Misplaced Pages

Revision as of 20:15, 29 August 2009

This is the talk page for discussing improvements to the Speed of light article. This is not a forum for general discussion of the article's subject.

Put new text under old text. Click here to start a new topic. New to Misplaced Pages? Welcome! Learn to edit; get help. Assume good faith Be polite and avoid personal attacks Be welcoming to newcomers Seek dispute resolution if needed Article policies Neutral point of view No original research Verifiability

Find sources: Google (books · news · scholar · free images · WP refs) · FENS · JSTOR · TWL

Archives: Index, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18Auto-archiving period: 7 days

	Speed of light is a former featured article. Please see the links under Article milestones below for its original nomination page (for older articles, check the nomination archive) and why it was removed.
	This article appeared on Misplaced Pages's Main Page as Today's featured article on October 29, 2004.
Article milestones Date Process Result August 17, 2004 Featured article candidate Promoted December 7, 2008 Featured article review Demoted
Current status: Former featured article

Physics: Relativity B‑class Top‑importance

This article is within the scope of WikiProject Physics, a collaborative effort to improve the coverage of Physics on Misplaced Pages. If you would like to participate, please visit the project page, where you can join the discussion and see a list of open tasks.PhysicsWikipedia:WikiProject PhysicsTemplate:WikiProject Physicsphysics B This article has been rated as B-class on Misplaced Pages's content assessment scale. Top This article has been rated as Top-importance on the project's importance scale. This article is supported by the relativity task force.

Template:WP1.0

Archives

Archive 1 (Up to end of 2004) Archive 2 (2005 – July 2006) Archive 3 (July 2006 – end of 2006) Archive 4 (2007) Archive 5 (2008) Archive 6 (Jan 2009 – February 2009) Archive 7 (February 2009 — July 2009) Archive 8 (—) Archive 9 (—)

This page has archives. Sections older than 7 days may be automatically archived by Lowercase sigmabot III.

This is the talk page for discussing improvements to the Speed of light article. This is not a forum for general discussion of the article's subject.

Put new text under old text. Click here to start a new topic. New to Misplaced Pages? Welcome! Learn to edit; get help. Assume good faith Be polite and avoid personal attacks Be welcoming to newcomers Seek dispute resolution if needed Article policies Neutral point of view No original research Verifiability

Find sources: Google (books · news · scholar · free images · WP refs) · FENS · JSTOR · TWL

Archives: Index, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18Auto-archiving period: 7 days

Section on Speed of light by definition

Perhaps it would be helpful to rewrite this section in a context where problems of succinctness are less severe. Then, with a clear statement of the situation in hand, perhaps a more accurate short intro sentence can be constructed?

I'd suggest this section begin with the pre-1983 definition:

“On October 14, 1960 the Eleventh General Conference on Weights and Measures redefined the International Standard of Length as 1,650,763.73 vacuum wavelengths of light resulting from unperturbed atomic energy level transition 2p₁₀ 5d₅ of the krypton isotope having an atomic weight of 86. The wavelength is

λ = 1 m / 1,650,763.73 = 0.605,780,211 µm

At different times some national laboratories used light sources other than krypton 86 as length standards. Mercury 198 and cadmium 114 were among these and they were accepted by the General Conference as secondary length standards.”

This beginning point has the merit of being perfectly understandable: define the length standard "metre" in terms of a fundamental length of nature, a particular wavelength.

Given this metre and a time measurement, the speed of light in m/s is readily understood as a length / time of transit, and is roughly 299 792 458 m/s.

That introduction then can be followed by an explanation of the 1983 switch to a defined speed of light and a length standard based directly upon time-of-transit. That explanation would clarify why the term "speed of light" in SI units is then to be understood only in the context of length determination as in length = 299 792 458 m/s × time of transit. The number 299 792 458 m/s is explained to be a convention adopted by CODATA, and caution given that it should not be confused with a numerical value for the physical speed of light, which can no longer be expressed in m/s, but can be expressed, for example, in wavelengths of some atomic transition / second.

Of course, references to all critical statements are to be provided, such as Wheeler; Jespersen; Sydenham. What do you all think? Brews ohare (talk) 21:14, 21 August 2009 (UTC)

I think you already know what we all think. Can you please stop disrupting and harassing this talk page? Thank you. DVdm (talk) 21:18, 21 August 2009 (UTC)

The reliable sources you quote are not critical of the current state of affairs, that is just a misconception on your part. Martin Hogbin (talk) 21:20, 21 August 2009 (UTC)

If you want to discuss this subject further, I have set up a page for it at User_talk:Martin_Hogbin/Speed_of_light_set_by_definition Martin Hogbin (talk) 21:22, 21 August 2009 (UTC)

Martin: You state I am under a misconception, but do not say what that misconception is. I have provided sources, and you do not address them. I have outlined an approach and you dismiss it without comment. Humbug. Brews ohare (talk) 22:15, 21 August 2009 (UTC)

This clearly goes under wp:nor#Synthesis of published material that advances a position. DVdm (talk) 21:24, 21 August 2009 (UTC)

Perhaps you could elaborate by stating exactly what has been said that goes beyond the cited sources? You have never addressed the sources, the issues; go beyond bald accusation and put some meat on the bones, eh?. Brews ohare (talk) 22:13, 21 August 2009 (UTC)

Well Martin et al.: I have explained this matter carefully to you all. I have made above a simple and clear outline of a proposed change. You all simply will not engage on a detailed basis, but continue to harangue in abstractions. That is not how to advance. The article is in flat contradiction to sources despite your unsupported claims to the contrary. It is logically unsound. It is factually incorrect. It makes the absurd claim that the speed of light is known exactly, in distinction (contradiction) to all other observations of nature. All this will become clear in due course, I hope. Brews ohare (talk) 21:31, 21 August 2009 (UTC)

The speed of light is known exactly because it is exactly defined. You're claiming that we can't know a square has four sides because we defined squares as having four sides. Headbomb {_{κοντριβς} – WP Physics} 01:04, 22 August 2009 (UTC)

The analogy with a square is a false one because we have defined "four" and "side" quite independently of "square" so when we make a statement about the square using four and side it is useful and meaningful. However, when we define the metre as 1/299.. of the distance travelled by light in a second and then state that the SoL is 299.. m/s we have a circular statement that is not a definition at all because, although we do have an independent definition of a second, we do not for the metre. To state it another way, "The SoL is 299.. m/s" says absolutely nothing scientific about the speed of light because we do not know what a metre is withought first knowing the SoL. Of course the statement is meaningful to a layman (and indeed to us all) because we all "know" what a metre is from its previous definition.Abtract (talk) 09:33, 22 August 2009 (UTC)

The numerical value of the speed of light in SI units is known exactly. The mechanical length of the meter has, as far as can be measured or detected, not changed.

Your statement:' "The SoL is 299.. m/s" says absolutely nothing scientific about the speed of light because we do not know what a metre is without first knowing the SoL' is not correct, we do not need to know the SoL, all we need to delineate the meter is some light, which we have. Martin Hogbin (talk) 09:47, 22 August 2009 (UTC)

Thanks. Perhaps you could enlighten (sorry couldn't resist) me as to how we delineate a metre using some light. Abtract (talk) 09:55, 22 August 2009 (UTC)

You, obviously, also need a means if realizing the second. In principle, all you then do is let the light travel for the appropriate fraction of a second and see how far it goes. That is your metre. Of course, in practice, this is done with interferometry and light of a measured frequency. Martin Hogbin (talk) 10:50, 22 August 2009 (UTC)

Thanks for confirming the circularity and apologies for previouly stating the problem badly. What I should have said is:

"The SoL is 299.. m/s" says absolutely nothing scientific about the speed of light because we do not know what a metre is without first defining it in relation to the SoL.

Abtract (talk) 13:17, 22 August 2009 (UTC)

I do not understand what you mean by 'nothing scientific'. The current article gives the speed in SI units. Obviously, if you do not know what a second and a meter are this will mean little to you. This applied just as much when the meter was defined by a metal rod. If you had never seen the rod or a meter rule, you would have no idea what the number meant. Martin Hogbin (talk) 15:12, 22 August 2009 (UTC)

If we had defined "four" as the number of sides a square has, and then state that a square has four sides, that would be a circular statement and not very helpful. Imho defining the metre as the distance travelled by light in 1/299.. metres and then stating that the SoL is 299.. m/s is equally circular and should be avoided without some reference to the way a metre is defined (which we have now done). That's all I'm saying. Abtract (talk) 15:29, 22 August 2009 (UTC)

As you say, the current lead briefly explains why the speed of light is an exact value so I suggest we leave drop the subject. Most people are getting rather fed up with. Martin Hogbin (talk) 15:51, 22 August 2009 (UTC)

Agreed, and thanks for your patience. Abtract (talk) 17:44, 22 August 2009 (UTC)

Proposed lead first para

In an attempt to overcome the circularity problem, I offer the following wording (second part taken more or less verbatim from the source):

The term speed of light generally refers to the speed at which light (or any other electromagnetic wave) travels in vacuum. This is a fundamental physical constant, usually denoted by the letter "c".

The metre has been defined (SI units) so that the speed of light in vacuum has the exact value:

c = 299 792 458 m/s ... which can be used as a standard. Abtract (talk) 22:29, 21 August 2009 (UTC)

This is close to what we once had before, but I am not sure why it is better that the wording we have now. There was a feeling that we wanted the fundamental constant bit first. Also, I do not understand what you mean by the 'circularity problem'. Also, I would prefer to say 'The metre is defined' rather than 'The metre has been defined'. Martin Hogbin (talk) 22:46, 21 August 2009 (UTC)

I have explained what I mean by the 'circularity problem' in the section above (Section on Speed of light by definition) Abtract (talk) 09:36, 22 August 2009 (UTC)

Is or has been, makes no difference to me; I think this phrasing is superior to what's there now, though I abhor staring with "the term". My suggestion would be:

In physics, the speed of light is a fundamental physical constant, the speed at which light (or any other electromagnetic wave) travels in an ideal vacuum. It is usually denoted by the letter "c". In the SI system of unit, the metre has been defined such that the speed of light has the exact value c = 299 792 458 m/s.

Dicklyon (talk) 00:54, 22 August 2009 (UTC)

Abtract, I can see why you do not like saying 'the term', it is rather ugly, but I think we have to say something like that. 'The speed of light' is, even in everyday usage, is a phrase with a specific meaning, namely the speed of light in free space. The natural English phrase 'the speed of light' can be used to mean many things, just like saying, 'the speed of a train'.

