Wikipedia:Reference desk/Archives/Science/2013 December 22

= December 22 =

Elephant flapping ears as a sign of aggression
I saw a GIF via Reddit wherein a person approaches an elephant that is flapping its ears and touches its trunk whereupon the elephant impacts the person, knocking them over. I would like to know whether the ear-flapping is a warning of aggression and whether it's a behaviour found elsewhere in the animal kingdom? --78.148.110.243 (talk) 14:19, 22 December 2013 (UTC)
 * This page from the Kruger National Park website has some appropriate information. If an elephant is startled, it will shake its head and flap its ears as a warning.  However, elephants also flap their ears to keep cool, so it's not always a sign of aggression.  See also animal communication for our general article. Tevildo (talk) 14:40, 22 December 2013 (UTC)

When have humans thought for the first time about a machine that can process information?
I don't mean a computer specifically, nor the name 'computer', but some sort of machine able to calculate independent of humans. Who was the first who came up with the idea? OsmanRF34 (talk) 18:10, 22 December 2013 (UTC)


 * The earliest known calculation device is probably an abacus, which the Sumerians used at the start of recorded history. -- Jayron  32  18:31, 22 December 2013 (UTC)


 * I wonder if the Antikythera Mechanism is relevant here? --TammyMoet (talk) 18:34, 22 December 2013 (UTC)


 * I had thought of that, but the abacus is much older. -- Jayron  32  18:43, 22 December 2013 (UTC)


 * The abacus doesn't fit the criteria of calculating "independent of humans" - A human has to perform every movement. Richerman    (talk) 19:03, 22 December 2013 (UTC)


 * Indeed, an abacus is a device in the same way as a paper notebook. OsmanRF34 (talk) 20:25, 22 December 2013 (UTC)


 * Hero of Alexandria made programmable devices about 70 AD, for instance a cart which carried automata on a path set by pegs on a rod. They also had automatic dispensers where you put in your money and it poured out some holy water. Dmcq (talk) 21:01, 22 December 2013 (UTC)


 * But a computer cannot operate independent of humans either. Humans program them... -- Jayron  32  04:03, 23 December 2013 (UTC)


 * ...and then they operate independent of humans. OsmanRF34 (talk) 18:49, 23 December 2013 (UTC)


 * Our article: Pascal's calculator has relevant information. ~E:71.20.250.51 (talk) 21:56, 22 December 2013 (UTC)


 * Until you pull the plug. ←Baseball Bugs What's up, Doc? carrots→ 06:48, 24 December 2013 (UTC)


 * The folks at the Computer History Museum in Mountain View spent a lot of time thinking about what Exhibit Zero ought to be. If you can't visit their showroom in person, here's the online version: they open up with the abacus and variations therein; written mathematics; the sector; slide rule; and so on.  (Plus, many more variants of pre-industrial and early-modern mechanical calculators!)
 * Personally, I think a critical knee-point in machine complexity occurred with the mechanical loom, circa the eighteenth century - weaving machines that could store the woven color-pattern and reproduce it on command. That's not what we call computation, but it seems to me that it's actually a very modern form of data processing - and arguably, more people use machines today for processing data (in the generic sense) than for computing numeric results.
 * No doubt, if we dig into details, we will find many more early inventions. Just like I described earlier this week, if we use weak and generic definitions for "data," "processing," and "calculation," then we can get some very strange answers.  Holding spaghetti, herding sheep, weaving thread - all of these are in some abstract sense "processing data" and may even be represented in Turing-complete mechanisms.
 * So, for the same reason, the folks at the computer history museum draw a sort of sharp line in the history section - once we get to "modern" computers of the 1940s and 1950s, the mechanisms and operations start looking very much more familiar and "computer-like". By the late 1960s, even the user-interfaces are nearly identical to modern-day machines: keyboards, visual graphical displays, touch-screens, natural-language interfaces; typed programs stored on nonvolatile memory; program compilers; host operating systems; networks.  All of that was pretty much in its present form by the late 1960s; from then forward, we've just been making the machines cheaper, faster, less buggy, more interconnected.  Oh, and hand-held/battery-powered!  Sometimes I forget this minor detail.  Nimur (talk) 00:52, 23 December 2013 (UTC)


