Talk:Spectrum (disambiguation)

Disambiguation

The discussion that follows originated on User talk:Smack, then moved to User talk:Worldtraveller.

Hi Smack - I've seen that you've changed quite a few occurrences of spectrum to frequency spectrum. I think in all the articles I've noticed, actually electromagnetic spectrum is the better link. The electromagnetic spectrum is the distribution of energy emitted by an object in X-rays, visible light, radio waves etc. Frequency spectra are something quite different and not necessarily at all related, never really fully understood them myself! Worldtraveller 28 June 2005 08:12 (UTC)

Hi,

I do not agree that Electromagnetic spectrum is a better choice than frequency spectrum. I'll try to explain the distinction by analogy with chemistry.

The electromagnetic spectrum is an aspect of nature. It's a description of the different forms that electromagnetic radiation can take. The periodic table of elements is another aspect of nature, describing the different kinds of atoms that can form. Of course, not every substance has the same chemical composition. Air, for instance, is mostly nitrogen, with a good bit of oxygen, a smidge of argon and smaller quantities of other elements. A lump of coal is mostly carbon and hydrogen.

Similarly, a sample of radiation generally comprises only part of the electromagnetic spectrum. The sun radiates most strongly in the visible band. If we set our lump of coal on fire, it'll probably emit predominantly near-infrared rays. Radiation is characteristic to the object that emits it.

Granted, it's confusing to constantly refer people with weak scientific backgrounds to the rather tricky notion of frequency spectrum. I'd started to have second thoughts of my own even before you messaged me. In the future, I'll use energy spectrum instead. (Since the frequency and energy of photons are closely related, light can be broken down by energy just as well as by frequency.) It just never occurred to me to use the energy spectrum, as I had encountered it only briefly before, and it wasn't present in the disambiguation page.

I've watchlisted your talk page, so we can continue this discussion here.

--Smack (talk) 1 July 2005 03:38 (UTC)

Hi, thanks for your response. I'll try and explain in more detail why frequency spectrum was the wrong link for the articles I noticed it on - I think they were Cat's Eye Nebula and possibly Hubble Space Telescope as well as some others - and why I think electromagnetic spectrum is still the best link. My background is in astronomy, and it's quite possible that different areas of science use slightly differing terminology.

My understanding is that a frequency spectrum is not anything to do with the frequency of the radiation emitted by an object, but is a term used in discussing time-related phenomena. For example, you could plot the frequency spectrum of the variability of a cataclysmic variable or something like that, and doing a fourier transform gives you a power spectrum which can reveal periodicity within the variation.

Taking the Cat's Eye Nebula link, the article stated that it was the first planetary nebula whose electromagnetic spectrum was measured, by William Huggins. He definitely did not analyse the frequency spectrum, as that is not really a meaningful concept for a planetary nebula. He analysed only the optical spectrum, which is of course only part of the electromagnetic spectrum - either link would be accurate, while frequency spectrum would not be. I am not convinced that energy spectrum would be right either - that redirects to energy level, which gives the reader an explanation of the physics behind the emission of electromagnetic spectra rather than information on what spectra themselves actually are.

I hope that's explained my view on this reasonably well - if not I will try and clarify further, but certainly for Cat's Eye Nebula, 'electromagnetic spectrum' was what I meant when I originally wrote just 'spectrum'. Worldtraveller 1 July 2005 09:55 (UTC)

FWIW, as an ex-astronomer and ex-physicist, I agree with Worldtraveller: frequency spectrum and energy spectrum could refer to any time-dependent phenomenon - phonons, for example. If we are thinking about photons and the electromagnetic radiation then we should refer to the electromagnetic spectrum. -- ALoan (Talk) 1 July 2005 11:01 (UTC)

Worldtraveller is right about the use of frequency spectra to analyze recurring phenomena. However, theoretically, the same concept applies to the rising and falling electromagnetic field in light. Of course, it's impossible to actually measure the field, but as far as information theory is concerned, the only difference is that the frequency is higher. As regards the use of 'electromagnetic spectrum' with reference to a particular sample of light, neither Merriam-Webster nor Dictionary.com confirms the legitimacy of this usage, so I still maintain that 'energy spectrum' is the best expression to use. --Smack (talk) 3 July 2005 00:30 (UTC)