Until we have defined our terms, I do not think that we can say, for example 'the speed of light is 299 792 458 m/s' because this is not always the case, the speed of light in glass is less, for example. My alternative suggestion was to say something like, "the speed of light is generally taken to refer to ...". — Preceding unsigned comment added by Martin Hogbin (talk • contribs)

Martin, that was me. So why did you put back "the term"? The lead defines the meaning in the context of the article, without that. I recommend we put back the more standard-form lead, and you tell us what it is in it that you "strongly object" to. You can certainly say, as there is a strong consensus to say, that 'the speed of light is 299 792 458 m/s' once you've said that 'the speed of light is a fundamental physical constant, the speed at which light (or any other electromagnetic wave) travels in an ideal vacuum'. Was I wrong in thinking that we had a consensus to go this way? Does anyone else object? Dicklyon (talk) 22:08, 22 August 2009 (UTC)

The problem is that the words "the speed of light" in the phrase The term "speed of light" generally refers to the speed at which light (or any other electromagnetic wave) travels in vacuum does not have the same meaning as it has in the phrase: The metre has been defined (SI units) so that the speed of light in vacuum has the exact value: c = 299 792 458 m/s .

The first occurrence refers to the physical, observable, real speed of light. The second occurrence refers to the factor used in the SI system of units to relate length to time-of-transit. This last number agrees with an approximate value for the physical speed of light as measured prior to 1983, now enshrined by definition to lie outside observation, forever immutable. Brews ohare (talk) 22:59, 21 August 2009 (UTC)

Brews, please stop dragging dead parrots into every discussion on this topic. Martin Hogbin (talk) 09:32, 22 August 2009 (UTC)

The numerical value 299 792 458 m/s is not measured in SI units

In SI units the number 299 792 458 m/s is used to relate time-of-transit to length in the formula length = 299 792 458 m/s × time.Jespersen and in Sydenham. Given this relation, the number 299 792 458 m/s is established by definition, and is not subject to confirmation by experimental observation. (See same references.) In short, the number 299 792 458 m/s in the SI system of units is not measurable.

In contrast, if a length is expressed in wavelengths of some particular atomic transition (for example, as per the definition of the metre prior to 1983), the time it takes light to traverse this length can be measured (as is done for the metre), and the speed of light then is measurable in terms of wavelengths/s. Needless to say, if the wavelength is expressed in metres, this number for the speed of light will be close to 299 792 458 m/s, but unlike 299 792 458 m/s, this measured speed of light will have an error bar associated with the errors of observation.

This measured value of the real speed of light is not used in the modern SI definition of length because (at the moment, anyway) the error bar introduced by use of the measured value of c is larger than that incurred using 299 792 458 m/s with the time-of-flight methodology. Brews ohare (talk) 23:17, 21 August 2009 (UTC)

Yes, so you've told us, and we get it. How does this relate to the article? I don't see anything about "measured" that needs to be fixed. Dicklyon (talk) 23:34, 21 August 2009 (UTC)

The number 299 792 458 m/s is not the measured speed of light in SI units, and it is not the "exact" speed of light in SI units; it's just the number in the formula length = 299 792 458 m/s × time of flight. The measured speed of light has an error bar, and is not known exactly, but is approximately 299 792 458 m/s. The statement in the lead In SI units, c is exactly 299,792,458 metres per second. is not correct, or perhaps I should say it is not correct if c is interpreted as the actual speed of light, and not just as the proportionality factor in the equation for length. Brews ohare (talk) 00:06, 22 August 2009 (UTC)

Brews, the scaling constant and the speed of light are the exact same things. I suggest you drop the stick before someone files an ANI request and get topic-banned like David Tombe just got. Headbomb {_{κοντριβς} – WP Physics} 08:07, 22 August 2009 (UTC)

Or you can continue the discussion somewhere else. Martin Hogbin (talk) 08:56, 22 August 2009 (UTC)

What was the measured speed of light in 1982 in m/s, before the meter was redefined to be based upon the speed of light? —Finell (Talk) 12:47, 22 August 2009 (UTC)

299 792 458 m/s. Martin Hogbin (talk) 14:43, 22 August 2009 (UTC)

Thanks, that's what I thought. Martin, I am sure that you can explain the fallacy of the DT-BO position in terms that they are able to understand, including demonstration (not bare assertion) of how they either misinterpret their sources or that they are not WP:RSs. Whether they will agree with it is a separate matter. While I would welcome their understanding and agreement, that is not my primary concern; my primary concern is the process. The explanation belongs on this Talk page and is the the appropriate way to close this discussion topic. As tendentious as they have been, they have been trying in good faith to improve the article and their thesis is relevant to the article's topic. Therefore, the error in the thesis should be explained, here. FYI, Jimbo looked at this talk page and agrees that neither side has addressed the other's points or supported their own position by showing what the RSs actually say on the subject (he doesn't claim to understand the topic, but he got the gist of the dispute). The explanation will serve in the future to show how the issue was resolved, if it arises again. Thank you. —Finell (Talk) 22:10, 22 August 2009 (UTC)

Reply to Finell.

I will make an attempt here to state my case in response to Finell with supporting references as a summary of how I see the matter. I invite the other editors on this page to support their views in a similar manner. The following is not intended to be argumentative, but a straightforward response to Finell.

Background

There are several different points to make. The first is that there does exist in the universe something called the speed of light that, according to relativity, relates to the structure of spacetime and the bounds on the transport of information and of matter. Second, the numerical value of this speed can be measured. One way to do this is to take a set length, for example the wavelength of an atomic transition, and determine the time it takes for light to transit this distance. The speed of light is then c = λ/t. That approach was the methodology used in the SI units prior to 1983. See this NIST timeline for details: definition:

“On October 14, 1960 the Eleventh General Conference on Weights and Measures redefined the International Standard of Length as 1,650,763.73 vacuum wavelengths of light resulting from unperturbed atomic energy level transition 2p₁₀ 5d₅ of the krypton isotope having an atomic weight of 86. The wavelength is

λ = 1 m / 1,650,763.73 = 0.605,780,211 µm

At different times some national laboratories used light sources other than krypton 86 as length standards. Mercury 198 and cadmium 114 were among these and they were accepted by the General Conference as secondary length standards.”

It seems to me doubtful that any of the parties disagree with this background. They probably also agree that this pre-1983 approach resulted in a value for c of approximately 299 792 458 m/s, and that this value was obtained only to within observational error (that is, this number is not an exact value).

SI units post 1983

In view of the status of the speed of light as a physical constant, independent of frequency, direction etc., it is possible to set up a laboratory to realize the speed of light (subject to some caveats about corrections needed to adjust for "nonideal vacuum"). That being so, lengths can be compared by comparing the times of transit of light along the lengths using a formula length = c × time of transit. Actually, one does not have to know the numerical value of c in order to make such length comparisons: one has to be sure only that the speed of light in vacuum has been realized. Thus length comparisons are simply ratios of transit times, and the actual value of c simply divides out in such ratios. For this reason, as pointed out by Jespersen of the Bureau of Standards (underscores are mine):

One fallout of this new definition was that the speed of light was no longer a measured quantity. The reason is that, by definition, a meter is the distance light travels in a designated length of time, so however we label that distance - the speed of light is automatically determined. ... turning c into a conversion factor whose value is fixed and arbitrary.

As stated by Sydenham in a rather definitive article that is often cited (underscores mine):

The numerical value of the speed of light c = 299 792 458 m/s is the result of numerical standards chosen for the standards of time and of length. Thus the speed of light as a numerical value, is not a fundamental constant.

Summary

We see then, that while the "speed of light" was initially a measurable quantity pre-1983, post-1983 is has become a "arbitrary" value that is not "a fundamental constant". This odd situation is understood by noticing that the speed of light still is a physical quantity, but the terminology "speed of light" has two different meanings: pre-1983 it was a measured value, and post-1983 it is a "conventional value" of 299 792 458 m/s.

This change in meaning is a peculiarity of the SI units. The speed of light is used in two senses: the actual physical experimental speed of light and the SI units conventional value for the speed of light. Because of the post-1983 definition of the meter, the actual physical speed of light can no longer be expressed in m/s; all that is possible is the conventional, defined value 299 792 458 m/s. To state the actual physical speed of light in SI units, that is the physically observable behavior of light, one must resort to expressing the speed of light in terms of some physical length, for example, wavelengths of some transition/second.

The debate on this WP page stems from confusion between the conventional defined value for the speed of light of 299 792 458 m/s, (which is an exact number chosen by definition to be a value chosen for convenience to be close to the measured value prior to 1983), and the actual physical speed of light, which cannot be expressed any longer in m/s because of the convention that the metre is defined so this number always is 299 792 458 m/s.

Within the new definition, the actual physical speed of light must be expressed not in m/s (because that results in tautology) but in terms of some physical length, for example in terms of wavelengths of some atomic transition/s, as was done prior to 1983. That is, c can be expressed as some number of λ's per second. The real, physical speed of light is not calculable from some basic theory, and so cannot be stated as a definite numerical value, but must result from measurement with an observational error bar attached to it.

I hope that this exposition lays out the situation clearly. It is not intended to be argumentative, but simply to point out the confusing usage of the term "speed of light" to refer both to a fact of physical nature (that is beyond human capacity to measure without an error bar) and also to an exact conventional "conversion factor" of 299 792 458 m/s Brews ohare (talk) 16:04, 23 August 2009 (UTC)

... and if everyone politely ignores this clearly laid out exposition of your confusion, will you then finally stop repeating it on this talk page? DVdm (talk) 16:47, 23 August 2009 (UTC)

DVdm: You have made what is simply a nasty uncooperative response to a sincere effort on my part. Your failure to be constructive and your snarky attitude does nothing to advance matters. My view is that your ideas about these matters are misinformed, and if you disagree, you should support your views, not badmouth earnest effort. Brews ohare (talk) 19:23, 23 August 2009 (UTC)

Perhaps I should have ignored it myself to begin with. My apologies. DVdm (talk) 21:23, 23 August 2009 (UTC)

True apology is accompanied by changed behavior. False apology is just a form of sarcasm. Brews ohare (talk) 02:01, 24 August 2009 (UTC)

This was not an apology to you but to the others, for not following my own hint to politely ignore your repetitive talk page disruptions. DVdm (talk) 10:00, 24 August 2009 (UTC)

Rest

I think that the speed of light, and specially its talk page, desperately needs some rest ;-) - DVdm (talk) 10:55, 22 August 2009 (UTC)

I completely agree that we should stop discussing the speed of light being fixed by definition. This is an impeccably sourced fact that must remain in the article. The fact that some people seem to be having difficulty getting their heads round this fact does not change it. I have offered a home for further discussion of the subject by those interested in it. Martin Hogbin (talk) 11:09, 22 August 2009 (UTC)

Martin's lead

In this diff, Martin took us back to saying "the term", removed a ref, adding a dangling "this", and made some other subtle wording changes. Does anyone understand or support this change, or can I put back the previous one? Dicklyon (talk) 22:12, 22 August 2009 (UTC)

Where is the dangling this and what was the previous version? Abtract (talk) 23:28, 22 August 2009 (UTC)

Click the diff link to see both versions, including "This is..." with no subject noun. Dicklyon (talk) 23:31, 22 August 2009 (UTC)

I have no preference either way. Abtract (talk) 23:38, 22 August 2009 (UTC)

We could change 'This is' to 'It is', perhaps and maybe make it one sentence.

