Wikipedia:Reference desk/Archives/Science/2009 May 11

Science desk
< May 10	<< Apr | May | Jun >>	May 12 >

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.

May 11

Quanta

Reading A Brief History Of Time and hearing from other sources, it says that even observing a quantum particle changes it. What is all that about? Does it mean the light hitting it from an electron microscope changes it? Or is there some other mechanism involved? Also, how do they know this, if they didn't see what it was like before they observed it (logical to me)?--KageTora (영호 (影虎)) (talk) 00:32, 11 May 2009 (UTC)

Well, that might not be strictly true. If you set up an electron to have spin 1 on any axis and measure spin along the same axis, there should be no change in the state of the electron. It is of course a bit pointless to measure something that you already know, but it disproves the assertion nevertheless.

On the second question: Simply setup a repeatable experiment where you can make two measurements. Run the experiment 100 times and measure both things, then run it 100 times and measure only the second thing. If the results of the second measurement differ significantly between both batches, the first measurement had an effect. —Preceding unsigned comment added by 84.187.75.89 (talk) 01:50, 11 May 2009 (UTC)

I don't think that works. Even if you measure something you already know, your measurement still changes it. The spin after you measured it is now in an unknown state. Nimur (talk) 06:01, 11 May 2009 (UTC)

"Observing a quantum particle changes it" means that (some) measurements don't commute, in other words if you do measurement A and then measurement B you get a different outcome for B than if you'd done B first. Since it's a statistical effect you have to repeat the experiment many times (on identically prepared systems) in order to see it. So I think anon's description was exactly right. It's also true that measuring something you already know doesn't affect the system (at least, quantum mechanics doesn't say it has to). -- BenRG (talk) 12:29, 11 May 2009 (UTC)

(EC) The book is probabily talking about wave function collapse. Your interpretation of it as being the effect of the light on the particle being observed is a real phenomenon usually refered to as observer effect. These are two different things. The second one happens wheather you take quantum effects into consideration or not. It even happens in other situations that have nothing to do with physics. See for instance observer's paradox. The wave function collapse is a much more misterious phenomenon in which a particle which is in a state of superposition of several pure states suddenly jumps into one of those states when it is observed. To this day there is still considerable debate about what exactly happens when the wave function collapses. But there is no doubt whatsoever that it happens. As to the second part of your question, the particle being observed may have been carefully prepared so that they know beforehand what state it is in, and then when the experiment is performed it is verified that the particle indeed jumped into one of the possible states compatible with its original state and the experiment performed. Dauto (talk) 02:07, 11 May 2009 (UTC)

There's no doubt that the model with the collapsing wave function works, but a lot of doubt that particles suddenly jump into the measured state just as you measure them. The only thing you can determine experimentally is that the system was prepared to be in one state and later measured to be in another; what happened in between is a matter of interpretation. In path-integral quantum mechanics there isn't anything that looks like wave function collapse, and the spreading out looks symmetric in time—the maximum indeterminacy is midway between preparation and measurement. That matches the experimental reality better. If you have a particle emitter and a particle detector at opposite sides of a room, you can investigate what sort of changes to the room will affect the rate of particle detection (e.g. adding a barrier or a block of glass), and the answer (both in theory and in practice) is that the rate will only be affected if you put things where the path-integral formalism says the particle "is", not in the much larger area that the wave-function-collapse picture says it "is" just before detection.

The other reason to be suspicious of the picture with the spreading out and the collapse is that it looks like classical probability. There's always some uncertainty in the preparation of a system and that uncertainty increases with time (for example, an uncertainty in momentum leads to increasing uncertainty in position). But when you measure the position several times in quick succession you obviously should get roughly the same answer each time, no matter how much the position had previously "spread out" in your model. In order for the model to predict that correctly, you have to "collapse" the system's position in the model after the first measurement, and the mathematical process for doing that looks just like the mathematics of the quantum collapse (a Bayesian update). In the many-worlds formulation of quantum mechanics that's exactly what the collapse is, a Bayesian update with no objective effect on the world. The objective effect of measurement that people attribute to the collapse is actually caused by quantum decoherence.

