost the entire film when I first saw it (long before I read the book) as a kid. LOL. Dad just found it hilarious. :( -- Obsidin Soul 22:47, 19 September 2011 (UTC)
There are a small number of bird species (e.g. the Hooded Pitohui) that can secrete poison in their skin and feathers for defence purposes - but they're not strictly 'venomous'. --Kurt Shaped Box (talk) 18:48, 19 September 2011 (UTC)
There is some evidence that Komodo dragons use venom (our article discusses it), so I don't think arguments of that sort are very compelling. Looie496 (talk) 23:41, 18 September 2011 (UTC)
No, I don't think one (debatable) example to the contrary really invalidates my arguments. Generally speaking we observe larger animals relying on physical attributes other than venom to bring down prey. ScienceApe (talk) 03:02, 19 September 2011 (UTC)
A BBC Radio 4 trailer I heard less than 2 hours ago, for a forthcoming episode of a current BBC TV series about dinosaurs, made reference to a flying or gliding dinosaur - non-avian but closely related to "true" birds - that was described as venomous: it may have been Sinornithosaurus as referenced above.
Whether or not this turns out to be accurate, I don't think that ScienceApe's argument above (in the topic's third post) is valid, because although extant birds are indeed descended from theropod dinosaurs (I would say they "are" therepod dinosaurs), they are only descended from one or at best a small number of species out of the very many other dinosaur species, theropod and otherwise, that existed. Their (alleged) lack of venom within the extant clade can therefore at the most only suggest that their particular ancestral species (or to be generous, genus) was probably non-venomous, without saying very much about their "non-ancestors." In practice I suspect that such a characteristic could both evolve and be lost multiple times in different lines over the couple of 100 million or so years that the scope of the OP's question. {The poster formerly known as 87.81.230.195} 90.193.78.49 (talk) 19:10, 19 September 2011 (UTC)
Sinornithosaurus was proposed to have been venomous, but this has been soundly debunked. It was based on blatant misinterpretation of the specimens (teeth out of socket, etc.). MMartyniuk (talk) 22:38, 19 September 2011 (UTC)
No, you misconstrued my post. I was using process of elimination. Basically what I was saying was that we can safely assume that none of the dinosaurs that birds evolved from had any venom. I wasn't talking about any of the other dinosaurs in regards to that argument, but again we have absolutely no evidence that any of them possessed any venom so there's no reason to believe any did. ScienceApe (talk) 23:40, 19 September 2011 (UTC)
Today there are poisonous reptiles, poisonous snakes, poisonous mammals, and at least one poisonous bird species. It seems rather absurd to suppose, a priori, that there would never had been a poisonous dinosaur, given that they were around for 160 million years, seemed to occupy a huge variety of ecological niches, had huge diversity in form (from the gigantic to the tiny). --Mr.98 (talk) 19:51, 19 September 2011 (UTC)
I never said there wasn't. I said there's no reason to believe there was because there's no evidence, nor any logical reason to believe that any did based on the rationale I gave before. ScienceApe (talk) 21:27, 19 September 2011 (UTC)
It's your logic that I'm disagreeing with rather explicitly. There is no reason to suspect that there wouldn't be at least one venomous dinosaur species. The rationale you gave is poor; dinosaurs came in all sizes and occupied all niches. The evidence is hard to parse on account of the fossil record; it doesn't prove the negative case whatsoever. --Mr.98 (talk) 22:30, 19 September 2011 (UTC)
It's possible, but there's no reason to think it's a certainty. It's a fun fantasy, but it's just that. You may as well make the same argument for, say, fully marine ornithischians. There needs to be some kind of evidence to make it science. To quote Futurama, "nothing is impossible if you can imagine it! That's what being a scientist is all about!" (he said, sarcastically). The bottom line is you need to realize there is a big difference between saying "there were never any poisonous dinosaurs" and "there is no evidence there were ever any poisonous dinosaurs." You can't prove a negative, but science is about disproving positives. MMartyniuk (talk) 22:35, 19 September 2011 (UTC)