Regarding 'the term', I believe that it is important to make clear that we are talking not about the normal English meaning of 'the speed of light' but a specific although widely used meaning of 'the speed of light in a perfect vacuum and an inertial frame (and maybe other conditions) that is the fundamental constant of the spacetime in which we live'.

I do not think that it is good English to write 'The speed of light generally refers to the speed at which light travels in free space'. Maybe we could have, 'The speed of light is a term that generally refers to the speed at which light travels in free space'. Martin Hogbin (talk) 12:20, 23 August 2009 (UTC)

In view of comments made elsewhere, I should also point out that this is not my lead it is the result of collaborative editing in the section above 'Proposed start to the lead'. Martin Hogbin (talk) 12:35, 23 August 2009 (UTC)

I don't think we want "generally" or "refers" or "term" in there at all; just say what it is, as in most other articles. And if any of those collaborators have an opinion, they should let us know. Dicklyon (talk) 15:07, 23 August 2009 (UTC)

I can live with the new version, but I prefer the original one, so afaiac, go ahead. - DVdm (talk) 12:51, 23 August 2009 (UTC)

I put back the standard-form opening. If anyone besides Martin think that "the term" is a better approach, they should say so here. I left Ruslik0's addition of "partly to allow more precise measurements of distance" in the lead paragraph, even though I think it's a bad idea to go off on this tangent in the opening paragraph. Dicklyon (talk) 16:05, 23 August 2009 (UTC)

I have no problem with the first sentence but I have restored the second one which says all we need to say. We do not need the date when the standard was changed in the lead, only the current standard. As Dick has said, we do not want just one of many reasons for the change. Also, as has been pointed out by Abtract, the speed of light is not strictly a defined quantity, it is fixed by virtue of the definition of the metre. Unsigned comment probably due to Martin Hogbin

OK, sounds like we're converged. I re-wikified, and changed "exact" to "defined" to avoid giving the false impression that the speed of light has been exactly measured or something like that. Hopefully, any further expansion or clarification of this bit of complication can be kept out of the lead. Dicklyon (talk) 19:55, 23 August 2009 (UTC)

Sorry, that last bit was left there by me by mistake. I have now deleted it. Have a look at the article. 86.142.238.242 (talk) 20:23, 23 August 2009 (UTC)

The statement "In SI units, the metre is defined such that the speed of light has the exact value of 299,792,458 metres per second." is technically accurate. However, this statement is not a simple one, involving as it does the notion that one actually is permitted to set an exact value for the speed of light, while normally one would think of it as a measured quantity. If one tries to explain this matter as a decision to redefine the metre so as to make it so, that again appears highly non-intuitive, and possibly circular. I do not find it satisfactory to leave the reader dangling, hoping they might find and read by some accident the subsection "Speed of light by definition". Brews ohare (talk) 20:34, 23 August 2009 (UTC)

Actually, the statement itself is simple, and that's what's good about it, given its position in the lead. Any physical, metaphysical, semantic, and philosophical complexity associated with it can well be deferred to the section about how and why it is a defined quantity. You have spent much of the last year pushing to include such complexity in the lead, and the consensus is clearly against you on that. Dicklyon (talk) 20:38, 23 August 2009 (UTC)

Call it simple to suggest that one can define the speed of light when intuition suggests the speed of light should be measured? Brews ohare (talk) 20:46, 23 August 2009 (UTC)

I guess I spoke too soon about convergence. Abtract has thrown it out and started over again. Abtract, can you please explain what and why? Dicklyon (talk) 23:05, 23 August 2009 (UTC)

Sorry about that but I started to make a couple of minor clarifications, then it came to me that the paragraphs were not in the ideal order and one thing led to another ... but it does read much better now imho. Abtract (talk) 00:27, 24 August 2009 (UTC)

The opening paragraph of three choppy little sentences reads better to you? And you're OK ignoring the long-fought consensus to put the exact value in the opening paragraph? Dicklyon (talk) 03:17, 24 August 2009 (UTC)

Definition and measurement

The WP article states:

In the International System of Units the metre is defined such that the speed of light in vacuum has an exact value where c = 299,792,458 metres per second.

That is an accurate statement, although it omits explanation of how it is possible to have an exact value for a physically observable entity the speed of light despite the fact that no observable entity has an exact value when that value cannot be calculated from scratch, but must be measured. That conundrum is only partly fixed by statement that this feat is accomplished by definition of the metre as a length that satisfies c = 299,792,458 metres per second. The natural response is "Huh? Isn't c measurable, and subject to measurement error? Isn't this 'definition of the metre' argument pure sophistry?".

Of course, to actually measure c in using the SI system, as opposed to defining c one must introduce a length other than the metre. One way to do that is to follow the older definition of length and use the wavelength of some specified atomic transition to do that, call it λ_atomic. Then one measures the time of transit of light through distance λ_atomic and expresses the speed c as c = λ_atomic / time-of-transit. That gives the actual speed of light in units of λ_atomic/s.

Is this number in principle 299,792,458 m/s? The answer is: "No, but it is pretty close". If we could measure λ_atomic in metres exactly and time-of-transit exactly, then we could convert this measured c to m/s, and it would not be exactly 299,792,458 m/s, but nearly so, because 299,792,458 m/s is the measured value (within experimental error) found before the 1983 definition of the metre. See NIST time line.

An interesting point is that if the present definition of the metre were altered so that the defined value for c were 300, 000, 000 m/s instead of 299,792,458 m/s, say, and with this new definition for the meter the wavelength λ_atomic were remeasured in terms of the new metre, the measured speed of light would still be exactly the same number of λ_atomic /s as before with the old metre, because λ_atomic would have the same transit time as before, so c in units of λ_atomic/(time-of-transit) would still be the same. However, in terms of the new metre, now c = 300, 000, 000 m/s. That is what Wheeler means by answering the question "A fundamental constant of nature?" by "Hardly! Rather, the work of two centuries of committees."; and what Jespersen; and Sydenham mean by statements like: the numerical value of c in SI units is a "conversion factor whose value is fixed and arbitrary." Brews ohare (talk) 00:53, 24 August 2009 (UTC)

Yes - 299,792,458 is not a fundemental constant of nature. However, 299,792,458 m/s is. EdwardLockhart (talk) 07:33, 24 August 2009 (UTC)

Your own words about the disputed statement in the lead are, 'That is an accurate statement'. Bearing in mind the length of discussion that there has been on this subject, do you really think that it is desirable or even possible to have an explanation of why the statement is correct in the lead? We have a statement that you agree is accurate. Let us leave it at that and work on an explanation elsewhere. Martin Hogbin (talk) 08:12, 24 August 2009 (UTC)

Hmm I'm beginning to see what the problem is here. Lockhart's statement above that "Yes - 299,792,458 is not a fundemental constant of nature. However, 299,792,458 m/s is." is surely incorrect. The SoL is a fundamental constant of nature (FCoN) ... yes; "c" is a FCoN ... yes; but this FCoN surely cannot be stated in SI unit because the standard unit of length has itself been defined such that the SoL has the exact value 299,792,458 m/s in SI units. This is the circularity problem I referred to previously and I think is what ohare keeps banging on about.Abtract (talk) 09:22, 24 August 2009 (UTC)

There is no circularity problem any more than there is if we define the metre using a metal rod. We mark up a rod with two lines which we define to be a meter apart. So, how far apart are the lines? One metre! How long is a metre? The distance between the lines! This 'circularity' is a consequence of our defining arbitrary standards. Martin Hogbin (talk) 10:33, 24 August 2009 (UTC)

Indeed but, if we then go on to ask how tall am I in metres or how fast does light travel in m/s, there would be no circularity because the metre had been defined independently of me or light. Abtract (talk) 12:06, 24 August 2009 (UTC)

Asking how tall you are is the same in both cases. If we define a meter with a metal rod then we can measure the speed of light based on our standard metre, we cannot measure the metre, it is set by definition. If we define the meter based on the speed of light we can measure out (delineate) the meter but we cannot measure the speed of light, it is fixed by definition. We choose to define some things, others are results of our definitions. Martin Hogbin (talk) 12:46, 24 August 2009 (UTC)

Martin: In the new SI units, how tall you are is how long it takes for light to travel your height in seconds. Converting this to metres by multiplication with 299,792,458 m/s (or any other exact conversion factor) adds absolutely nothing to the experimental info in use. It is still a time of transit, not a measured length. You have not dealt with the three sources Wheeler, Jespersen and Sydenham. If, however, you measure your height in wavelengths, you have added some experimental info and the real, physical, entity speed of light is included in that new info. See the last paragraph of the above comments. Brews ohare (talk) 13:52, 24 August 2009 (UTC)

Do we have a consensus on the first paragraph?

Abtract and I have just made some minor changes to the first sentence.

Abtract removed the brackets, which I agree do not look good in the first sentence.

My changes were:

Change 'electromagnetic waves' to 'electromagnetic radiation'. I think that this is a more general term in that it covers the quantum model better.

Change 'ideal vacuum' to 'vacuum' with a link to 'free space'. This is another case where we cannot hope to explain a much discussed topic in one or two words, best to have a link to the page that discusses the topic fully. I do not think the term 'ideal vacuum' is that standard, it is not used in the vacuum article.

If there are objections to my changes can we please discuss then here and, if there is a consensus, just revert them rather than rewriting the whole paragraph from scratch again.