In short, the situation is far too complicated and ambiguous to conclude that wave function collapse is physically real. Given how much it looks like a Bayesian update and how little it looks like other physical laws, I'd bet pretty strongly that it isn't. -- BenRG (talk) 12:29, 11 May 2009 (UTC)

Yes, I agree with that point of view. But that's still not a universally accepted interpretation of what happens at the wavefunction collapse. My statement intended to convey the fact that there is still dabate about that specific point. I think the term "Wave Function Collapse" is here to stay even if the decoherence interpretation eventually becomes universally accepted. Dauto (talk) 16:06, 11 May 2009 (UTC)

(Incidentally - an electron mictroscope uses electrons - not light - to image its target. Not that this changes the answers.) But basically, yes - there is no way to measure something without somehow affecting it. For large scale objects, the consequences are negligible - but for things like electrons, the consequences are significant - just one more weird thing that happens at the quantum level. SteveBaker (talk) 12:14, 11 May 2009 (UTC)

(Except that the rules do permit some measurements to leave the system unaffected—for example, a remeasurement of a property that you already know, but there are also less trivial examples.) -- BenRG (talk) 12:29, 11 May 2009 (UTC)

If you really want to enter into the "was the information really there before we measured it" rabbit hole, the most interesting places to start are EPR Paradox, Bell's theorem, and, somewhat differently, my favorite, Wheeler's delayed choice experiment. --140.247.10.164 (talk) 14:20, 11 May 2009 (UTC)

Clearly there is a high probability that Schrödinger's cat will eventually be mentioned somewhere in this thread. In fact, it was in a superposition of mentioned/not mentioned states - until I just mentioned it, which collapsed its wave function ... Gandalf61 (talk) 14:43, 11 May 2009 (UTC)

Hypocenters

I know that you can calculate the depth of a hypocenter by p waves, but is it possible to calculate the location of a hypocenter? —Preceding unsigned comment added by 174.6.144.211 (talk) 02:00, 11 May 2009 (UTC)

Yes. Dauto (talk) 02:23, 11 May 2009 (UTC)

how? —Preceding unsigned comment added by 174.6.144.211 (talk) 02:54, 11 May 2009 (UTC)

By comparing the time of arival of seismic waves at different seismographs around the world you can find out which seismographs were closer and which ones were further away and therefore reconstruct where and when the earthquake happened. Dauto (talk) 04:21, 11 May 2009 (UTC)

See hypocenter and particularly epicenter#Epicentral distance for an explanation. Astronaut (talk) 17:55, 11 May 2009 (UTC)

Chimera

In the Chimera dab page, somebody took out an entry for the wrong reason. My knowledge of genetics is pretty limited, so it's just possible that there is already an entry there that covers the same topic. The two items are:

• Chimera (protein), a hybrid protein made by splicing two genes (still there)
• Chimera (EST), a single cDNA sequence originating from two transcripts

Advice? Clarityfiend (talk) 03:26, 11 May 2009 (UTC)

I think you misunderstand the purpose of disambiguation pages. They are not intended to provide a comprehensive list of all of the meanings of a word - that's the job of a dictionary - and for that we have Wiktionary. The purpose is (as its name implies) to remove ambiguity when there are multiple articles with the same (or very similar) name. Since there is no article Chimera (EST), there is no ambiguity - so there does not need to be a DAB entry for it. Now, if you are arguing that there SHOULD be such an article - then go write one and add it to the DAB page. But until that happens, it's perfectly OK to remove the entry in the DAB. SteveBaker (talk) 12:06, 11 May 2009 (UTC)

A red main link is not a valid reason to delete a dab entry (see MOS:DABRL). Granted, the bluelinked article that went with it didn't refer to "chimera" as the guideline requires, but there are sufficient ghits both inside and outside of Wikipedia to make me think the topic is more than just a dictionary definition. If the two dab items refer to the same concept, there's no problem. However, if they are different, a related bluelinked article might be found or chimera might be mentioned in cDNA. Clarityfiend (talk) 21:22, 11 May 2009 (UTC)

I don't understand where you're coming from with that. MOS says: "A link to a non-existent article (a "red link") should only be included on a disambiguation page when an article (not just disambiguation pages) also includes that red link." - so is there an article out there (not a dab) that links to "Chimera (EST)"? If not, then the MOS clearly says that it shouldn't be in the DAB page. If I do a search on "Chimera (EST)" and click on "all pages that link to "Chimera (EST)"" in the search results - there is a talk page and two entries for the ref-desk. No articles. Hence the DAB entry most certainly should have been deleted per MOS:DABRL. What's to argue about? SteveBaker (talk) 12:10, 12 May 2009 (UTC)