Mr.98 you basically just pulled out the argument from ignorance, which is commonly used by theists in regards to claims that god exists. They say that "Yes I can't prove god exists, but you can't disprove it either", which is an absurd argument. You essentially did the same exact thing. Yes dinosaurs lived in all kinds of environments and came in all kinds of sizes, but there's absolutely no proof that any of them had venom. Does that mean that none of them were venomous? No, but there's no reason to believe any of them did either. I don't have to disprove something when there has not been a positive made in the first place. You need solid evidence in order to make that claim. ScienceApe (talk) 23:40, 19 September 2011 (UTC)

The short answer is: we don't know, but perhaps our research will yield evidence to support the hypothesis. Lots of diverse critters have evolved toxic bites, stingers or spurs such as insects, octopuses, stingrays, platypuses and snakes. In addition, even though birds are now rather benign towards us, they are not a good representative sample of what we should expect to find before the mass extinction occurred (most dinos were nothing like today's birds). Therefore, hypothetically, if I were to find some lively pint-size odd-looking but-cute dinosaur being playing about in the bushes or near me, I'd be cautious enough to treat it as if it were a lethal viper, for I would believe it to be possible. --Modocc (talk) 16:47, 20 September 2011 (UTC)

Was the format of the Voyager Golden Record's decipherability tested with humans?[edit]

After coming up with what to draw on the surface of it as well as the format to be used for encoding images in the analog grooves of the record, did they ever see if humans who were not "told the answers" could figure out what the drawings meant and could read the images off the record? Peter Michner (talk) 20:24, 18 September 2011 (UTC)

Humans would, of course, recognize the figures as humans, and would have a head-start in many other ways. However, a technologically advanced alien civilization might be better equipped to rig up a record-player than some isolated tribe of humans. StuRat (talk) 20:37, 18 September 2011 (UTC)
Actually, that particular record at the link provided doesn't seem to be the one with two naked Homo sapiens. I mainly want to know the factual yes-or-no whether they presented that record in the image to some other Homo sapiens prior to the launch and said, in so many words, "Can you decode this?" Peter Michner (talk) 20:42, 18 September 2011 (UTC)
The main problem with testing the golden record is that essentially any modern human is going to recognize that it is a phonograph, and immediately be able to make huge leaps and bounds based on that alone. I'm not sure how you could "test it on a human" for that reason: if you gave it to a human who you presume could work it out (a scientist, to be sure, not your average day laborer), they'd know quite a lot just by the appearance of it. More tricky would be the Arecibo message that StuRat refers to — even recognizing the human figure as human (which you wouldn't be able to do unless you figured out what the width/height of the signal was supposed to be) wouldn't really make the rest of it obvious to you without a lot of patient study. I think the idea behind both is that the alien equivalent of scientists would spend some time hemming and hawing over it, just as we did the Egyptian hieroglyphs. It's not expected that they'd be able to decode it in a day. --Mr.98 (talk) 23:00, 18 September 2011 (UTC)
I agree any success of a human doesn't say anything about aliens' success, but, simply, did they give a human a crack at it before they sent it up? That's all I want to know. I'm asking about the existence or non-existence of an event before a certain time in history. Peter Michner (talk) 23:50, 18 September 2011 (UTC)
I didn't recognize it was a record the first time I saw it; in fact, I didn't know what a record was until much later. I seem to remember that the answer is yes, and that the bright scientists it was presented to couldn't decipher it. However, the closest thing to a confirmation I found on Google was this highly sketchy website:http://www.thesupernaturalworld.com/forum/index.php?showtopic=3996
"'I've shown the plaque to lots of people, including some really bright students,' said Park. "They couldn't make heads or tails of it.'"