Yes, we have my consensus :-)

I changed "in an vacuum" to "in vacuum" though. Feel free to change into "in a vacuum" if you prefer that. DVdm (talk) 09:09, 24 August 2009 (UTC)

I have made a few clarifying changes. In addition I think the statement of the actual number should be moved down until after the history para ... there would then be a natural progression. Abtract (talk) 09:20, 24 August 2009 (UTC)

As Dick has pointed out there is a strong consensus not to move the numerical value from the first paragraph. Martin Hogbin (talk) 09:36, 24 August 2009 (UTC)

Although my own preference was to start with the approximate value in the first paragraph, since all that many readers will want to know. The exact value, and the fact that it is exact, is an extra complication that I'd just as soon defer. But I think I lost that argument already. If someone wants to hold another referendum, we can revisit that; otherwise, let's don't. The best we can hope for at this point is to satisfy everyone except Brews and David, I think. Dicklyon (talk) 02:03, 25 August 2009 (UTC)

Unfortunately, there is not unanimity; clarity of exposition has not been reached. The natural response to the lead paragraph as now constituted is: "Huh? Isn't c measurable, and subject to measurement error? Isn't this 'definition of the metre' argument pure sophistry?" Brews ohare (talk) 13:58, 24 August 2009 (UTC)

I agree that readers may find the fact that the speed of light has an exact value surprising. That's why we explicitly say that it is a consequence of the way the metre is defined, and why we supply links to metre and SI units for interested readers to follow. EdwardLockhart (talk) 14:07, 24 August 2009 (UTC)

Brews, that may be your reaction but it is not everyone's. You have agreed that what is written is correct. Considering the pages of discussion we have had here, you must surely accept that it is simply not possible to provide any kind of explanation or justification for the change in the standard in the lead section. Martin Hogbin (talk) 14:08, 24 August 2009 (UTC)

Martin: The main source of this impossibility is a lack of understanding of the real role of 299,792,458 m/s as explained by the three sources Wheeler, Jespersen and Sydenham. If the editors would concede that this number is an arbitrary value set by committee and is not a physical constant of nature, the wording could be rearranged to explain that fact, and the sources provided. However, a number of editors are not yet aware of this fact, and continue to think 299,792,458 m/s is somehow an exact value for the real physical speed of light. This situation could be improved if all editors addressed the three sources instead of inventing new protests of personal construction. Brews ohare (talk) 14:17, 24 August 2009 (UTC)

You see a contradiction where there is none. Yes, 299,792,458 is an arbitrary number set by committee which could equally have been something else. But nontheless, 299,792,458 m/s is an exact value for a fundamental constant of nature. EdwardLockhart (talk) 14:23, 24 August 2009 (UTC)

It is a source of amazement to me that no-one (other than ohare?) seems to understand the circularity of that last statement. A degree is ⁄₃₆₀ of a full rotation; a full rotation is 360 degrees ... duh. Abtract (talk) 14:36, 24 August 2009 (UTC)

EdwardLockhart: As I understand what you are saying, it is within the power of a committee to set a fundamental constant of nature. In other words, if we are dissatisfied with the limitations on space travel set by the speed of light, we can simply declare it to be faster? If that is not what you mean, how do we determine what is the fundamental limitation upon space travel? I've suggested that it can be measured independent of man's machinations as c = λ_atomic/time-of-transit. This suggestion is not my invention: it is exactly what was done prior to 1983, and is the origin of the number 299,792,458 m/s (although post-1983 the error bars have been stripped). Again, I plead with you to address the three sources and to avoid shooting from the hip based upon your immediate understanding of the situation. Brews ohare (talk) 14:40, 24 August 2009 (UTC)

How about taking the parallel with temperature? The Celsius scale was originally defined with the freezing point of water at zero. Using this scale, it was not possible to measure the freezing point of water - it had been arbitrarily set by committee. It was, however, possible to measure the triple point of water, which was approximately 0.01 degrees. However, in 1954, the situation was reversed - the triple point became the defined value (at 0.01 degrees), and the freezing point was now able to be measured (it's about zero).

None of this is committees dictating to nature - we are merely making convenient choices about how to define our arbitrary units. Since we use physical phenomena as our yardsticks, some phsyical phenomena will end up with defined values rather than measured values. Which phenomena these are will depend on the choice of definitions for our units. EdwardLockhart (talk) 14:59, 24 August 2009 (UTC)

EdwardLockhart: We could digress to discuss temperature scales and just what is the analogy between that and the speed of light. However, that strikes me as the long way around. It would be simpler and more direct to address the specific points raised, address the three sources, and state why you do or don't agree. Brews ohare (talk) 15:15, 24 August 2009 (UTC)

I agreee with the sources, e.g. Sydenham's "the speed of light as a numerical value is not a fundamental constant", and Jespersen's "the speed of light was no longer a measured quantity; it became a defined quantity".

You said to Martin:

"If the editors would concede that this number is an arbitrary value set by committee and is not a physical constant" - I think we are all in agreement on this point.

"a number of editors continue to think 299,792,458 m/s is somehow an exact value for the real physical speed of light" - yes, this is the consensus of the editors. It is also what the sources say. EdwardLockhart (talk) 15:34, 24 August 2009 (UTC)

EdwardLockhart Do you also agree with Wheeler answering the question: "A fundamental constant of nature?" by "Hardly! Rather, the work of two centuries of committees."; and Jespersen "c is a conversion factor whose value is fixed and arbitrary"?" Doesn't your acceptance of Sydenham ""the speed of light as a numerical value is not a fundamental constant" stand in contrast to your earlier remark that " But nonetheless, 299,792,458 m/s is an exact value for a fundamental constant of nature." ? Brews ohare (talk) 15:39, 24 August 2009 (UTC)

I agree with Wheeler that there is nothing special about the number 299792458 except that a committee happens to have picked it to be, by definition, the speed of light in SI units.

There is no contradiction between the two statements of mine you quote, or between them and the sources you mention. EdwardLockhart (talk) 15:46, 24 August 2009 (UTC)

Well, EdwardLockhart, I hesitate to parse the two sentences ""the speed of light as a numerical value is not a fundamental constant" (Sydenham) and "But nonetheless, 299,792,458 m/s is an exact value for a fundamental constant of nature." (EdwardLockhart), but in my mind they are absolutely contradictory. Perhaps you can bring out the nuance that separates their meanings? Brews ohare (talk) 15:52, 24 August 2009 (UTC)

The first statement refers to the number 299792458. There's nothing remotely special about this number - it is a purely arbitrary choice. And although there were good reasons to pick this particular number, they too arose from previous arbitrary decisions.

The second statement refers to the speed of light, which is a fundamental constant of nature. And in SI units, it has an exact value, by virtue of the way the definitions have been chosen. This value is 299,792,458 m/s.

But all this has been explained before, so I am not hopeful that this will help any. EdwardLockhart (talk) 16:02, 24 August 2009 (UTC)

EdwardLockhart: Here is the difficulty: the number 299,792,458 m/s was selected by committee. It is fixed. This number was measurable pre-1983 and was 299,792,458 m/s ± Error. It was not known exactly. Fixing this number by fiat in 1983 at 299,792,458 m/s does not make it an exact numerical value for "c"; it makes the conversion factor from transit time to metre an exact value. These are not the same things. That is what the sources say: see the words arbitrary value, not a fundamental constant? Brews ohare (talk) 21:04, 24 August 2009 (UTC)

There is a pretty strong consensus to leave the lead section as it is with respect to this subject. Perhaps we should discuss how the relevant section should read to make the exact situation as clear as possible. My first question is this. We start with:

In 1983 the 17th Conférence Générale des Poids et Mesures defined the metre to be the length of the path travelled by light in vacuum during a time interval of 1⁄299,792,458 of a second. The reasons for using this definition are stated in Resolution 1.

The first sentence is a simple statement of fact, whether or not we like it,but should we also state the reasons given in resolution1 in the text: verbatim, in our own words, not at all? Martin Hogbin (talk) 16:28, 24 August 2009 (UTC)

There is no unanimity upon the first sentence beyond the belief that it will spark incredulity among some readers. That reaction could be avoided by simply pointing out that the approximate value for the speed of light of 299,792,458 m/s that resulted from measurement at an earlier time has been adopted today as a conversion factor for exact translation of times-of-transit to lengths in metres, thereby changing the definition of the metre. Brews ohare (talk) 21:26, 24 August 2009 (UTC)

You must know that your "demand" for unanimity goes against WP:CONSENSUS, and so will be ignored: no single editor has an absolute right of veto over changes. You do not OWN this article, and you would do well not to obstruct its improvement. Physchim62 (talk) 04:17, 25 August 2009 (UTC)

Physchim62: There is no demand for unanimity made here; just the rather mild observation that unanimity has not been achieved. I am not obstructing the improvement of this article, but inviting actual discussion of sources in place of belligerence and attempted intimidation. Brews ohare (talk) 05:05, 25 August 2009 (UTC)

Instead of attitude, how about addressing the issue: There is no unanimity upon the first sentence beyond the belief that it will spark incredulity among some readers. Brews ohare (talk) 11:23, 25 August 2009 (UTC)

Last paragraph of lead

Here is my suggestion for the last paragraph of the lead, which I suggest is made the second:

The speed of light is an important constant connecting space and time in the unified structure of spacetime. It defines the conversion between mass and energy and is an upper bound on the speed at which matter and information can travel. It is the speed of travel of all electromagnetic radiation in free space, and is believed to be the speed of gravitational waves. In an inertial frame, light in vacuum always travels at c. However, when light passes through a transparent medium, such as glass or air, its speed is less. The ratio between the speed of light in vacuum and the speed of light in a medium is called the refractive index of the medium.

I also suggest that, 'regardless of any differences in the frequency, polarization, intensity, or propagation direction of light' is too detailed for the lead. Martin Hogbin (talk) 15:55, 24 August 2009 (UTC) Maybe it should be moved to the 'Constant speed in inertial frames' section. We currently have more detail in the lead than the body. Martin Hogbin (talk) 15:58, 24 August 2009 (UTC)

Now edited to reflect my comments above. If nobody objects I will insert this section as the second paragraph. Martin Hogbin (talk) 21:29, 24 August 2009 (UTC)

IMO, the intro reads better with this paragraph at the end, rather than as second paragraph. The presently preceding paragraphs are more qualitative and general and easier to read. This paragraph is more detailed and requires more of the reader. Brews ohare (talk) 23:01, 24 August 2009 (UTC)

That is one 'no' so far. The reason I would like to change this paragraph is that it currently states many facts as if they were disconnected, 'The speed of light is a constant of spacetime', 'according to special relativity that', 'EM radiation travels at c', etc. These effects are all manifestations of the fact that the speed of light is a constant of our spacetime. Martin Hogbin (talk) 10:37, 25 August 2009 (UTC)

I did not vote "no" on the paragraph restructuring, only on a change of placement. Brews ohare (talk) 10:52, 25 August 2009 (UTC)

Sorry, I misunderstood you. Martin Hogbin (talk) 16:33, 25 August 2009 (UTC)

Terrell-Penrose rotation

Should we have a brief mention of this subject in 'Practical effect of the finite speed of light'? Martin Hogbin (talk) 16:00, 24 August 2009 (UTC)

This subject is found on the WP page Terrell rotation and the effect refers to the apparent geometry of a moving object. I think a mention is useful. Brews ohare (talk) 23:04, 24 August 2009 (UTC)

I don't think so - it's a relativistic effect, not a finite-light-speed one. EdwardLockhart (talk) 09:11, 25 August 2009 (UTC)

Quite the reverse. It was years before it was noticed that if we were to look at a passing, rapidly moving, object we would not see it Lorentz contracted, as predicted by relativity, but rotated (simply speaking) due to the combined effect of the Lorentz contraction (the expected relativistic effect) and the time delay in light reaching our eyes from different parts of the object (a simple and easily calculated effect due only to the finite speed of light). It is similar, in principle, to aberration of starlight. Martin Hogbin (talk) 10:35, 25 August 2009 (UTC)

So it is. Yes, I think it would make sense to include it. EdwardLockhart (talk) 21:40, 25 August 2009 (UTC)

I have added a paragraph on the subject. Martin Hogbin (talk) 22:28, 25 August 2009 (UTC)

There are some nice simulations of this effect. Should we add a link to one? Martin Hogbin (talk) 22:30, 25 August 2009 (UTC)

Thanks for the copyedit Dick. Just one thing, I really do not want to get into a discussion of whether it is Terrell, Terrell-Penrose or Penrose-Terrell rotation, the various names are mentioned in the linked article. Can we pick just one name, I do not care much which, and stick to it here. Martin Hogbin (talk) 08:47, 26 August 2009 (UTC)

Approximate value of c in first paragraph

Quote from above discussion:

Although my own preference was to start with the approximate value in the first paragraph, since all that many readers will want to know. The exact value, and the fact that it is exact, is an extra complication that I'd just as soon defer. But I think I lost that argument already. If someone wants to hold another referendum, we can revisit that; otherwise, let's don't. The best we can hope for at this point is to satisfy everyone except Brews and David, I think. Dicklyon (talk) 02:03, 25 August 2009 (UTC)

I believe this view, to start with an approximate value of c, to be the best compromise I have seen. It avoids sparking incredulity among readers induced by startling claims about defining the metre so that the speed of light cannot be measured in SI units, and allows this topic to be deferred to the section on "Speed of light by definition" where there is room to go into matters.