What's the problem, Steve? I came here looking for an expert's opinion on whether this was something that should be salvaged. All I've gotten from you is a botched search and a flippant suggestion. Well, guess what? I've taken you up on that. Thanks for your "help". Clarityfiend (talk) 05:11, 13 May 2009 (UTC)

That was a rather rude reply. Steve was correct: there was nothing to link to, hence, no reason for it to be included on the DAB page. You've made an article now, so there's a good reason for it to be on the DAB page. Note also that Steve was talking about a search on Wikipedia itself, not a Google search, so your "botched search" comment is out of line. — The Hand That Feeds You:^Bite 13:04, 13 May 2009 (UTC)

If you'll notice, my search was restricted to Wikipedia, plus I had already specifically stated that I found "sufficient ghits" in Wikipedia, so I stand by my characterization. Steve apparently missed the whole point of what I was after, and I found the tone of his responses to be rather unworthy of him.

As it turns out, I appear to have gotten a bum steer from Spinningspark as well. After reading the other article and based on what I've learned, I conclude that it is the same concept, necessitating the blanking of my work in favor of the more complete article. A real fiasco all around. Clarityfiend (talk) 20:58, 13 May 2009 (UTC)

I don't know much about genetics either, but they are clearly two different things; one is protein and the other is DNA. Also expressed sequence tag is probably a better target for the blue link as this is the subject that gives rise to Chimera (EST). But as Steve says, the red link has no incoming links from other articles and none of the potential blue links discuss it. Having an entry on the dab page only misleads readers into thinking Wikipedia has information when there is none. SpinningSpark 22:16, 12 May 2009 (UTC)

"God probably doesn't exist"

Buses in London carried for some time ads endorsing atheism. They said - "God probably doesn't exist". Does it make sense to say "probably" in this case? Isn't this a case where we cannot apply any meaningful statistics? PS: this question is not about the existence of God, so don't hijack the topic.--Mr.K. (talk) 15:54, 11 May 2009 (UTC)

You can't apply frequency probability, you can apply Bayesian probability. --Tango (talk) 15:57, 11 May 2009 (UTC)

(after edit conflict) I think you can get quite close to a frequentist probability. Culturally, there have been many gods (Zeus, Ra, etc). If you're willing to accept that Zeus doesn't exist (as most people will) then you can simply count the number of gods, and count the number that you're quite sure don't exist. This gives you a low estimate of the probability that a given god doesn't exist. I'm not saying this is valid reasoning, but it is a way to apply frequency probability.

You can also generalize this to all instances of a particular type of claim, which I think is closer to what people actually do and explains the connection with the informal use of the word "probably" here. As you go through life, you encounter various types of claims, and for many you eventually decide whether they're true or not. For instance, you will encounter many spectacular claims for which there is no strong evidence, ie. claims that require faith. And you may be undecided on some and decide to reject others. This gives you a probability distribution over all such claims.

This, I think, is more or less why many people will dismiss things like ESP, horoscopes etc. offhand, without careful scientific investigation. A thorough investigation would cost far too much energy, so they rely on their internal probability model (ie. experience) to reject these things.

It is in this sense that atheists might say that God "probably" doesn't exist. Their experience tells them that claims of that nature often turn out to be untrue (by their personal concept of truth), and they judge subsequent claims by that experience, in this case because there is no way to achieve a scientific level of certainty.

I should also point out that in this case the odd use of the word "probably", is exactly what lends the phrase its humor. Much like Carlsberg tag line: "Probably the best beer in the world". risk (talk) 16:24, 11 May 2009 (UTC)

I think the example with Zeus and others is not appropiate. They were deities, but they were not omnipotentt beings, creators of the universe, etc. All the so called "gods" from Greek and Roman mythology were just like some superhumans. The current concept of God is that of an omnipotent force or being, who or what designed the laws of physics. I think the question is not about weather he/it exists, but how we interpret it. These advertisments however don't seem to address questions about neither philosophy nor probability, they seem to be just a provocation, just like gay pride parades.--131.188.30.106 (talk) 08:59, 12 May 2009 (UTC)

The poster's a nice piece of provocation, and is a response to a similar campaign from the opposite camp. Note that according to I think therefore I am, the bus doesn't exist either. --Dweller (talk) 16:06, 11 May 2009 (UTC)

Since when has me existing implied buses don't? --Tango (talk) 16:15, 11 May 2009 (UTC)

I disagree; the logic of cogito ergo sum merely implies that while I definitely exist, the bus might not exist. It makes no more concrete statement as to the degree of certainty of the existence of the bus; except that it is less definitely extant than I am. Nimur (talk) 18:19, 11 May 2009 (UTC)