I'm really not sure I follow your logic here. To put it simply, William Huggins analysed electromagnetic radiation from the Cat's Eye Nebula, and to study it, he dispersed it into a spectrum by means of a grating. So, he analysed the electromagnetic spectrum (but only a small part of it, of course - the optical spectrum). A frequency spectrum, in common scientific usage, is something quite different to what Huggins analysed. And energy spectrum redirects to energy level: transitions between the energy levels of the ions and atoms in the nebula give rise to the emission of electromagnetic radiation, but that's one step removed from what Huggins studied - knowledge of the structure of the atom was still decades away when he took his spectra. So, frequency spectrum and energy spectrum are both inaccurate terms to descibe what Huggins studied. Worldtraveller 3 July 2005 00:53 (UTC)

1. Common scientific usage is not a guide on this question. In common scientific usage, the notion that we're talking about is simply called a 'spectrum'. Since Wikipedia structure disallows us to use this general term, we're left with the task of finding a more-specific term that matches the concept as closely as possible.

2. The redirect at Energy spectrum is a big mistake. As soon as I have time, I'll put a stub there. Neither Webster nor Dictionary.com has this term, but Google confirms that it can be used in this way. --Smack (talk) 3 July 2005 03:06 (UTC)

I agree the redirect to energy level is not very accurate. But I still don't understand what your objection is to disambiguating, at least in the case of the Cat's Eye example, to electromagnetic spectrum? Worldtraveller 3 July 2005 07:47 (UTC)

Because, unless you can convince me otherwise, 'electromagnetic spectrum' is not a property of an object. We can speak of the E. S. as the set of all possible electromagnetic radiation, but we cannot speak of an E. S. associated with a specific body.

If you've been paying close attention, you should have noticed an inconsistency in my argument: first I cited dictionaries to support my case, but then I rejected common scientific usage. So now I have to come up with a new basis. The electromagnetic spectrum and the generalized "spectrum" that we're talking about here are two different things. If we try to put this topic at the E. S. page, we'll be faced with a new question of disambiguation. --Smack (talk) 3 July 2005 16:20 (UTC)

We can and do speak of the electromagnetic spectrum of an object. Astronomers often talk about a 'spectrum' without a qualifying adjective, but we implicitly mean an electromagnetic spectrum; it's so common a usage that it's not often actually spelt out. However, here's some examples from the literature of papers using the expression 'electromagnetic spectrum of...': [1], [2], [3], [4], [5], [6]. The electromagnetic spectrum can refer to the whole range of radiation from radio to gamma ray wavelengths, but there is no more accurate term for the dispersed spectrum of electromagnetic radiation emitted by an object than its electromagnetic spectrum. Worldtraveller 3 July 2005 22:36 (UTC)

By the way, in the second abstract I've linked to above, mention is made of the energy spectrum of electrons emitted by the pulsar in the Crab Nebula - this shows that energy spectrum refers to something quite distinct from the electromagnetic spectrum. Worldtraveller 4 July 2005 10:56 (UTC)

That will teach me to shoot my mouth off without checking the context properly. If we are talking about an analysis of the amount of electromagnetic radiation that is emitted or absorbed by the object as a function of frequency (or, equivalently, as a function of wavelength or of energy), as in the second paragraph of the lead section in Cat's Eye Nebula, I think it is clearer to refer to spectroscopy (since that kind of analysis is precisely what spectroscopy is).

Actually, in some of the articles that Smack has done some disambig on, frequency spectrum is more appropriate than electromagnetic spectrum and better than just spectrum which is a {{disambig}} page. However, a better article in general is spectral density, since it actually says in words what it means, and the mathematics is that usually associated with Fourier analysis and the like which is what is usually meant by 'spectrum'. I've been wondering if we should merge the relevant parts of frequency spectrum and spectral density somehow? Also, in several places, you've linked to spectrum (disambiguation) when the relevant one from that page might be better (e.g. in spectral radius, a link to spectrum of a matrix would be much better). On the other hand, there might be some where the disambig page is better than a questionable link to a a particular spectrum type (no particular e.g. in mind here).