I see absolutely no necessity for introducing the number 299,792,458 m/s in the lead, and in fact many discussions of speed of light simply use an approximate value, because the main point is that it is fast. See Google search.

I believe the desire to introduce 299,792,458 m/s at bottom stems from a misplaced notion that this number has fundamental significance. The underlying reason for this belief is that it is referred to as an exact value, which confers a mystical awe among some editors. This despite the fact that 299,792,458 m/s is an arbitrary value, exact only because it is defined by committee as a conversion factor between length and time-of-transit. Half a dozen sources have been cited and quoted that say this, and explicitly indicate the lack of any mystique associated with this number. It could just as well be 300, 000, 000 m/s. Brews ohare (talk) 11:32, 25 August 2009 (UTC)

Congratulations! You just confirmed Dicklyon's phrase "The best we can hope for at this point is to satisfy everyone except Brews and David, I think."

Now please stop harrassing this talk page? Thank you. - DVdm (talk) 12:09, 25 August 2009 (UTC)

Your input is not relevant to the issues outlined above. You seemingly are unable to distinguish between harassment and discussion. Of course, to participate in discussion one must be able to articulate a point of view, not just rely upon hostility. Brews ohare (talk) 12:27, 25 August 2009 (UTC)

There is nothing more to discuss here, Brews. We have two definitive sources BIPM and NIST stating: 'The meter is the length of the path travelled by light in vacuum during a time interval of 1/299 792 458 of a second', and Note that the effect of this definition is to fix the speed of light in vacuum at exactly 299 792 458 m·s-1.

This is clearly what must go in the article. If you want to discuss why this decision was made, the science behind it, why you do not agree with it, or your bizarre interpretation of the three sources which repeatedly quote then please do so on my talk page. Martin Hogbin (talk) 16:45, 25 August 2009 (UTC)

Martin; Of course BIPM and NIST say that, but they are talking about the SI system of units within which c refers only to the conversion factor between transit time and length. This is a classic case of a term being used in two senses: the speed of light as a fundamental constant of nature and the speed of light as a conversion factor in the SI units. Please consult Jespersen and Sydenham and Wheeler as I have suggested to you innumerable times without success. They point out that 299 792 458 m/s is an arbitrary value, a conversion factor set by committee decision that is not a fundamental constant of nature. Of course the actual speed of light is a fundamental constant of nature and has a specific value, not an arbitrary one. As Tombe, Brews-ohare and Abstract have said over and over again, you cannot measure this fundamental constant of nature within the SI system of units in units of m/s, because it is a defined value. You could, however, measure it in units of wavelengths per second. Brews ohare (talk) 19:26, 25 August 2009 (UTC)

Brews the definitive sources are quite clear. I will only continue discussion on my talk page. Martin Hogbin (talk) 21:46, 25 August 2009 (UTC)

First, your definitive sources (NIST and BIPM) are only one source, as both sources simply quote the same 1983 committee determination. Second, this 1983 determination is specified by this source as placing 299 792 458 m/s out of the reach experimental determination as "never to be measured again". Third, the three sources Wheeler, and Sydenham, and Jespersen classify 299 792 458 m/s as arbitrary, a conversion factor and not a fundamental constant of nature. Thus, reference to 299 792 458 m/s is clearly not a reference to the fundamental constant of nature, and yet still is called in the SI units context the speed of light.

Is there any reason to impugn these three sources? I'd say they are at least as definitive as the committee source, and more general as they discuss the speed of light in a broader context than the SI units conversion factor. Brews ohare (talk) 00:34, 26 August 2009 (UTC)

It has more information because your comparing things. "X units" doesn't tell you anything unless you compare it to something else. "My car moves at 10 m/s" tells me that your car is moving 10 299792458th of how fast light is moving. "This particle moves at 299792458 m/s" tells me that your particle moves as fast as light moves. Likewise "The chunk of metal in Paris has a mass of 1 kg" doesn't tell you anything. Headbomb {_{κοντριβς} – WP Physics} 19:49, 26 August 2009 (UTC)

What the introduction says (or doesn't)

While I agree with Brews that it would be better to start with the approximate number, I don't agree with his reasoning or logic for why. The present lead says "In physics, the speed of light is a fundamental physical constant, the speed at which light and all electromagnetic radiation travels in vacuum. It is usually denoted by the letter "c". In the International System of Units, the metre is defined so that c has the exact value of 299,792,458 metres per second." which certainly does not suggest to anyone that 299,792,458 is a fundamental constant. And I agree that this number has to be in the article, and almost certainly in the lead; I just don't think that the somewhat distracting fact of the defined value should appear in the first paragraph. Many readers would be better served the way many books do this (not like Brew's sources, but some gentler sources), starting with the approximation that a reader can remember and use. Dicklyon (talk) 03:53, 26 August 2009 (UTC)

Let me understand this carefully: WP says: the speed of light is a fundamental physical constant ... usually denoted by the letter "c". Moreover, in the SI units, "c" has the exact value of 299,792,458 metres per second.

However, these WP statements (according to Dicklyon) in no way: suggest to anyone that 299,792,458 is a fundamental constant. Well tell me about the way this conclusion flows from the WP statements, please. Will the reader who is not a Philadelphia lawyer tease out this conclusion from the premises? The WP intro presents incorrect information. Brews ohare (talk) 05:38, 26 August 2009 (UTC)

So you're saying a speed being a fundamental constant of nature implies that its numerical value in SI units is a fundamental constant? Yes, I can see why that would confuse you, since you think of it differently. Oh, well, we'll have to live that that duality. Dicklyon (talk) 05:44, 26 August 2009 (UTC)

Dick: Not only does your statement completely misstate my stance, which I will explain further below if you wish, but you have not dealt with your incorrect statement about the implications of the WP introduction. The WP intro presents incorrect information, despite your remarks. Brews ohare (talk) 13:32, 26 August 2009 (UTC)

To spell out the misinformation: WP says by implication: 299 792 458 m/s is a fundamental constant. The sources Wheeler; Jespersen; Sydenham say explicitly the contrary. Such errors certainly will impede attainment of featured article status for this page. Brews ohare (talk) 13:57, 26 August 2009 (UTC)

h is a fundamental constant of nature. 6.62606896×10 isn't. Should we rewrite the Plank constant article because we state that h is a fundamental constant equal to (6.62606896±33)×10 J·s in SI units, out of fear that readers will think that 6.62606896×10 is a fundamental number? h and c are as fundamental as the other, both are scaling factors. Accept that you are the minority view here Brews, and move on. Headbomb {_{κοντριβς} – WP Physics} 15:03, 26 August 2009 (UTC)

Headbomb: There is no majority or minority opinion involved here: WP states that 299 792 458 m/s is the exact value of a fundamental constant in contradiction to sources Wheeler; Jespersen; Sydenham. That is neither opinion nor conjecture. Brews ohare (talk) 15:23, 26 August 2009 (UTC)

These sources say that the numerical value itself is arbitrary and not "fundamental", not that the physical concepts upon which they are based on cease to be "fundamental". You can define c or h to be whatever you want (other than 0 or infinity), all you're doing is scaling units ('cause you know, these are fundamental constants). Quantization doesn't stop because you've set h to be 10×10 instead of 10×10. The only fundamental constants which aren't scaling factors are dimensionless constants such as G_F and α, and we have a name for those: dimensionless constants. Headbomb {_{κοντριβς} – WP Physics} 17:25, 26 August 2009 (UTC)

Of course there is opinion involved. Brews believes the sound bytes carelessly quote-mined from the sources support his fringe viewpoints/conclusions, and most others do not. The Wheeler quote, for example, comes from a book where the predominant unit system used has c=1. Such a system having c=1 uses c as an exact value, without error bars. The idea that Brews is the only one using logic or arguing from sources is a gross violation of good faith. Brews' opinion is a minority view, is against current consensus, and his continuing tendentious editing is the problem here. Tim Shuba (talk) 18:07, 26 August 2009 (UTC)

I've yet to see more than private opinion on your side. The only sources you quote are repeating the 1983 committee decision, which cannot settle the matter as it is confined to the SI units system.

Apparently it is hairsplitting for you guys. Of course the underlying physical entity "speed of light" that occurs in physical theory such as the theory of relativity is fundamental. That was never an issue. It also happens to be true that this speed of light has a real physical value in (say) wavelengths of some transition /s, and that this value can be measured and does have an error bar. In contrast, as I am sure you agree, the so called speed of light in SI units is a convention, can have any value whatsoever. It is not measured, and has no error bar. The metre is adjusted to make this value of so-called speed of light to whatever you choose. The only experimental input to the metre is just a transit time of light. There is no fundamental length unit in the SI system. In no way can this arbitrary number for so-called speed of light be taken as indicative of the physical entity "speed of light" because the metre definition precludes ever getting a value different from the defined value. The units system is "elastic". It cannot express the concept of length traversed per unit time, as there is no length input in this system, only time input.

To put it differently, the speed of light in (wavelengths of some atomic transition) / (transit time) contains more physical information than does the statement of the speed of light in m/s. Brews ohare (talk) 18:39, 26 August 2009 (UTC)

I think missing from all this argument is *why* the meter is defined this way - it's precisely *because* the speed of light appears (by all experimental evidence) to be a real, universal, physical constant. So the two uses of c, the fundamental physical constant and the 'conversion factor' are not independent - the second was a direct result of the first. For example, a similar definition of the meter would not work at all if the speed of sound was used instead of the speed of light. LouScheffer (talk) 02:37, 27 August 2009 (UTC)

I'd agree with that, Lou. The fact one is permitted to use a conversion factor to relate length to time of transit is a result of the confidence with which the speed of light can be realized, even if you do not know what that speed is, and make no attempt to measure it. (One complication is realizing "vacuum".) However, the numerical value of this conversion factor is arbitrary, and is independent of whatever the actual speed the speed of light may have in vacuum. Brews ohare (talk) 05:09, 27 August 2009 (UTC)

I agree with Lou's too. We could say more about why the metre was defined the way that it was, but anything we say must reflect current mainstream scientific thinking, as described in reliable sources, in the subject. We do have the reasons that the standard was changed in 1983, from the people who actually made the change, see my reply to Edward below. Martin Hogbin (talk) 11:37, 27 August 2009 (UTC)

You are of course correct that "the speed of light in (wavelengths of some atomic transition) / (transit time) contains more physical information than does the statement of the speed of light in m/s". With definitions expanded, the former is a comparison of two different physical phenomena, the latter is a statement of a conventional scaling factor.