However, this refers to the Pioneers plaques, not the Golden Record. Sorry for not having a better answer; hopefully somebody will, because I'm also curious. --140.180.16.144 (talk) 03:33, 19 September 2011 (UTC)
That made me curious, can anyone understand this:
AB●C●●D●●●E●●●●F●●●●●G●●●●●●H●●●●●●●I●●●●●●●●J●●●●●●●●●
AKA BKB CKC EKE FKF GKG HKH IKI JKJ
ALBKB BLCKD CLDKF DLEKH ELFKJ
AMAKA BMBKB AMBKA BMCKB CMDKG
I'm using arbitrary symbols and the principles outlined in Carl Sagan's Contact, heh. -- Obsidin Soul 06:19, 19 September 2011 (UTC)
For the record I interpret A to M as... (see comment in Wiki source) but this relies heavily on our cultural conventions for writing such things. Even tiny changes would make it harder. Wnt (talk) 21:43, 20 September 2011 (UTC)
I actually thought of using something other than base-10 (for aliens which have 12 fingers, or aliens which use the number of their eyes to count, etc. LOL), but well... that was a quick test and I can't exactly transmit it with pulses of light etc., heh. And saw the error now.. eesh. In Sagan's novel the primer was also all in binary. -- Obsidin Soul 02:27, 21 September 2011 (UTC)
As for the message. I found an interview of Frank Drake that mentions him testing out the Arecibo message on Sagan. Quote:
I took [Carl Sagan] to lunch one day and presented him with the message already decoded. I asked, “Can you understand this?” This was a test to see if the message was understandable to a very knowledgeable Earth scientist. We eventually learned that nobody could interpret all of the message; each scientist only could interpret the part relevant to their discipline. So Carl got the numbers right and the planetary system right, but he didn’t get the DNA chemistry.
It's not the Golden Record though :/ -- Obsidin Soul 06:19, 19 September 2011 (UTC)
A-J are 0-9, K is =, L is + and M is *. That's just basic arithmatic, though, and I was expecting you to be doing basis arithmatic because I'm familiar with Sagan's work. It's not completely unreasonable to assume that a technologically advanced alien culture would also have the same expectation. Try explaining anything more complicated and it gets a lot harder. --Tango (talk) 18:13, 20 September 2011 (UTC)
Yay! Yep agree of course. But I'm happy I made contact. ;) Obsidian Soul phone home. ROFL. -- Obsidin Soul 02:27, 21 September 2011 (UTC)
The Golden Record was shown to humans before being put aboard Voyager. Many different humans were involved in designing it and deciding if it will provide any benefit to anyone who might find it. I believe there is confusion about what is on the Golden Record. On one side is primarily information about getting audio and video from the record. If you don't work with audio or video electronics, you won't get it. If you do, you will recognize the waveform representations and, possibly, figure out how to get audio and/or video from it. Once you do that, the record contains sounds and pictures of things around Earth. There is nothing much to decipher. You may not recognize which specific species of bird is making one of the bird calls on the record, but you'd recognize it as a bird. You may not know who one of the photos is of, but you'd recognize it as a photo of a human. -- kainaw 18:46, 20 September 2011 (UTC)

Genital cutting reduces/increases the risk of contracting HIV?[edit]

According to this, http://en.wikipedia.org/wiki/HIV#Sexual it says that male circumcision reduces the risk of contracting HIV, while female circumcision may increases the risk of a female contracting HIV. That doesn't make sense to me. Why would there be any correlation at all? ScienceApe (talk) 23:12, 18 September 2011 (UTC)

Male and female circumcision are two quite different procedures. Hence, no wonder that they produce different results. In the case of male circumcision, it reduces the possibility of infections and therefore it reduces the possibility of having more open ways for the HIV virus to enter the body. Quest09 (talk) 00:00, 19 September 2011 (UTC)
How does it reduce the possibility of infections? How does removing the clitoris increase the possibility of infection? ScienceApe (talk) 02:57, 19 September 2011 (UTC)
Unsurprisingling, the former is discussed extensively at Male circumcision#Sexually transmitted diseases and Circumcision and HIV. It doesn't exactly seem surprising there is some effect. Nil Einne (talk) 04:59, 19 September 2011 (UTC)
Doesn't say why though. Just that there's a correlation. ScienceApe (talk) 09:47, 19 September 2011 (UTC)
You're mistaken. Our article says:
Langerhans cells are part of the human immune system. Three studies identified high concentrations of Langerhans and other "HIV target" cells in the foreskin[97][98][99] and Szabo and Short suggested that the Langerhans cells in the foreskin may provide an entry point for viral infection.[28] McCoombe, Cameron, and Short also found that the keratin is thinnest on the foreskin and frenulum.[100] Van Howe, Cold and Storms criticised Szabo and Short's suggestion as "pure speculation".[101] Fleiss, Hodges and Van Howe had previously stated a belief that the prepuce has an immunological function.[101] Waskett criticised their specific hypothesis on technical grounds.[102] A study published in 2007 by de Witte and others said that langerin, produced by Langerhans cells, is a natural barrier to HIV-1 transmission by Langerhans cells.[103]
Dowsett (2007) questioned why it was just males that were being encouraged to circumcise: "Langerhans cells occur in the clitoris, the labia and in other parts of both male and female genitals, and no one is talking of removing these in the name of HIV prevention."[23]
and
A newer study, published in PLoS in January, 2010, points out that gross changes in the penis's microbiome occurs following circumcision, and this may play a role in protection from HIV and other sexually transmitted diseases
It's true it doesn't say precisely why there's a correlation, this shouldn't really be surprising since there isn't even unanimous consensus there is a correlation or more particularly to who and in what circumstances, how great it is and in particular, what the effect would be of greater circumscion and taking these, whether it should be recommended as a means to reduce the spread of HIV. So most studies concentrate on testing these hypothesis. However it seems clear no one is that surprised a correlation either way may exist since circumscion is a fairly significant change and you can come up with plenty of hypothesis on why the correlation is there either way.