I feel I should be WP:BOLD and do this myself, but I don't want you to think that I'm wikistalking you or something! Maybe we can work together to decide on appropriate disambigs on things one of us is unsure of? — a good example might be sound pressure level, where you've redlinked spectral response but I think absorption spectrum might be better but am unsure. -Splash July 3, 2005 22:39 (UTC)

A seperate comment: I'm not sure about amending people's comments on talk pages; that's for them to do, really, I tend to think. -Splash July 3, 2005 22:39 (UTC)

I think one would generally refer to a "spectrum" in this kind of situation, without making it entirely clear whether that word was shorthand for an "electromagnetic spectrum" or an "energy spectrum" or a "frequency spectrum". Using the term "electromagnetic spectrum" in this context does some rather odd - you would usually think of the electromagnetic spectrum (radio waves to gamma rays) rather than an electromagnetic spectrum (the electromagnetic radiation emitted by a particular astronomical object). On the other hand, "energy spectrum" or a "frequency spectrum" also look a bit odd and potentially misleading. Would "spectroscopy" be a suitable unambiguous alternative? -- ALoan (Talk) 4 July 2005 14:53 (UTC)

To Worldtraveller: Point taken. Now tell me, don't we need to disambiguate somehow between the general electromagnetic spectrum and particular electromagnetic spectra?

To ALoan: No, it wouldn't. Spectroscopy is a branch of science. It's a set of theoretical and practical devices for the study of (drumroll) electromagnetic spectra. Substituting 'spectroscopy' for some kind of 'spectrum' is like substituting 'carpentry' for 'table'. --Smack (talk) 4 July 2005 21:03 (UTC)

Ok, I'm going to jump in here, with a number of points. Excuse my verbosity in advance:

• The article at frequency spectrum is rather peculiar. What it's really talking about is the discrete fourier transform, and this is the bulk of the mathematics it gives. It ought to be merged there somehow. The usual meaning of frequency spectrum is "how much power there is at each frequency in a signal" i.e. the spectral density of the signal, or the power spectrum. I use signal in a loose sense to mean a "time-varying quantity", but my background is communications, so I often mean literal signals. The frequency spectrum in this sense can, as a result, be used for spotting periodicity, but that is not its only use.

• There are, as I think has become evident, many different types of spectra. Even matrices have a spectrum, and there are also power-delay spectra in wireless comms that have nothing to do with frequencies or electromagentism! The word "spectrum" has no unique meaning in the physical sciences and engineering; it is usually to be inferred from context if not explicitly assigned.

• I'm reasonably sure that the electromagnetic spectrum referred to by Worldtraveller is really the emission spectrum, (the opposite is the absorption spectrum) or the energy spectrum. I suspect vernacular use of terminology in the cited examples; this doesn't make them wrong, it just means they might be using the 'usual' terminology which may not be the absolutely strict one. It may also be to emphasise that they are talking about the spectrum of the light from the star rather than any other kind of spectrum a star may exhibit (though I don't know what these may be). Granted, these two articles refer to materials, but they could as well refer to stars.

• Energy spectrum needs work. The last sentence really describes an emission spectrum (the photons must have come from somewhere) which highlights the fact that Worldtraveller may be talking about such spectra. I've not come across the rather abstract meaning in the first sentence (which may just hand-wavingly duplicate the second). Energy spectrum can be and is used to describe the quantized nature of the possible frequencies that an atom can emit. Quantum mechanics only permits certain energies to the electrons in an atom and, via Einsteins $E=hf$ only certain frequencies. Frequencies imply an interpretation in the electromagnetic spectrum — the energy spectrum is those portions of the electromagnetic spectrum that may be so emitted.

• In terms of ALoan's comments, spectroscopy wouldn't do as a general disambig. In the context dominating the discussion here, spectroscopy is the process of determining an emission (or absorption) spectrum, but it has other uses I believe.

• In short, there is no particular meaning of "spectrum" that will do for all its occurences, and careful research is needed to determine the correct usage. For example, in the spectral radius article, Smack has linked to the disambig page (as it already was), but spectrum of a matrix would be better. For some more abstruse uses of "spectrum" not even touched on in this discussion, take a look through Category:Spectral theory, which has a very mathematical leaning!

-Splash July 4, 2005 22:52 (UTC)

This is an interesting discussion. I was not advocating a redirect from spectrum to spectroscopy or general disambiguation, simply saying that, in this context, I think a piped link ([[spectroscopy|spectrum]]) makes sense: the relevant paragraph reads thus:

It was discovered by William Herschel on February 15, 1786, and was the first planetary nebula whose spectrum was investigated, by the English amateur astronomer William Huggins in 1864.