Given the post-1983 definitions, the physical content that is contained in the first statement is now expressed in the statement "the wavelength is x m", which previously would have been an exact statement of a conversion factor, with no physical content. EdwardLockhart (talk) 05:37, 27 August 2009 (UTC)

Actually, Edward, the pre-1983 definition of the meter contains exactly the same information as the later one, or at least the 17th Conférence Générale des Poids et Mesures thought so (See point 7 in this link ]). The specified atomic transition of krypton emits light of a particular frequency. The wavelength of this light depends on the speed of the light, thus the specified krypton transition was actually being used as a frequency standard, with the meter effectively being defined as the distance light travels in a specified number of periods of the krypton radiation. Later advances in technology meant that the frequency of the krypton radiation could be measured against the existing caesium frequency standard. By stating the time of travel in seconds rather as a number of periods the krypton radiation it became possible to use other, more stable, sources of light, and the existing caesium time (frequency) standard to define the metre. You can read about it in this link kindly provided by Brews ]. Martin Hogbin (talk) 11:15, 27 August 2009 (UTC)

I think you've misread ]. Point 7 says that the various *proposed* standards mentioned in point 6 are equivalent, not that the new standard is equivalent to the old one. I believe the following two pairs of statements are equivalent:

Pre-1983 - the speed of light is approx 299,792,458 m/s, the wavelength of is exactly 1/1,650,763.73 m.

Post-1983 - the speed of light is exactly 299,792,458 m/s, the wavelength of is approx 1/1,650,763.73 m.

In each case, one statement is a definition of our length unit (but phrased in a curious way), and the other statement is a physical observation using that unit. The 1983 definition change flipped the roles of these two physical phenomena. And in so doing, it made a philosophical (if not a practical) difference to the statement of the speed of light in SI units. EdwardLockhart (talk) 12:04, 27 August 2009 (UTC)

I was referring to this statement, 'that these various forms, making reference either to the path travelled by light in a specified time interval or to the wavelength of a radiation of measured or specified frequency, have been the object of consultations and deep discussions, have been recognized as being equivalent and that a consensus has emerged in favour of the first form'. They are clearly saying here that using the wavelength of a specified radiation and using the time travelled by light in a specified time are equivalent. Martin Hogbin (talk) 13:50, 27 August 2009 (UTC)

I agree with EdwardLockhart that equivalent to the BIPM means equivalent within the context of length comparisons using the SI units, not physical equivalence. As the actual, physical speed of light simply cancels out in any length comparison:

${\displaystyle {\frac {\ell _{1}}{\ell _{2}}}={\frac {c\ t_{1}}{c\ t_{2}}}={\frac {t_{1}}{t_{2}}}\ ,}$

any length comparison is "equivalent" to a "time-of-transit" comparison. The "fundamental discussion" for the BIPM is simply whether the length ratio or the time ratio has the greater error bar. Brews ohare (talk) 12:43, 27 August 2009 (UTC)

BIPM statement

Martin has brought up some sources for discussion. One of these sources is BIPM, which states:

that these various forms, making reference either to the path travelled by light in a specified time interval or to the wavelength of a radiation of measured or specified frequency, have been the object of consultations and deep discussions, have been recognized as being equivalent and that a consensus has emerged in favour of the first form

The metrology advantage of replacing length comparisons with time-of-transit comparisons is that the measurement of time-of-transit involves only a time measurement error, and this error is small compared to length comparison based on fringe-count (interference pattern) comparisons.

Of course, the BIPM is a metrology organization concerned with the SI system of units, that is, with practical criteria of convenience and upon accuracy sufficient within the SI system of units, and unconcerned with "equivalence" of methods in the more strict sense of physical law. They are concerned with only length comparisons and are not concerned at all with the actual speed of light, which is unnecessary to their chosen time-of-transit methodology.

In other words, use of a conversion factor, chosen arbitrarily as 299 792 458 m/s but equally any other number, simply reduces length measurement to being identically the same as measurement of transit time. The actual speed of light is not necessary, and indeed is not even a knowable (measurable) within this approach. It is simply a defined value, arbitrary, chosen for convenience. Brews ohare (talk) 12:32, 27 August 2009 (UTC)

NIST document

Martin has brought up the source NIST paper for discussion. Martin has suggested that this source supports this view expressed by Martin: "By stating the time of travel in seconds rather as a number of periods the krypton radiation it became possible to use other, more stable, sources of light, and the existing caesium time (frequency) standard to define the metre."

This statement of Martin's is not (to me) terribly clear. What tipped the scale to comparisons of "time-of-transit" for comparing lengths, rather than comparison of "counted interference fringes" was that time measurement errors became much smaller than fringe-counting errors as technique improved.

As the actual, physical speed of light simply cancels out in any length comparison:

${\displaystyle {\frac {\ell _{1}}{\ell _{2}}}={\frac {c\ t_{1}}{c\ t_{2}}}={\frac {t_{1}}{t_{2}}}\ ,}$

any length comparison is "equivalent" to a "time-of-transit" comparison.

It is totally irrelevant what the actual value of c might be, so long as one can be sure that the same value appeared in both lengths. Consequently, one is free to choose any value for c whatsoever. In addition, one never needs to measure the value of c .

A consequence of this

definition is that the speed of light is now a defined

constant, not to be measured again. NIST paper

The definition of the metre by BIPM and NIST is really not a length definition at all, but the identification of a standard time of transit namely t_standard = 1/299 792 458 s. When one says a length is so many metres, one really is saying the length has a transit time in vacuum of so many t_standard units. Brews ohare (talk) 12:55, 27 August 2009 (UTC)

When you add the unit of metre you are putting the definition of c back into the value for distance.

Thanks for the quote from NIST; it shows that one credible body, at least, believes that c is defined as exactly 299 792 458 m/s. This is a credible enough source that I think we can include that value in the lead now without any objections. TStein (talk) 14:12, 27 August 2009 (UTC)

TStein: A bit more care is needed here before rushing ahead with this quote. This whole mess is just a result of the confusion engendered when the literature uses the same exact wording for different concepts. It happens again and again. Examples are centrifugal force and Faraday's law and electromotive force. In this case, BIPM refers to an arbitrary conversion factor with absolutely no physical significance as the speed of light, while the structure of spacetime also uses the notion of speed of light in an entirely different way.

By introducing wavelength, a true length is invoked. On the other hand, the present-day metre is actually not a length, but a new name for a standard transit time of 1/299 792 458 s. This time is converted to a length by a fictitious dimensional conversion factor.

The key to this jumble is that the actual, physical speed of light simply cancels out in any length comparison:

${\displaystyle {\frac {\ell _{1}}{\ell _{2}}}={\frac {c\ t_{1}}{c\ t_{2}}}={\frac {t_{1}}{t_{2}}}\ ,}$

any length comparison is "equivalent" to a "time-of-transit" comparison. It is totally irrelevant what the actual value of c might be, so long as one can be sure that the same value appeared in both lengths. Consequently, one is free to choose any value for c whatsoever. Obviously, the same is not true of the relation:

${\displaystyle \ell =c\ t\ ,}$

so length (e.g. atomic wavelength) depends critically upon the actual speed of light, a matter no longer of any concern to NIST or BIPM, at least wrt the SI units. Brews ohare (talk) 14:50, 27 August 2009 (UTC)

Third paragraph

What does this paragraph add to the lead? It seems to me that it is of passing interest only and should be relegated to the body of the article. Abtract (talk) 20:34, 25 August 2009 (UTC)

The lead is meant to be a summary of the article. In this case it does seem that the lead has some odd points in it that are not covered in the article. So, I agree in general. The questions are, where should the information now in the lead be put and should we have a very brief summary of it in the lead? Martin Hogbin (talk) 21:52, 25 August 2009 (UTC)

Google summary

A google search for speed of light returns this page (good), but the extract is: "Here, laser light in air is traveling at about 99.97% the speed of light in a vacuum (the refractive index of air being about 1.0003). ..." (bad). I assume because this is the first mention in the article text. I tried moving the picture & caption to after the lead paragraph, but that made the layout worse. Does anyone know how to fix this? EdwardLockhart (talk) 04:24, 26 August 2009 (UTC)

I don't think that "fixing" the google snippets is among the wikipedia editing guidelines; leave it alone. Dicklyon (talk) 04:29, 26 August 2009 (UTC)

Lead picture

This seems to add nothing to the article except maybe confusion by talking of light travelling at  % of the SoL. IMHO we would be better with no pic than this one. Abtract (talk) 21:55, 26 August 2009 (UTC)

We need a picture of some sort. I see nothing wrong with the current one. Maybe it is the caption that you do not like. This is also the cause of the problem above. Martin Hogbin (talk) 22:53, 26 August 2009 (UTC)

I think the caption is good, as it clarifies that, in air, light travels at less than the speed of light. Or maybe it should say 99.97% of c if that's more clear. Dicklyon (talk) 23:54, 26 August 2009 (UTC)

I see no reason why we need a picture of some sort (Speed of sound has none in the lead) but, even assuming we do, why on earth would we choose one that showed light travellling at less than c? I much better picture imho would be of a celestial object with the statement that "light takes xx years to travel from the zzz galaxy to earth" or maybe of the moon or sun with a similar caption. An alternative would be a picture of the equipment used in one of the historic experiments to measure the SoL. Abtract (talk) 08:35, 27 August 2009 (UTC)

I think we need a lead image for FA status. I am not fighting for this particular picture, if you can fine a better one or some candidate images the please present the here where we can argue endlessly about them ;-) Martin Hogbin (talk) 11:20, 27 August 2009 (UTC)

I have replaced it with a picture of the sun, the source of our natural light. Abtract (talk) 12:59, 27 August 2009 (UTC)

I like that. However, in order to hint at the article's subject (a speed), lets replace the caption "Light takes 499 seconds to travel from the Sun to the Earth" with "Light takes about 500 seconds to cover the 150 million Km distance from Sun to Earth". DVdm (talk) 13:10, 27 August 2009 (UTC)

Why not, Travelling at 299,792,458 m/s, light takes about 500 seconds to cover the 150 million km distance from Sun to Earth, This get round the problem that Edward noticed. Martin Hogbin (talk) 13:42, 27 August 2009 (UTC)

Excellent. I gofered it. DVdm (talk) 14:36, 27 August 2009 (UTC)

By the way, an even better picture would be one where we see Earth on the foreground and a distant Sun on the background, like this... DVdm (talk) 14:48, 27 August 2009 (UTC)

Excellent; I like that useful picture and the meaningful caption but while we are rounding we may as well do a complete job. Abtract (talk) 15:28, 27 August 2009 (UTC)

I'd vote for the exact value, but I wouldn't have a fight over it ;-) - DVdm (talk) 15:35, 27 August 2009 (UTC)

The new "L2" illustration is terrible here, as it's dominated by graphical information that's off topic and goes unmentioned. Dicklyon (talk) 16:23, 27 August 2009 (UTC)

Let's edit the L2 part out of it or find a better one. Here is the edited version. I don't know how to add it as another alternative version at - DVdm (talk) 16:52, 27 August 2009 (UTC)

I edited the picture (originally from a govt source) and enlarged the titles. See what you think. Brews ohare (talk) 14:15, 28 August 2009 (UTC)

Exactly what I had in mind :-) DVdm (talk) 14:17, 28 August 2009 (UTC)

introduction to the speed of light

may a make the same observation i made some time ago in the torque article . this reads like a thesis for a physics degree and is not very interesting to most people . i would suggest to make the introduction readable and understandable for a ten year old , pretend it is your child who wants some information and keep the highly sophisticated stuff for the rest of the article .