 * If you're going with mechanical looms like the Jacquard loom, then you'd have to also go with things like music boxes and player pianos and the like; the first automated musical instrument (according the article music box) with interchangeable musical programs dates to 9th century Baghdad, and has all the same data processing and programmability as does the Jacquard loom some 1000 years earlier. -- Jayron  32  04:06, 23 December 2013 (UTC)


 * Excellent observation! I have to agree, Jayron - those machines have a stored program and a means for running it...
 * I think the one element missing from a music player is a conditional branch, though. More recent incarnations of mechanical looms could execute instructions conditionally, which (in tandem with its other features) make some of them Turing complete.  (But, on closer inspection, conditional instructions were not part of Jacquard's loom, circa 1800; which is why Babbage's design is so widely heralded as an important leap forward).   Nimur (talk) 04:20, 23 December 2013 (UTC)


 * I'd suspect the calendar and lunar cycle to be much earlier than any of those. When to plat, when to harvest most likely predates mechanical devices.  Many religions are based on lunar calendar which is very important for future climate prediction.  --DHeyward (talk) 07:25, 23 December 2013 (UTC)


 * Good point. Stonehenge and other structures like it were designed to tell people the day of the year, and thus when to plant, harvest, etc.  It didn't require any input from humans, after it's construction, and continued to "process information" from then on, with the input being the location of the Sun in the sky and the output being the current day of the year.  Of course, stone henges (and wood henges) also had a religious/spiritual meaning. StuRat (talk) 09:19, 23 December 2013 (UTC)


 * Stonehenge, seriously Stu? It is a theory, but a weak one. There are far easier methods of telling roughly what week it is. Planting and harvesting are not a date dependent processes, they vary according to climatic conditions. Clearly you are no gardener ;-) How would the dissipation of information work. Runners with messages, personal visits ("I'm going to see what day it is, I'll be back in five sunrises") or what method. I think the construction of the edifice does not allow for such fine detail. I'll accept the general premise of measuring time. Richard Avery (talk) 08:18, 24 December 2013 (UTC)


 * They had henges all over the place, and several remain today. Who knows how many there were.  The wood ones, particularly, weren't likely to last long.  So, there were probably plenty and they didn't have to travel far.  And I imagine there was lots of travel between villages anyway, for trade.


 * And the reason they were so big is that they also served a religious and political purpose. That's like asking why cathedrals are so big when they could fit as many seats in a far smaller building.


 * And I don't get what you mean by the date not mattering for planting. Of course the local climate determines which date is ideal for planting, but you certainly don't want to just say "well, it seems warm enough now, so let's plant", as it might well freeze again and kill off your crops, if it's too early in the season yet. StuRat (talk) 11:34, 24 December 2013 (UTC)


 * Like I said, you're no gardener. There are many natural signs like leaves and shoots appearing on trees, perennial plants shooting, the behaviour of the birds and animals, the height of the sun in the sky that all help decide when to plant. No gardener who is half experienced plants things according to dates, for the very reason you quote - the variability of the weather in temperate climes. This spring in the UK was a good example with a very cold snap with snow in May. I can't understand why people in neolithic times needed to know so frequently what day it was. Once you have popped down to your local henge to check the date you would keep count with notches on a stick or pebbles or something, wouldn't you? Richard Avery (talk) 14:06, 24 December 2013 (UTC)


 * You can't rely on when plants start to bloom, as they frequently get tricked by mother nature, bloom too soon, and then die when a late frost comes along. So, for each particular location, farmers would have determined the best date to plant a given crop, and use the nearest henge to determine when they hit that date.  And sure, they could count days using whatever method was handy, but counting out 365 days doesn't work so well, as without leap days the quarter day soon adds up and you end up planting at the wrong time.  They probably tried that method first, but eventually figured out it was no good.  And again, you're not considering the religious significance of things like this.  The solstices and equinoxes weren't just for farmers, they were considered to be part of the god's plans, and therefore a time for celebrations, etc. StuRat (talk) 14:32, 24 December 2013 (UTC)
 * ? It is worth re-reading Archaeoastronomy and Stonehenge, and some of the external references.  With Stonehenge in particular, it is very common that pseudoscientific fringe-theories are better-known to the public than actual facts and theories based on proper archaeological research.  If Stonehenge was a sundial, it wasn't a very good one; and if Stonehenge were an astronomical calculator, it was a very bad one.  The oft-mentioned, rarely-cited, "astronomical alignments" are actually quite difficult to specify.  If a "high degree of accuracy" is required - let's say, "±1 day," (about ±1 degree of arc), most of these alignments fall apart; and if we don't mandate this level of accuracy, then any arrangement of objects will align with almost anything.
 * And there's also the pesky problem that the stones have been moved; some fell over; and each time they get replaced according to a modern interpretation of antiquarian drawings. All these movements occurred long before the "astronomical archaelogy" trend in the 1960s, on which many of the "ancient astronomy" theories rest.  Nimur (talk) 17:19, 24 December 2013 (UTC)