Nil Einne (talk) 03:38, 20 September 2011 (UTC)
Yea, those two procedures have little to do with each other. Female "circumcision" is about preventing girls from having premarital sex. So, if it works, that should prevent disease transmission. Male circumcision is more about improving hygiene, by preventing the accumulation of smegma under the foreskin. Since that smegma could contain contagions after sex, eliminating it should reduce the infection rate. StuRat (talk) 05:23, 19 September 2011 (UTC)
What's stopping them from washing their penis? ScienceApe (talk) 09:45, 19 September 2011 (UTC)
Many people don't wash their hands after going to the toilet. What's the chance of them washing their penises? --Colapeninsula (talk) 10:38, 19 September 2011 (UTC)
Well then it seems like this study is rather deceptive because this would be an issue of proper hygiene reducing the risk of contracting HIV rather than circumcision reducing the risk of contracting HIV. ScienceApe (talk) 15:23, 19 September 2011 (UTC)
That's a bit like saying that obesity is really about poor food choices and has nothing to do with the availability of cheap junk food. Obviously, both play a role. StuRat (talk) 19:14, 19 September 2011 (UTC)
I don't think so. If the issue is about preventing the accumulation of smegma and improving hygiene then just cleaning your penis would accomplish the same thing. Not having foreskin just allows a lazy person to have approximately the same amount of protection as a man who cleans his penis. ScienceApe (talk) 21:05, 19 September 2011 (UTC)
Even if it were just about hygiene (and that seems unlikely), it is not necessarily "laziness" that prevents people in sub-Saharan Africa from meeting Western standards of good hygiene. Less than half of Africans have access to safe drinking water. If you are hauling a 50 pound water barrel on your back for miles from the river / well, and you have to boil it to drink it, then it isn't surprising that you wouldn't often want to spend much of it on washing yourself. A simple intervention that cuts HIV transmission by 50% in African populations is still a big deal regardless of the circumstances. Dragons flight (talk) 21:26, 19 September 2011 (UTC)
I'm not actually talking about Africa though. I'm talking about circumcision in general, that just so happens where they decided to do the study. The point is, the WHO is saying that circumcision is a good way to prevent the contraction of HIV when it's not because of that, it's because of good hygiene. If it's not just about hygiene, then I want to know what causes it. That's what the original question was asking. ScienceApe (talk) 22:32, 20 September 2011 (UTC)
The WHO makes policy wide considerations and aren't really there to provide individual advice. If you want that, I suggest you see your doctor. From a policy wide view point, circumcision may very well be a good way to prevent the contraction of HIV, regardless of the reasons (altho it's clear from the sources there is some question about how effective it is likely to be as a public health measure and whether the benefits always outweighs the costs and ethical issues). Finding out the reasons may provide other avenues for some reduction in some instances, it doesn't necessarily make the recommendation wrong particularly since we live in the real world not your magic world where people will always do what you say. I think it's clear from the sources many suggest other behaviour changes like greater condom usage and reduction of the number of concurrent sex partners will likely provide a more effective benefit but again because we don't live in this magic world, it isn't a magicly achievable goal. P.S. Note that is with others, I'm not suggestion hygiene improvement is the sole or even main reason for the reduction, I'm just pointing out even if it is your argument doesn't make sense. P.P.S. Note that because we live in the real world, you'll notice the sources often do discuss the risk of behavioural changes like a reduction in condom usage and other riskier behaviours that may arise from circumcision and a false sense of security even though there's clearly shouldn't be considered in your magic world. Nil Einne (talk) 23:43, 20 September 2011 (UTC)