Furthermore, I think the cogito ergo sum reasoning proposes several "equally likely" scenarios; one possibility is that the entire universe also exists, and is as it seems; another possibility is that none of it exists, but it appears as such because some entity (God?) is deceiving me; or alternatively, that I am deceiving myself (and that I am the entire universe, including God). My personal belief is that the first case is the most interesting one. This case must be true. If it isn't, then none of you exist, so your opinions of what is true or interesting are irrelevant; and I can single-handedly decided what is true and correct anyway. Nimur (talk) 18:23, 11 May 2009 (UTC)

<-Chaps, it was a joke, based on the fact that buses can't think. Never mind. --Dweller (talk) 20:30, 11 May 2009 (UTC)

My English could be failing me but isn't "God probably doesn't exist" actually a categorical statement that "God does not exist in a probable manner" rather than a dubious statement on his (her) existence? Isn't that a truism, since whether or not God exists it could only be in a definite manner? I seem to remember getting sentences starting "Hopefully, xyz" marked through in red at school for this reason....--BozMo talk 16:17, 11 May 2009 (UTC)

I think that would be "God doesn't exist probably"(note the lack of a comma), or better yet: "God doesn't exist probabilistically". I suppose particles in quantum mechanics can be said to exist probabilistically, so the case could be made. risk (talk) 16:29, 11 May 2009 (UTC)

Acording to the linguist Steve Pinker there's nothing wrong with saying "Hopefully, xyz". I agree with him. Dauto (talk) 17:43, 11 May 2009 (UTC)

I think he is at odds with Fowler's_Modern_English_Usage which has to be the higher authority. As you say he is only a linguist not a grammarian, although his book the language instinct is in my top ten books of all time.--BozMo talk 18:19, 11 May 2009 (UTC)

Which article in Fowler? (For those who don't know, it's notoriously ill-organized.) — All I remember of The Language Instinct, a few years after reading it, is vague disappointment. —Tamfang (talk) 01:22, 13 May 2009 (UTC)

There's an article, needless to say. --Sean 20:12, 11 May 2009 (UTC)

However the article seems to be entirely about US usage whereas we were discussing a London bus I thought. Misuse being widespread in the US isn't really relevant or unusual. It is however very difficult to care about it. --BozMo talk 21:03, 11 May 2009 (UTC)

The use of 'probably' is not in the mathematical sense, it is in the everyday sense of meaning 'most likely'. They're not saying they've done some stats and the probability says X, rather they are saying it is 'most likely' that god doesn't exist. Their reasoning will be based on a huge multitude of factors, not least a lot of what modern science has uncovered about the beginnings of the universe. Anyhoo the point is - this is more likely a advertiser looking for a 'catchy' comment than a statement of fact/reasoning. ny156uk (talk) 16:39, 11 May 2009 (UTC)

The bus ad uses a rhetorical device that can be called "amplification of slight doubt" that has a venerable tradition in an English operetta:

CAPTAIN: I am never known to quail

At the fury of a gale,

And I'm never, never sick at sea!

CHORUS: What, never?

CAPTAIN: No, never!

CHORUS: What, never?

CAPTAIN:"Hardly ever! Cuddlyable3 (talk) 17:06, 11 May 2009 (UTC)

The inclusion of "probably" was, I believe, required to allow the advertisement to meet the requirements of the ASA that absolute claims made must be verifiable. To omit it would require proof that God does not exist. Bazza (talk) 13:18, 12 May 2009 (UTC)

Actually in a radio interview the instigator of the ads said the word "probably" was included to be inclusionist about the atheists supporting it. Many were not "fundamentalist atheists", who contend that it is utterly inconceivable that God exists, but were more moderate atheists who had concluded that the existence of God was unlikely. (If you think about the other unsupported claims that are allowed in adverts then the ASA theory isn't likely). DJ Clayworth (talk) 17:22, 12 May 2009 (UTC)

It is worth noting that a complaint was made to the ASA and rejected. I don't know if the presence of the word "probably" was significant in that decision, but I doubt it. --Tango (talk) 17:48, 12 May 2009 (UTC)

In the UK it is actually illegal to place in an advert a statement that you cannot prove. Hence, the 'probably' in the Carlsberg Beer adverts.--KageTora (영호 (影虎)) (talk) 01:09, 13 May 2009 (UTC)