Wdl1961 (talk) 15:03, 27 August 2009 (UTC)

May I suggest that you use proper capitalisation and punctuation in your talk page messages? Imagine the influence your way of writing might have on these kids. DVdm (talk) 15:13, 27 August 2009 (UTC)

Wdl1961:How about some concrete suggestions on how to simplify the lede.

Remember that this article is primarily about the constant c which relates space and time and is not simply the speed at which light travels. For that reason the inherent difficulty level of this article is akin to that of special relativity; I am not quite sure how advisable it would be to write the lead of special relativity so that it is understandable by a 10 year old. The same is true for this article. If you can do it or make a suggestion that works then more power to you, though. TStein (talk) 15:26, 27 August 2009 (UTC)

The best approach so far to simplifying the lead is that of Dicklyon: introduce the speed of light with an approximate value; avoid startling terms like exactly 299 792 458 m/s and the concept of defining the metre to make this so; and follow the lead with the "Speed of light by definition" subsection where matters can be discussed carefully without severe length limitations. Brews ohare (talk) 15:54, 27 August 2009 (UTC)

I am trying a rewrite on my name page.

One suggestion is to rename it as "C" the speed of light Abtract (talk) 16:00, 27 August 2009 (UTC)

You might find it more convenient than replacing your user page to create your draft article by making a new page User:Abtract/Speed of light which you can link to your user page by inserting this link at the top of your user page. Brews ohare (talk) 16:09, 27 August 2009 (UTC)

Proposed new lead section

My suggested rewrite can be found here (I don't use my user page for anything else but thanks for the idea). IMHO this is more inline with wp:lead, is more user friendly (tells a story) whilst still being scientific, and brings all(?) the salient points to the fore. Whether it will bring together the two sides of the divide remains to be seen but it clarifies the circularity problem. Naturally it needs refinement even if you have some liking for it. Abtract (talk) 17:11, 27 August 2009 (UTC)

It looks acceptable to me. I'd suggest using something like "In physics, the speed of light 'c' " as being more in keeping with a standard intro line, and allowing the use of a lower case c. Brews ohare (talk) 17:32, 27 August 2009 (UTC)

Personally I prefer to leave out "In physics" because that somehow seems to diminish the importance of the SoL, but let's see what others say. I do agree with the lower case c so I have reordered the beginning and maybe the article might be better named The speed of light ... "c"Abtract (talk) 08:36, 28 August 2009 (UTC)

"For many years thought to be infinite," - I think it would be more accurate to say it wasn't known if it was infinite or simply very fast. Or even for that matter whether it made sense to talk of a speed at all. EdwardLockhart (talk) 05:55, 28 August 2009 (UTC)

I have changed it along those lines. Abtract (talk) 08:36, 28 August 2009 (UTC)

Your proposal now says that the speed of light was previously assumed to be instantaneous. If you read the relevant section of the article| it should be clear that this is not the case - from Empodocles onwards, some thinkers on the subject have thought that light takes some time to travel. Equally others, such as Aristotle, didn't believe there was any movement going on, meaning there would be no such thing as a speed. EdwardLockhart (talk) 08:46, 28 August 2009 (UTC)

OK it is quite difficult to get the idea across in a few words but I will try again. Abtract (talk) 10:26, 28 August 2009 (UTC)

No way The current lead is the result of pages of discussion, cooperative editing, and a considerable degree of consensus amongst editors here. There is no possible benefit to our readers in not giving the exact speed of light at the start of the article, and this seems to be a plan to avoid doing that. The only reason for not wanting to give the exact speed is that, despite extensive discussion, some editors still do not understand why the speed is now exact. I do not think the style of the proposed lead is terrible encyclopedic and it contains a number of dubious or unclear statements. No doubt it could be improved but that would then start the whole cycle of editing and arguing all over again.

I am not against careful rewriting of the lead to improve the English and style and but the content has been discussed in great detail and I suggest that it should not be changed dramatically like this. The content should evolve slowly as a result of discussions between editors to reflect consensus views. Martin Hogbin (talk) 09:30, 28 August 2009 (UTC)

I don't know how to respond to Hogbin's accusatory dismissal of my proposal except to say I thought it was a "careful rewriting". There is no "plan" just a reflection of my, and I thought others', opinion that starting with a very close approximation would be much easier on the eye and therefor more reader friendly, and by then taking the reader gently through a brief history we could lead them to the exact figure and explain how it arose. The final para explains the importance of c. All this imho is what a lead should do and seems to accord with wp:lead. I await other comments with interest. Abtract (talk) 10:26, 28 August 2009 (UTC)

Sorry Abtract that my remarks appeared accusatory. Let me explain the individual points that I was trying to make.

There was a very long discussion on what numerical value we should show in the lead followed by an RfC. There was there a clear consensus to show the exact value first (I believe that this same consensus should apply to the picture caption). The two editors who were against doing this took that view because they believed that the exact value was, in some way wrong, a position clearly not in accordance with the generally accepted scientific view. I proposed adding that 300,000,000 m/s was often used as a convenient approximation, but it seemed that most editors were against this. For many readers there is a simple question, 'What is the speed of light', and to this question there is a simple answer, '299,792,458 m/s'. I cannot see how anything is made simpler by saying it is about 300,000,000 m/s followed some while later with 'well actually it is '299,792,458 m/s'. For those interested in why the value is exact (many may have no interest at all in this fact) we have a section devoted to that subject in the article.

We should be trying to improve this article by cooperative editing. This cannot be achieved by every editor rewriting the lead in their own words. We all have different talents and it is better to try to use them together to produce something better than any individual could have done. Martin Hogbin (talk) 14:16, 28 August 2009 (UTC)

I'm in broad agreement with Martin - the exact value of c in m/s should be at the top. This was the concensus of lengthy previous discussion, which I think you missed. I'm not convinced either that so much history of mistaken ideas and imprecise estimates should be in the lead - other articles on scientific topics don't generally take this approach. EdwardLockhart (talk) 13:11, 28 August 2009 (UTC)

The proposed lead by Abtract in its original version contained the SI units conversion factor exactly, as seems to be desired by EdwardLockhart & Martin Hogbin, and contained the wording:

For technical reasons in 1983, the metre was redefined using this value as a conversion factor. This had the effect of 'fixing' c at exactly 299,792,458 metres per second in the International System of Units.

This wording is accurate. The present wording:

In physics, the speed of light is a fundamental physical constant, the speed at which light and all electromagnetic radiation travels in vacuum. It is usually denoted by the letter "c". In the International System of Units, the metre is defined so that c has the exact value of 299,792,458 metres per second.

juxtaposes two different usages for the term "speed of light" without distinguishing between them, and is in fact incorrect on that account. More detail on this point is found at NIST document and at Some math.

The first form of wording by Abtract should be adopted. Brews ohare (talk) 14:13, 28 August 2009 (UTC)

We cannot pander to the views of those who do not understand the consequences of the current definition of the metre. The page must reflect current mainstream scientific thought on the subject. Martin Hogbin (talk) 14:16, 28 August 2009 (UTC)

Martin Hogbin: The word "pander" is, of course, inflammatory. You make the further unsupported claim that you understand the implications of the SI units conversion factor.

The problems with your view of this factor have been carefully explained to you here, and on the page you created for discussion of this point, and you have not succeeded in supplying any source or argument that supports your views.

What you have done is continually reiterate your unsupported view, namely, that the SI units conversion factor is the numerical value of the physical constant called the speed of light throughout special relativity and the rest of physics. You have never advanced argument or sources, not attempted to meet sourced objection. You do not understand the 1983 BIPM definition of the SI units conversion factor, nor its implications, nor its limitation to the SI units. You have not addressed the three sources by Wheeler, Sydenham or Jespersen that refute your viewpoint . Brews ohare (talk) 14:38, 28 August 2009 (UTC)

I did indeed miss it and, although I have looked at some of the more recent discussions, there seems to be so much heat generated that I have tried to put together a lead which, as it should be, is a summary of the article, brings out the salient points and makes the reader want to know more. IMHO the salient points are: 1) how we reached the counterintuitive conclusion that light has a speed 2) some idea of how the SoL is measured 3) what the measured speed is (300,000 km/s will be a good degree of accuracy for 90% of readers and is much simpler to read), 4) how we came to have a fixed exact figure 5) why c is so important and in what fields. All these points are to be found in the article in much more detail ... I don't think I have missed any important points. One more point on the use of 300 early in the lead with the exact figure saved for after the story has unfolded ... it isn't as though 300,000 km/sec is incorrect, it is the correct figure stated to three significant figures. I quite fail to see why showing c to nine significant figures in the first paragraph should arouse such high feelings. Pi is approximated in the lead as 3.14159 but given much more accurately later in the article; interestingly e is not shown as a number at all until the 4th paragraph and the Gravitational constant is first shown in numeric form at the end of the 3rd paragraph; Planck constant is not mentioned numerically in the lead, being relegated to a side box; the Fine-structure constant doesn't really count because the lead is only one very short paragraph; there are two with a high degree of accuracy early in a decent sized lead Elementary charge (1st para) and Vacuum permittivity (2nd para). All this demonstrates that other editors in other similar articles have taken a pragmatic approach and used approximations and more accurate figures where they fitted the particular story ... that is what I have tried to do, hoping it would satisfy the warring factions here. This is not pandering. Abtract (talk) 14:36, 28 August 2009 (UTC)

Your present proposed lead is acceptable to me. You appear to have dropped a reference for the speed of gravitational waves: Wei-Tou Ni (2005). "Empirical foundation of the relativistic gravity" (PDF). Intl J Modern Physics D. 14: 901–921. Brews ohare (talk) 15:41, 28 August 2009 (UTC)

So why not just restore the reference? Martin Hogbin (talk) 16:12, 28 August 2009 (UTC)

Gentlemen, according to current, generally accepted, mainstream scientific thinking there is only one speed of light and it is exactly 299,792,458 m/s. Pi and e are transcendental numbers and must always be approximated in any finite numerical representation. Martin Hogbin (talk) 16:10, 28 August 2009 (UTC)

Martin Hogbin: Abtract's present lead is accurate and appears to me to meet all your requirements. Why, specifically & in exact detail, do you not wish to adopt it? Brews ohare (talk) 16:16, 28 August 2009 (UTC)

Since no-one seems to have any objection tomy proposed lead except for the rounding in the first para, I am going to be bold put it in. The only change I will make is that I will put the exact figure in the first para but without the word "exactly" which I hope may 'satisfy' both sides (although why there are 'sides' in this beats me). Abtract (talk) 09:41, 29 August 2009 (UTC)

I have a strong objection to your lead and I have clearly stated it above. It starts with No way! Martin Hogbin (talk) 10:22, 29 August 2009 (UTC)

Yes so it appeared at first sight but, having overcome my shock at the tone, I could only find one actual point - the use of an approximation - I have altered that. What more do you require? Abtract (talk) 10:57, 29 August 2009 (UTC)

As I have just said in reply to Tim Shuba it is not that I think your lead is that bad but there are a few things about it that need changing. However, there is absolutely no point in me or any other editor doing this because the likelihood is that someone else will rewrite it from scratch again. If this is going to be the ethos of editors here than we will never reap the benefits of cooperative editing. What I would like is for you to discuss exactly what was wrong with the original lead (which was far from perfect) so that we can all work together to improve it.