Anyway, I'll leave you guys to sort it out. I'll be very glad if it leads to improvement in the various spectrum articles. (Sorry, Worldtraveller, for taking up so much space on your talk page - I guess we should move this discussion to a talk page somewhere - Talk:Spectrum or Talk:Cat's Eye Nebula?) -- ALoan (Talk) 5 July 2005 11:28 (UTC)

Suggest we replace all occurrences of spectrum with can of worms :) Nice to see a Splash as well as a Smack here, any other onomatopaeic users around with an interest in spectra? No probs about taking up space here - I'll archive my page soon, it's getting large, but happy to copy the discussion elsewhere also.
Anyway, I agree that each occurrence of Spectrum needs careful thought as to what the most appropriate disambiguation would be. I don't necessarily agree with Smack that there is a need to distinguish between 'the' electromagnetic spectrum and a spectrum emitted by a body - suggest just adding to the electromagnetic spectrum a section saying that it can be used in this sense.
As for emission and absorption spectra, they're more like terms to describe different types of electromagnetic spectra, in astronomical usage at least. Stars typically show an absorption spectrum - a continuum with absorption lines superimposed. Nebula have an emission spectrum - continuum with emission lines. But many stellar spectra also have emission lines and some are dominated by them, while many nebular spectra have absorption lines.
From what I know, the new article at energy spectrum is mostly accurate, except that astronomers wouldn't really talk about the energy spectrum of photons. While radio astronomers talk in Hertz and optical astronomers talk in Angstroms, X-ray and γ-ray astronomers talk in KeV and MeV, but 'energy spectrum' is used to describe the energy distribution of particles rather than photons (hopes no-one wants to start bringing wave-particle duality into this), as in the second of my example links above which talks about the electromagnetic spectrum emitted by the crab nebula depending on the energy spectrum of electrons emitted by the Crab pulsar.
I think I agree, Splash, with your assessment of the frequency spectrum article, but I've always got a bit nervous at the word 'fourier' so would defer to those with more expertise than me in this area!
Linking to spectroscopy (or astronomical spectroscopy from the word 'spectrum' might not be very helpful to the reader, I think, as we then have the problem with that article of what spectra we are talking about (that article currently links to frequency spectrum; might be better to link to the disambig page as spectroscopy could mean analysis of many different types of spectrum). But there probably should be a link to spectroscopy in most cases where spectra are mentioned.

My thoughts on the way ahead - I've seen several articles linking to frequency spectrum where I think a link to electromagnetic spectrum would be more appropriate. If we are all agreed that electromagnetic spectrum is the appropriate link for articles like Cat's Eye Nebula, then I will change the link in similar situations. In many cases the appropriate link will be different - I would only be changing links on astronomical-related articles, other people might wish to check on their area of interest. We might yet need to think about whether the articles at energy spectrum etc need to describe the differing uses of the terms (if that is the case) in different areas of science. Worldtraveller 5 July 2005 11:58 (UTC)

To User:Splash: Correct me if I'm wrong, but it seems to me that the gist of your comments was that disambiguating 'Spectrum' requires more thought than a search-and-replace operation.

To User:Worldtraveller and all others: I'm going to try to articulate a particular point of view that I've developed over the course of my wiki participation. Maybe I ought to write an essay about it sometime (shameless plug for the essay I wrote some time ago).

My personal bias with regard to the task of writing an encyclopedia is that I have little patience for convention, particularly if it becomes a stumbling block for accurate description of theory. I'm having a hard time articulating my point of view, partly because it's subtle and partly because it's late at night, so I'll try an example. Consider wire. It is obvious that wires come in different thicknesses. In the United States, wire thicknesses are standardized according to the American wire gauge (AWG) system. AWG is so firmly established that engineers, mechanics and anyone who works with wire usually speaks of the gauge of a wire rather than its thickness.