There is an exception made for things that are obviously subjective or obviously not intended to be taken seriously. "Joe Bloggs is ugly" cannot be proven, but I could say it in an advert because no reasonable person would think it was supposed to be a statement of fact. I can also exaggerate wildly as long as it is clear to any reasonable person that I am exaggerating. I think the God statement is acceptable under the former exception. --Tango (talk) 11:03, 13 May 2009 (UTC)

In any case the Carlsberg thing is less than convincing as it doesn't seem to address the issue. If they're banned from stating outright that they have the best beer because they can't prove it, well, how exactly do they plan to prove that they have "probably" the best beer? --Trovatore (talk) 20:20, 13 May 2009 (UTC)

Does half the voltage mean half the power?

Does the same lightbulb in a country with a 110 V standard burn half as bright as in a country with 220 V? And does the same go for a battery charger? Will it take twice as long to charge batteries with 110 V instead of 220 V? DirkvdM (talk) 19:03, 11 May 2009 (UTC)

Actually, if you assume that the lightbulb's resitence is Ohmic The power drawn at 110V should be four times smaller from the formula $P={\frac {V^{2}}{R}}$. In reality the bulb's resistence dependes quite a bit on its temperature which influences the result above. If the battery charger is designed for both 110V and 220V which is quite common nowadays, there should be no change on the time taken to fully charge the battery. Dauto (talk) 19:13, 11 May 2009 (UTC)

That power formula is for direct current. I believe all countries use alternating current. The power is based on voltage and current. Without knowing the current, it is impossible to make a judgement about power. See AC power. -- kainaw™ 19:16, 11 May 2009 (UTC)

The formula I presented works for both direct current and alternating current, provided that the load is an ohmic resistor. As I pointed out, that's not the case here. Dauto (talk) 21:26, 11 May 2009 (UTC)

(edit conflict with both above) A simple resistive load will draw 1/4 the power when connected to 110 V, as compared to 220 V; see Electric power. But neither of the devices you mention are simple resistive loads. The resistance of a light bulb's filament depends on its temperature, which (circularly) depends on the power it's radiating. See Lamp rerating for some guidelines. A battery charger like this one is a complicated microprocessor-based device, which is in no way a pure resistive load. The one I linked to is designed to accept either input voltage, and its power supply likely uses approximately the same power in either case (drawing twice the current at 110 V, compared to 220 V). It would be designed to apply the same charging current to the batteries in either case, and hence take the same time to charge them. -- Coneslayer (talk) 19:17, 11 May 2009 (UTC)

A light bulb filament, whether tungsten or carbon, is not an ohmic resistance. When the temperature changes, the resistance changes, quite dramatically. The source [1] indicates that a tungsten filament at half the rated voltage passes just under 2/3 the current it passed at full voltage. The resistance from the chart would be about 2000 ohms at full voltage and about 1530 ohms at half voltage, for the particular bulb referenced. (Milage may vary). Edison (talk) 02:54, 12 May 2009 (UTC)

What are the benefits of a tree structure?

The Transhumanist 20:22, 11 May 2009 (UTC)

A tree diagram can show more levels of hierarchy than other types of chart. NeonMerlin 21:04, 11 May 2009 (UTC)

What are the benefits of showing more levels of hierarchy? The Transhumanist 01:35, 13 May 2009 (UTC)

In computer software, it seems that a vast number of types of data are best stored in a tree-like structure (or one of it's generalizations or specializations). Why this is, is hard to say - but it pops up all over the place. Take, for example, a phone book. A long list of names with associated numbers. Real phone books are essentially just linear arrays - but when you store a phone book on the computer, it turns out to be handy to use a tree structure. The 'root' of the tree might have 26 branches - one for each letter of the alphabet. Each one of those has 26 branches - and so on for perhaps 100 levels corresponding to the longest name + address in the phone book. At the leaves of the tree, we place the phone number that corresponds to the name you spell out as you 'walk' along the tree to get to that leaf. Obviously, the leaf that you get to by following XQWZXGKWRTQ doesn't exist - so we 'prune' the tree to keep only the branches that actually lead somewhere. You'd obviously want branches for numbers and space characters and stuff - but the idea is basically there. Now, when you ask the computer to look up a phone number, it just follows the branches letter by letter and in just a couple of dozen choices, it finds the phone number. Compare that to the number of names you have to look at to find an entry in a paper phone book! Furthermore, when someone new has to be added to the phone book, it's easy to create the extra branches and tack on another leaf. Doing that with a simple linear array would require that you shuffle a large amount of data down the array to make a space for the new name. Furthermore, while you're doing that, the computer can't look up any more phone numbers...but with a tree, one part of the program can be adding or deleting entries while another part is looking up phone numbers. Tree structures are VERY cool. SteveBaker (talk) 02:13, 12 May 2009 (UTC)