 * ±1 day would be plenty accurate for planting. Heck, you could be several days off.  However, what you can't have is a quarter day each year which accumulates over decades, eventually putting your planting date weeks off.  So, some type of astronomical calendar gets the job done nicely, even if not particularly accurate.  Also, reversing your logic, if many of the stones have been moved, then perhaps it was originally far more accurate than it is now.  And finally, did they account for the precession of the equinoxes in determining what the accuracy was, when the henges were built ?  StuRat (talk) 16:55, 26 December 2013 (UTC)


 * You're missing Nimur's (and the linked article's) point. It is not that Stonehenge was insufficiently accurate to be used as a calendar calculator, which seems to be the argument you're disputing.  It is that if one allows enough slop in comparing Stonehenge's various sightlines – ±2 degrees in 1963's Stonehenge Decoded, for instance – to a large assortment of potential astronomical alignments, then the odds are actually pretty good (about even) that you'll get a large number of apparent 'hits' just by chance.  In other words, if you plunked down a Stonehenge-ish arrangement of stones at random in a field, you're as likely as not to produce an astronomical calculator that 'works' at least as well as Stonehenge, entirely by accident.  Absent either better quality alignments or independent archeological evidence, we're left with no pressing need to invoke the idea of an astronomical calculator to 'explain' the positioning of features at Stonehenge, and no need to conclude that Stonehenge's various builders were aware of the complicated astronomical features their construction could, in principle, be used to predict. TenOfAllTrades(talk) 18:30, 26 December 2013 (UTC)


 * If we break out the razor, the question then comes up as to what is the most reasonable assumption to make. Since we have many other archeological artifacts of the time which are indisputably astronomical calendars, and they would certainly have use for such a device, and it seems that the placements are consistent with such a calendar, albeit not a terribly accurate one, that seems a far more reasonable assumption to me than that they were just randomly placed for no apparent reason.  Considering the effort such a construction must have taken, it seems unlikely to me that they would do it for no reason.  Or, to put it another way, while "extraordinary claims require extraordinary evidence", the claim that Stonehenge is a calendar is in no way extraordinary. StuRat (talk) 18:41, 26 December 2013 (UTC)


 * That's sort of like saying that since some stone buildings from the Middle Ages are known to have been cathedrals, every stone building from that era – no matter how shoddy or distinct its construction, or how poorly it would serve as a place of worship – should reasonably be considered a cathedral. (I will leave aside the semantic argument about the distinction between a 'calendar' and a 'calculator'.)  For that matter, we should not assume more ignorance than necessary about the use and function of Stonehenge.  Centuries before the first stones were even raised, the site was in active use for ceremonial burial purposes; burials on the site continued after many of the stones were arranged.  Astronomical or calendar functions could have been incidental or even coincidental to the site's religious and ceremonial uses.  To go back to the example of cathedrals, virtually all French Gothic cathedrals are oriented such that their apse is able to directly face the point of sunrise on the horizon on a couple of days of the year: .  While this feature could in principle be used to mark a calendar, it is incidental to the building's function.  TenOfAllTrades(talk) 21:29, 26 December 2013 (UTC)


 * An, admittedly, tangential answer: Gregor Mendel 1866.  Genes contain a “stored program” and process information independent of humans.  Conditional branches (if Trisomie 21, then …) certainly exist.  Applying some fuzzy logics, one may even argue that gametes are Turing complete. Of course, you may not consider them to be biological "machines".   --Cookatoo.ergo.ZooM (talk) 10:41, 23 December 2013 (UTC)