Nevertheless, the fundamental property of a wire is its thickness, and the gauge is merely an expedient quantification thereof. In writing a technical document, one would of course refer to wire by its gauge. However, I believe that it would be a big mistake if a Wikipedist wrote something like, "For [some use], wire of different gauges may be used." I don't want to make a similar mistake by adhering to conventional usage of spectral terminology. We should base our decisions on textbook definitions. I'll admit that this viewpoint may be only the personal bias of a greenhorn student at a decidedly theoretical university, but I've seen a good many instances of convention being passed off as fundamental truth.

--Smack (talk) 7 July 2005 04:47 (UTC)

In response to your comment: I don't really understand what you mean. You seem to be saying we should not use conventional language. You also seem to be saying you think that just linking all the uses of spectrum to the disambig page is fine; I can't believe that's what you mean though. However, I think we absoluetly should use conventional terminology, but with explanation of what it means. If we don't go conventional, we're effectively engaging in POV and/or original research. The partic. example you give calls for an explanation of the meaning of "gauge of a wire" and then the use of the correct terminology from them on. To make up our own terminology or not use conventional wording would be to mislead our readers. -Splash 7 July 2005 18:20 (UTC)

Yes, it would be slightly misleading, but the misunderstanding is only a trivial one of terminology. I'm willing to live with that if the alternative is an unclear or convoluted explanation that risks leaving the reader with no understanding at all. --Smack (talk) 9 July 2005 04:56 (UTC)

I'm still not sure what you're suggesting. Do you mean we should make up new words for Wikipedia? Surely not. Or do you mean to use wikilinks to the generic disambig page when you're unsure of the appropriate correct disambig? The second I'd be ok with, as long as you were happy with them gradually being replaced with correct disambigs. However, we should not mislead the readers one inch: this is an encyclopedia, not a place for coining new terminology or original research. As for your wire example, it needs to be taken to its logical conclusion, otherwise you are effectively saying that Wikipedia should never use technical language if even it is explained elsewhere in the 'pedia (which is what WIkilinks are for):

"For [some use], long, thin bits of metal with plastic on the outside of different thicknesses may be used."

And we wouldn't want that, I'm pretty sure! Anyway - like I said, I'm just not sure what you are suggesting: if you are suggesting removal of references to spectra, that would not be sensible, if you are suggesting a careful disambig, that'd be fine. -Splash 9 July 2005 13:25 (UTC)

If I see a link to a redirect to a disambig page, and I'm unsure of the proper way to disambig it, I will change it to point to the disambig page, but the reasoning behind that is unrelated to what we're discussing here. As regards your "long thin bits of metal", I see that I still haven't made myself clear. Anyone over the age of five knows what a wire is. There's no problem with that. I'm not advocating radical changes of terminology. Here's another example. It's somewhat different from the wire example, but since I haven't crystallized my thoughts on this question, it'll have to do.

I recently saw an article about some particular type of signal. It referred to the frequency spectrum of the signal and then said, by way of explanation, that the frequency spectrum of that signal may be computed by the FFT algorithm. IMO, this is equivalent to saying that the product of two numbers is what you get when you punch them in on a calculator, separated by the multiplication symbol. It's accurate, and it corresponds to the most practical way to perform Fourier transforms, but it's remarkably unhelpful in understanding what the frequency spectrum and Fourier transform actually are. There are only two signal-processing articles that I can think of that would be truly incomplete without a link to the FFT, and those are the Fourier transform and the discrete Fourier transform. No other article even needs to know about the existence of the FFT. --Smack (talk) 9 July 2005 16:04 (UTC)

Ok, I see what you mean. And I'd certainly agree with you on the example you quote above. However, the strength of the WP is that we can include a mention to the (F)FT in every article that it relates to (which would, I think be many sigproc ones) without having to clutter the rest of article, since we just use a wikilink. Those that care can follow it, those that don't, don't. We can have depth and clarity. The FT is so fundamental to much of sigproc (and telecomms) that I'd be shocked if all the links to it suddenly disappeared. -Splash 9 July 2005 17:07 (UTC)

The Fourier transform as a theoretical device is fundamental to sigproc. The fast Fourier transform computational algorithm is fundamental to nothing. Drawing again from my multiplication analogy, linking to the FFT every time the FT occurs is like linking to hand calculators every time you see a formula involving multiplication, such as $F=ma$ or $E=h\nu$. --Smack (talk) 03:54, 10 July 2005 (UTC)

Spectra of particles?