OK, so we have:

1. Branching for efficient (faster) navigation
2. Scalability for ease of expansion
3. Modularity to enable multi-tasking

Humans using a phone book (a linear array) still have the benefit of "branching" because they can skip straight (or close enough) to the letter the name they are looking for starts with, etc.

If the phone directory is on rolodex cards (still a linear array), it's scalable, as the user can add cards anywhere in the collection.

Such a phone directory is modular, each index card being a separate module, allowing two people to use or work on the index at the same time. (E.g., one person can update the "M's" while you use the rest of the index).

What benefits are there to having a knowledge tree that is used directly by humans, say, on a word processor or on sheets of paper or in a published printed book? The example in the article on tree structure is an encyclopedia. Let's say that example was complete, with the whole encyclopedia rendered into a tree and published in a book or on a webpage. What would the benefits of that be?

The Transhumanist 01:34, 13 May 2009 (UTC)

Tree structures don't occur very often in the physical world - I'm having a hard time thinking of any, actually. A phone book with tabs for each of the letters A-Z is a kind of super-primitive one-level tree. The chapter list at the front of a book is somewhat like that. I guess in big technical documents, people sometimes do deep chapter/section/sub-section/sub-sub-section numbering - and then the index is a tree of sorts. But trees inside computers are typically dynamic, vibrant things - where data is added and removed easily and (often) frequently. I can't think of a physical analog of that. The lack of a physical analog of a tree structure means that these are not things that people are generally used to using. The file system on your PC is a tree structure - folders inside folders with files inside that - but aside from the initial structure of "My Documents/My Pictures", etc that is imposed upon you - most people have all of their files in gigantic, ugly flat lists. I don't think there necessarily IS a huge benefit for humans to deal directly with tree structures...in my mind they are the things of computers - and for that they are often ideal. Humans seem to prefer a structure like the Wiki or the World Wide Web where there is an arbitary mesh of connections going every which way. SteveBaker (talk) 02:22, 13 May 2009 (UTC)

"Tree structures don't occur very often in the physical world" - Steve, think again! Tree of Life, plain Tree, of course, nearly all river systems,... --Stephan Schulz (talk) 02:40, 13 May 2009 (UTC)

The tree of life is not a structure in the physical world - it's an artificial construct for storing data - and even that is pretty controversial because of the issue of genes crossing from one species to another. "The Tree of Life" should really be called "The Directed acyclic graph of Life"...but that doesn't trip off the tongue quite so well! Real trees are (of course) shaped like "tree structures" (although they are upside-down compared to what most computer geeks have in their heads) - but they don't organise information - and humans don't interact with them as tree-structures. We don't often start at the trunk of a real tree and follow the branching pattern to a specific leaf. I maintain that there is no physical analog of a computer's tree structure that humans interact with in that way. SteveBaker (talk) 14:11, 13 May 2009 (UTC)

Actually, there are more than one: Family trees, taxonomies, and organizational charts. The base structure of mind maps and concept maps are hierarchical. But the type of trees I'm most interested in are outlines. Such as Britannica's Outline of Knowledge (in its Propaedia, which includes over 700 outline sections). The Dewey Decimal Classification is a tree structure and de facto outline of knowledge, and so is the Library of Congress Classification. And of course Wikipedia's Outline of knowledge (which is likely to surpass the others in breadth and depth very soon), includes about 500 component outlines such as Outline of health, Outline of history, Outline of philosophy, Outline of robotics, Outline of Japan, Outline of Alabama, etc.). It will soon include hundreds more.

I've been having a hard time putting our outlines' benefits to readers over and above regular articles into words, but the input of editors such as yourself is helping a lot. Thank you.

The scalability and modularity benefits apply to articles in general, including outlines. But Wikipedia's outlines certainly take advantage of branching to a greater extent than regular articles, and therefore they are faster to use for navigation. They also show the structure of a subject, that is, they identify a subject's subtopics and show the relationships between those topics in a way that is easy and quick to view and comprehend. The relationships are metaphorically familial, that is, possessive (what belongs to what, like family ties). The relationships are shown through the arrangement of "parent", "offspring", and "sibling" topics. Articles do explain the relationships between their subtopics (sometimes), but an article's prose takes longer to read than an outline's hierarchy of topic names.