Nil, you're clearly agitated. I recommend laying off of the personal attacks especially regarding "my magical world", be mature please. You are arguing a strawman when you accused me of wanting individual advice. Please don't do that, I never asked for that. You are now making the argument that if the issue were solely about hygiene improvement then my argument doesn't make sense. Actually yes it does, because if WHO recommends circumcision as a good method to reduce the incidence of contraction of HIV, then it's deceptive because circumcision doesn't reduce it. Proper hygiene does. The main problem with advocating this, is that circumcision advocates will use this as a reason to justify circumcision in males, when in fact it has to do with hygiene, not circumcision. Like I said, it's deceptive. ScienceApe (talk) 03:16, 21 September 2011 (UTC)
And a predictable percentage won't engage in proper hygiene, even where convenient, and can then infect others, even if those others do engage in proper hygiene. For another example, we could deliver contaminated drinking water in pipes and just say "well, they should know to boil it before they use it", but you could still predict a higher rate of disease than if it's properly sterilized by the water department.
Health authorities have known for a long time that just telling people what to do is insufficient to stop disease. If it were that simple, many contagious diseases would have been eliminated long ago. StuRat (talk) 22:04, 19 September 2011 (UTC)
That's not really a good example because they are providing a service to people. Sex is not a service the government is providing to people, it's something that they do on their own. If we were to use your analogy, then it would be more akin to the government providing prostitutes for people. If they were crazy enough to do this, they would at least test their prostitutes regularly, have a brothel or whatever that has health codes and enforce proper hygiene, require their customers to use condoms, etc. ScienceApe (talk) 22:32, 20 September 2011 (UTC)
Study Finds Circumcision Helps Prevent HIV and Other Infections says: "Families of anaerobic bacteria, which are unable to grow in the presence of oxygen, are abundant before circumcision but nearly disappear after the procedure. The researchers suspect that in uncircumcised men, these bacteria may provoke inflammation in the genitalia, thereby improving the chances that immune cells will be in the vicinity for HIV viruses to infect." Alansplodge (talk) 18:57, 19 September 2011 (UTC)
The article doesn't affirm that the bacteria causes inflammation. There's also aerobic bacteria on the penis. Where's the evidence that the anaerobic bacteria causes inflammation while the aerobic bacteria does not? It seems like another hygiene issue. A normal circumcised penis does not get inflamed unless there's an infection. That doesn't really make sense in any case. The penis becomes enlarged during sex anyway. ScienceApe (talk) 03:16, 21 September 2011 (UTC)
Morison, Scherf, Ekpo, Pain, West "The long-term reproductive health consequences of female genital cutting in rural Gambia: a community-based survey" Tropical Medicine and International Health 6:8 643-53 mentions a number of possible risk factors resulting from female circumcision. The labia normally has a protective function covering the vagina, and female circumcision often involves removing the labia, thereby exposing the vagina to infection, etc. There's a risk of scar tissue, keloid formation, cysts etc on vulva, which could tear or become inflamed during intercourse or childbirth. There's also a risk of damage to the urethra, perineum, or anal sphincter, which could result in fistulas or other complications. Anything that results in an open wound around the vagina will increase the likelihood of HIV infection. --Colapeninsula (talk) 10:48, 19 September 2011 (UTC)
I don't know. The labia has a protective function, but the foreskin does not? Wouldn't the labia also trap in vaginal fluids and such that may contain contagions? There's also scar tissue from circumcision as well. Seems suspect that most of the arguments given for why foreskin increases the risk of contraction of HIV, is reversed for the labia. Scar tissue, open wounds, complications are also ignored with male circumcision is regarded. Don't you think there might be some bias involved here? ScienceApe (talk) 21:16, 19 September 2011 (UTC)
The female reproductive tract is, of course, designed to retain "foreign fluids" (semen), so it's not going to make much difference one way or the other. The male reproductive system doesn't need to retain any foreign fluids at all, though, especially in an anaerobic environment. StuRat (talk) 23:25, 19 September 2011 (UTC)
It wouldn't? I don't really believe that. Opening up the vulva causes semen to drip out. I know, because I've seen it. ScienceApe (talk) 22:21, 20 September 2011 (UTC)