I've gotten back to cleaning up the massive Whatlinkshere page for Spectrum, and I just ran across a reference to a spectrum of particles. What is the right page to link to for this term? --Smack (talk) 00:42, 18 July 2005 (UTC)

This page is a mess

It is also not a disambiguation page and should not be so labeled. It's nowhere near as bad as it used to be a year ago, but it's not at all well organized. I'm going to move its material back to spectrum. Maybe later I'll work on organizing it. Michael Hardy 21:23, 18 July 2005 (UTC)

Someone actually deleted the section on ghosts ("spectral evidence", etc.). That section is needed. Without it, the material that follows it cannot possibly be understood.

This page really is a horrible mess. The person who moved it all from spectrum must not have attempted to read the whole thing. Michael Hardy 21:38, 18 July 2005 (UTC)

You're right that it's a very bad disambiguation page, but that's my fault, not the article's. I didn't know what to do with the "general description", so I left it dangling. It should be parceled out to more specific pages. I guess it would be possible to just put it back at Spectrum, but that's a very messy and confusing way of doing things.

Also, I'm going to put back the odd-sounding statement about electromagnetic spectrum. You're right that it uses two different definitions (the one in "The infrared part of the spectrum..." and the one in "The observed lines in the spectrum of hydrogen..."), but it was arrived at in the very long discussion above that there's no satisfactory way to distinguish them. If you have alternate suggestions, I'd be glad to hear them. --Smack (talk) 03:49, 19 July 2005 (UTC)

P.S. Here's what's wrong with having a non-redirect "main" page at Spectrum. Wikipedia allows for disambig pages to coexist with non-disambiguated content if there is one dominant usage of a word. However, what you have created does not fit the bill. The information that your "main" page contains is part etymology, part history of science. There is no actual usage of the word 'spectrum' that it corresponds to. It's good information, but it's exceedingly unlikely to be what a reader is looking for when he follows a link. Thus, he has to jump through two hoops to get to where he needs to be: the very long disambig page, and the inconspicuous link in the intro of the "main" page. BTW, I should probably tell you that I've been plugging away doggedly in an effort to remove all links from the article namespace to that blasted page. (My eyes are thoroughly glazed over.) See my contributions for details. --Smack (talk) 04:33, 19 July 2005 (UTC)

P.P.S. Yes, I did read the whole thing, but I decided that what I did was better than letting the status quo sit and fester until a bunch of people could be gathered and induced to hash out the most prudent course of action. --Smack (talk) 04:39, 19 July 2005 (UTC)

Hello there. I've just touched up the formatting a bit on this page. Overall, I think it's fine, but there are some details (such as the itallics around every link and the large, self-referrential lead-in paragraph) which didn't quite fit with the Wikipedia style guidelines for disambiguation. I think that given the work you two have done, and these minor changes, this is a very nice way to introduce the broad topics at hand. Nicely done, folks. -Harmil 20:31, 21 September 2005 (UTC)

Lead-in paragraph

The problem with the lead-in paragraph (and the reason I replaced it) is that it tries to explain how other articles are structured, and one thing that can be guaranteed about Wikipedia is that those other articles will change over time. Saying that the spectrum article is "brief", for example, can't be assumed with any accuracy.

That said, there's also a lot of verbosity in the current version of the intro paragraph that doesn't really convey any extra information. "This page links to longer articles on narrower instances of things called spectra:" is not really telling us much more than, "the following are possible specific applications:".

The edit summary on the last revision was, "No, it is NOT about "explor[ing] the word itself"!" Besides being overly emphatic, it seems to be incorrect. The text you put back was, "The spectrum article explains why the seemingly disparate topics referred to by this same name are not in fact disparate" Again, a very verbose way of saying the same thing that I did, "The spectrum article explores the word itself." If you have an objection to the word, "explores", then fine, but can we please do away with the dramatic lead-in that seems to imply that there's some controversy that, as far as I'm aware, never actually existed? Keep in mind that disambiguations are supposed to hold the reader's attention for a short a time as possible, quickly directing them to the page that they actually wanted. Thanks. -Harmil 10:47, 22 September 2005 (UTC)

Primary topic

Discussion of primary topic here. Consensus is that there is a broad primary topic: spectrum. --Kkmurray (talk) 18:57, 15 August 2014 (UTC)