Is that description easy to understand? How can it be improved?

Any more observations about the benefits of tree structures, including outlines, would be most appreciated.

The Transhumanist 01:44, 14 May 2009 (UTC)

Why do these need to be separate articles though? For 'top level' topics like these, you can still have a conventional, readable article and put your 'sea of links' in a subsection. Take a look at two that I worked hard on: Computer has a large section at the bottom called "Computer#Further_topics" which tabulates practically every other Wikipedia page about computers into a handful of tables with links broken out into sections and subsections. I did the same thing for Automobile - and subsequently, the tables got moved out into separate template files what are shrunk down into one like "show" blocks in the "Automobile#See_also" section. These are tucked in to the bottom of the articles - which still read nicely - but which have shorter sections that summarize second tier articles that show up as links to more detailed articles: eg "Main article: Petrol engine". This means that when someone doesn't know that there is the concept of an "Outline" document (which 99% of readers won't - no matter how well you promote it) - they'll find the "main" article by typing in a really obvious search term like "Computer" - and arrive at a simple, easy-to-read article - which may be enough. If they find a section that interests them - then, right there, they have a link to a Main article that expands that short section into another full article. The "Further topics" section at the end contains the most links possible in the smallest space - but sorted into areas of interest. So if you're looking for articles about careers in computing-related fields, you'll find a table of "Computer-related professions" with a bunch of links that are split into hardware and software jobs. Another way to organize this vast amount of linkage is by packing them into a navigation box with expandable sections. I helped set one up that way for Video game industry - look at the box at the top-right of all video game industry-related articles and you'll find this standard box that cross links to every other article on the subject that we have. Other top-level articles such as Physics have taken a hybrid approach using both a top-right-corner "Nav Box" with expanding sections (just like Video game industry does) - AND a set of expandable sections at the bottom of the article (just like Automobile does). Notice that (for example) all of the 'top level' articles on the natural sciences have the same set of cross-links to other fields in that same expandable box format.

IMHO, any one of those three techniques is better than having a separate "Outline" article - which is harder to find and requires extra mouse clicks to navigate to. Better to pack that stuff into the bottom of every relevent article as an expandable section and cut out a bunch of unnecessary extra navigation. Also, packing the lists of bulleted links into a table allows for people to find the link they are looking for with much less scrolling up and down the page. If all you have is a list of links - then categorize them and pack them into a table in alphabetical order. SteveBaker (talk) 04:26, 14 May 2009 (UTC)

The benefits of a tree structure are twofold: First, it's easier for beginners. Most writing courses teach people to make an outline first, and it's a good way for them to start out. Second,, I tend to think and write that way myself--possibly due to some experience in intercollegiate debate. I always list things, mentally at least, before I write or speak about them. But there's two benefits of having separate the outline articles separate. First, so that different people can do it their own way. There's a corollary of NOT PAPER: we're a BIG encyclopedia. The second is that a pure tree structure with collapsable boxes hinders universal accessibility, or so I've been told, because reader programs cannot follow it. And I've learned this from my work as a librarian: however you organize things, someone will want them differently. They don't usually have the economic possibility of multiple methods--though even they try, because the subject headings are deliberately different from the classification on the shelves. DGG (talk) 15:18, 14 May 2009 (UTC)

Thank you for the feedback Steve, this is exactly what we need so that each design element is well considered.

Outlines are separate pages for a number of reasons. Outlines are integrated (connected) into a whole, which is harder to do (and confusing) if they are article subsections, to enable easy browsing of the tree of knowledge itself. Formatted tables are much more labor intensive (slower), both to build and maintain. Tree structures are simple and they are easily scalable.

To include lists in articles in the way you propose, would require rethinking and changing many of Wikipedia's fundamental guidelines.

"Further topics" is synonymous with "See also", and falls under the WP:SEEALSO guideline. Someone is bound to notice this and consolidate the two sections. See also sections are subject to moderation (i.e., being moderate), and they aren't intended to display redlinks. Links already included in the body of the article (in its text) are generally not repeated in "See also" (which means "related but not covered here"), making that section inappropriate for outlining the subject.