I think your missing the point. Firstly our article notes there may be an increased risk if men engage in sexual intercoure before full healing. More importantly, if there is indeed a correlation, there is a correlation whatever the reasons. There are plenty of reasons either way it could make a difference. Nil Einne (talk) 03:30, 20 September 2011 (UTC)
Actually I think you missed the point. I was asking what those reasons were, because I'm aware they are saying there is a correlation. Ok so then by insinuation, that risk associated with scar tissue with female circumcision should be alleviated after full healing has occurred as well. ScienceApe (talk) 22:21, 20 September 2011 (UTC)
I don't care what you're asking here as I've already provided other answers as have others. You're suggesting there is some 'bias' and that 'complications are also ignored'. You're missing the point that researchers are working back from the evidence and trying to determine the cause (but not that hard since it isn't that important to most of them at this stage) so the issue of 'bias' or 'complications are also ignored' doesn't come in to it. The researchers aren't there to prove something to the satisfaction of ScienceApe, particularly if SA isn't even going to completely read our articles (and the real sources would anyway be the peer reviewer articles not our likely inaccurate summary of them). Also no one said there is no risk after full healing, my point was simply that your statement the risk was ignored was ignored was clearly wrong. And you still don't seem to understand that the different nature of the procedures (and FGC isn't one procedure anyway, there is a large variety of procedures which go under that name) means the healing etc could be rather different. Nil Einne (talk) 23:41, 20 September 2011 (UTC)
You're getting angry now. You need to calm down. If you don't care, please don't participate in the discussion. You have not provided any answers that do not relate to hygiene, which is what I asked in the previous post I made. No, it seemed like there was bias in the reasons I've been given. A claim was made that female circumcision increases the risk of contraction of HIV because of scar tissue. However that would also be present in male circumcision so it doesn't make sense. A claim was made that foreskin traps fluid that may contain contagions, but so would the labia so it also doesn't make sense. Based on the answers I was given, things didn't make sense so I was questioning whether there may be bias in the rationale given. There might not be, but it's possible given that there is a history of people advocating circumcision for cultural reasons and then using bad science to give the illusion that their advocacy has practical benefit. Your personal attacks are really uncalled for. Please stop that or don't participate in this discussion. ScienceApe (talk) 03:08, 21 September 2011 (UTC)
The answers given thus far make sense. There appear to be many ways in which circumcision can affect the risk of HIV (both proven and hypothesised), including hygiene, scar tissue, Langerhans cell density etc. That different effects dominate in vastly different environments is completely reasonable, and is apparently backed up by the statistics. If you're looking for evidence of cultural bias influencing the science, you would be best off examining the papers that focus on the statistical links between HIV and circumcision. Your current approach of asking questions to which no one knows the answer (read carefully: this has been mentioned a few times) has not and will not get you far. Teshmanesh (talk) 10:36, 22 September 2011 (UTC)
No they don't because they contradict each other as I already stated. I'm not looking for evidence for cultural bias (that's a strawman logical fallacy), I just suspect it given the poor rationale given for why there is a correlation. Yes, no one knows the answer to my question, so what? There's nothing wrong with me asking questions, in fact more people should have been asking questions about this because the study is rather poor. The study wasn't carried out very well with proper controls, namely controlling for hygiene to ensure that it wasn't a factor. ScienceApe (talk) 11:55, 22 September 2011 (UTC)

I am not a scientist or doctor, but I can't help feeling that you are all missing one vital fact. HIV is contracted through blood or bodily fluid intermingling. With uncircumcised males the foreskin has a high potential for this form of transmission due to its delicate nature and high risk of damage during sexual acts. Ergo remove the foreskin and you remove a high proportion of the risk it posed. Simples yess!!Petebutt (talk) 23:40, 22 September 2011 (UTC)