The division of Wikipedia into separate articles and stand-alone lists is a fundamental design feature of the encyclopedia. Embedded lists (including those at the end of an article) are intended to complement the prose of an article, not replace it. The purpose of articles is to explain. Therefore, lists of links (which are intended for browsing) should usually have their own entries. See WP:EMBED and WP:STAND.

Outlines tend to be long, and due to the size constraints of articles, "seas of links" (long lists) are not generally included in articles. See WP:SPLIT.

"Computer hardware" and "computer software" are major subtopics of "computer" and should therefore be covered in the Computer article. And because those subtopics are extensive, they are subject to Wikipedia:Summary style. Therefore, the hardware and software sections of the Computer article should be truncated versions of the articles of the same names.

I've done some experimentation with using navigation boxes in place of sublists in outlines. Those outlines aren't as convenient to use because you can't scroll down and "scan" all the contents. Clicking on show links to expand navigation templates is inconvenient and time-consuming, and disrupts scanning.

It takes one click to navigate to an outline from an article on the same subject. A subject that is chopped up into several collapsed navboxestakes several clicks to view.

Lack of awareness of outlines is a problem, but currently, it is a problem that is deliberate. To give the design team time and latitude in refining a core of the Outline of knowledge (and standards for its pages), we haven't heavily "promoted" the OOK nor integrated it extensively into the encyclopedia via links. To support the development of outlines, a guideline and explanation, and a projected model (showing future branches to be developed), are being drafted at this time. If we were to heavily promote or link to the pages of the OOK before the support pages were ready, we wouldn't be able to handle the volume of queries from confused editors. Chaos would reign.

To say that 99% of readers will never be aware of Wikipedia's outlines regardless of what we do, is sheer "can't be done" pessimism. They said the same thing about flying, and about sending a man to the moon. Sending readers to corresponding outline pages is much easier to accomplish. I'm confident our integration efforts will be successful.

Each of Wikipedia's navigation systems has strengths and weaknesses. But they also have a strength by coexisting: they're synergistic, playing an eternal game of leapfrog - in coverage and innovations. See WP:CLN for details.

The Transhumanist 19:51, 14 May 2009 (UTC)

The main benefit of a tree structure or hierarchy is that it's easy to conceptualize, because humans naturally group or divide things into sets, and hierarchies just do this in a recursive manner. It also has the benefit that it can be navigated with only constant working memory (you don't have to remember where you've already been to avoid going in loops). Directed acyclic graphs also share this property. Dcoetzee 03:35, 19 May 2009 (UTC)

Mutants leaving home to adapt

In what species, if any, have mutants been known to leave their home environments for places to which they were better adapted? NeonMerlin 21:03, 11 May 2009 (UTC)

Er, humans? David Vetter, for example. Rockpocket 21:31, 11 May 2009 (UTC)

Actually, Vetter was born in that environment, so perhaps Ted DeVita would be a better example. But there are lots of less dramatic examples of this: people with albinism moving to countries with less sunlight or people with arthritis moving to countries with a warmer climate. Rockpocket 21:41, 11 May 2009 (UTC)

Zoonoses must have done so at one point. Some species are rather polymorphic and thus may utilize different habitat niches. It's more likely that the "conquest" of a new habitat by a population subgroup with an adaptive trait takes place gradually; it would be hard to pass on your genes if you had no one to mate with in your new home. Also it may take some generations of selection for the organism to refine its adaptations. I'll try to find a specific non-human vertebrate example. In the meantime you may want to see Biological dispersal and rafting event. Sifaka ^talk 22:53, 11 May 2009 (UTC)

More a hybrid than a "mutant", but certainly different from your average polar bear: Grizzly–polar bear hybrids might have different habitat preferences than their parent species, but there is no information due to them being very rare. Sifaka ^talk 23:10, 11 May 2009 (UTC)

One of his descendants looks like a good bet [2]. Can't give you a species name, though. Since lots of scientists think that the first lifeforms were anaerobic the first bacteria taking steps toward becoming aerobic bacteria would probably also qualify. 71.236.24.129 (talk) 03:41, 12 May 2009 (UTC)

One of whose descendants? (more direct link) —Tamfang (talk) 01:41, 13 May 2009 (UTC)

Gogonasus or any other Tetrapodomorpha whose descendants left the water. One of them must have started it. It would probably be impossible to draw a line when exactly a species moved "to an environment to which it was better adapted". The transition will more likely have been gradual. (Think e.g. of a hippopotamus is it "better adapted" to land or water?) 71.236.24.129 (talk) 04:22, 18 May 2009 (UTC)