It is generally true that a lead written by one person is likely to read better than one edited by several people over a period of time. I would therefore suggest that we should finalise the content and order of the lead and then, when that is agreed, one person should rewrite it to improve the style and presentation. Martin Hogbin (talk) 11:06, 29 August 2009 (UTC)

Apologies, I thought I had made it clear previously ... the previous lead was not bad but it failed in a couple of key aspects imho 1) it did not bring out the history suffieciently well (I have expanded it), 2) it overemphasised the "exactness" too early and failed to explain in properly later (this is all tied up with my 'circularity' point and ohare's point too I believe), 3) it did not 'read' very well (maybe your point about one editor v many). If you compare the two versions, I haven't actuallt changed it much just sharpened it up , reordered slightly and extended the history including clarifying the exactness point ... and, maybe important to you,I have delayed using the word exact until later in the piece. Abtract (talk) 11:40, 29 August 2009 (UTC)

So what exactly is you 'circularity point'? Martin Hogbin (talk) 12:30, 29 August 2009 (UTC)

As it was the last time we discussed it: defining the metre in terms of the SoL and then stating the SoL in units of m/s says absolutley nothing about the SoL (except that we all have in our mind a vision of the metre as previously defined independently of the SoL). I don't think we can escape this circularity but we can 1) not bother the reader with it too soon, and 2) explain it at the end of the historical story so that it makes some sort of sense in context. I have tried to do both. I have no axe to grind in this matter except producing a readable, useful and scientifically correct article ... Abtract (talk) 13:27, 29 August 2009 (UTC)

If the metre is defined in terms of the second and the speed of light, an obvious and inevitable consequence of this definition is that the speed of light will have a fixed numerical value when measured in metres per second. If this is what you meant by 'circularity problem' then yes, this is true, get over it! This is not a world shattering revelation or a dark secret revealed only to the chosen few, it is a fact that is quite obvious to nearly everyone.

If you meant that nobody knows roughly how long the meter is, it is about this long , so from an intuitive perspective giving the speed of light in metres per second is no different from giving it in furlongs per second; you have to know roughly how long a furlong is for the value to have any intuitive meaning.

If you mean that nobody knows exactly how long a meter is then you are right. All issues of experimental error and imprecision manifest themselves in the practical realization of the meter. Mind you, they can get pretty close these days.

If you are going to assert that there is a 'circularity problem' with the exact speed of light you must find a reliable source that states, in plain English, that there is a circularity problem (or indeed a problem of any kind) with the current definition of the metre. You have seen the statements from the world and US standards authorities stating the exact value of the speed of light and they are quite clear. If you want to argue with this you must find a source that plainly and simply disagrees with what BIPM and NIST say. Bizarre interpretations of sentences plucked from good sources are not acceptable. Martin Hogbin (talk) 13:52, 29 August 2009 (UTC)

I am really sorry that you do not understand my point or indeed the topic, as evidenced by "If the metre is defined in terms of the second and the speed of light, an obvious and inevitable consequence of this definition is that the speed of light will have a fixed numerical value when measured in metres per second" which would only be true if you substituted "stated" for "measured". I think, under these circumstances, we had better stick to discussing content. What changes would you like to see to the current lead, and why? Abtract (talk) 14:21, 29 August 2009 (UTC)

You are quite right, 'measured' was a poor choice of words. Change it to 'stated'. Now, what was your problem? Martin Hogbin (talk) 14:25, 29 August 2009 (UTC)

I don't have a problem ... what changes would you like to see? Abtract (talk) 14:33, 29 August 2009 (UTC)

I would like to see it restored to its earlier consensus version which said, 'In the International System of Units, the metre is defined so that c has the exact value of 299,792,458 metres per second'. How does your 'circularity' point apply to this?

Imho it is best not to burden the reader with the counter intuitive "exact value" until it can be placed in context later in the lead. According to wp:lead "The first paragraph of the introductory text needs to unambiguously define the topic for the reader, without being overly specific". I believe the words "exact value" are "overly specific". Perhaps you could explain why it is important to you to include those words in the first para?Abtract (talk) 16:28, 29 August 2009 (UTC)

within the SI system of units

User Headbomb removed the phrase within the SI system of units from the following: "The effect of this definition is to fix the speed of light in vacuum at exactly 299,792,458 m/s; thus, within the SI system of units, the speed of light is now a defined constant and can no longer be measured."

with the cryptic comment "untrue". However, the source supports the statement, and the use of a defined value for c certainly applies within the SI system, and not necessarily for other systems of units (for example, not for the pre-1983 system of units). Brews ohare (talk) 17:17, 27 August 2009 (UTC)

Mathematical elaboration

User Headbomb removed this mathematical follow up of the references by Boyes and Smolin with the comment "untrue"

"Measurement is basically comparison between the quantity being measured and the standard unit; it is a matter of ratios. Setting the speed of light to a definite numerical value when measured in the SI units of m/s simply means comparisons of length become comparisons of transit times of light. Mathematically, comparison of two lengths ℓ₁, ℓ₂ with times-of-transit of light, t₁, t₂ may be expressed as the ratio:

${\displaystyle {\frac {\ell _{1}}{\ell _{2}}}={\frac {c\ t_{1}}{c\ t_{2}}}={\frac {t_{1}}{t_{2}}}\ ,}$

which is independent of the the speed of light c."

This mathematical elaboration explains further what these sources have said, and is certainly "true". Thus, at a minimum Headbomb's deletion is poorly justified. Brews ohare (talk) 17:25, 27 August 2009 (UTC)

No it is not true, because you are not comparing transit times, you are comparing a length (of say a table) with another length (the distance covered by light in vacuum in 1 / 299972458 seconds). That l_table/l_meter = t_{light across table}/t_{light across meter} only illustrates that is it equivalent to compare transit times. Looking back reverting may have been overdramatic. Headbomb {_{κοντριβς} – WP Physics} 17:49, 27 August 2009 (UTC)

Headbomb: If t₂ = 1 / 299972458 s then ℓ₁ is in metres. Brews ohare (talk) 18:28, 27 August 2009 (UTC)

Some math

The wavelength of a given frequency of light is exactly equal to the period of that radiation multiplied by the actual, physical speed of light, which is not the BIPM arbitrary conversion factor. BIPM is interested only in length comparisons, which for a fixed speed of propagation c are transit-time comparisons (See footnote ):

${\displaystyle {\frac {\ell _{1}}{\ell _{2}}}={\frac {c\ t_{1}}{c\ t_{2}}}={\frac {t_{1}}{t_{2}}}\ ,}$

where c can be any arbitrary speed, so long as it is the same speed for both transit-time measurements. In the SI units, for example, if t₂ = 1 / 299 792 458 s then ℓ₁ is in metres.

At least insofar as the SI units are concerned, BIPM is not interested in lengths per se :

${\displaystyle \ell =c\ t\ ,}$

which length requires the actual, physical speed of light (preferably measured using 2009 techniques, rather than a 1983 estimated value), not an arbitrary conversion factor set by committee.

Footnote: Measurement is basically comparison between the quantity being measured and the standard unit; it is a matter of ratios. See Smolin Brews ohare (talk) 14:58, 28 August 2009 (UTC)

We have a consensus

I am not going to get involved in yet another edit war about the same subject - whether the speed of light is exact or not. There was a clear consensus to quote the exact sped at the start of this article and I see no reason why this does not apply to the caption of the lead image. Martin Hogbin (talk) 16:47, 28 August 2009 (UTC)

It simply is bad practice to state a rounded-off distance in the figure, indicate a rounded-off time, and then relate the two in the caption to a speed calculated to nine figures. It is stupid. Brews ohare (talk) 17:01, 28 August 2009 (UTC)

Martin Hogbin: Abtract's present lead is accurate and appears to me to meet all your requirements. Why, specifically & in exact detail, do you not wish to adopt it? That would replace your so-called "consensus" (actually a cramming down of your own wording) with unanimity. Brews ohare (talk) 17:06, 28 August 2009 (UTC)

I see the speed has now been removed from the caption. There was a reason that it was there, which is Edward's observation that this caption is the text that appears when you do a Google search on the speed of light. It was agreed that it might be useful to have the value of the speed of light in this text. I am not particularly that fussed about that but my real concern is this: The structure and wording of this article are still being influenced by a minority of editors who, despite the clear statements of impeccable sources to the contrary, insist that there is something wrong with stating the exact speed of light in m/s. I would like some support in taking a stand against this nonsense, which is still taking up pages of discussion and causing silly arguments about the wording of the article. Martin Hogbin (talk) 20:26, 28 August 2009 (UTC)

And the speed disappeared from the caption long ago. And cooperative discourse does not involve terms like "nonsense" and "silly arguments" to characterize what, in fact, are solid arguments supported by sources, not to be dealt with by rounding up a lynch mob. And we have an acceptable proposal in front of us at User:Abtract for a lead that Martin Hogbin refuses to discuss at all. Brews ohare (talk) 00:13, 29 August 2009 (UTC)

Aggressive editing spree now over

Ah, that was refreshing. I am calm now, and I will expect and enjoy any and all excoriations and reversions. Tim Shuba (talk) 04:05, 29 August 2009 (UTC)

Commentary for Tim Shuba: In the first line of your lead is the phrase: "In physics, the speed of light is a fundamental physical constant, the speed at which light and all electromagnetic radiation travels in vacuum."

I agree with that.

The second sentence is: "It is usually denoted by the letter ‘c’."

I agree with that.

The third sentence is: "In the International System of Units, the metre is defined so that c has the exact value of 299,792,458 metres per second."

From this concatenation, I'd take this 'c' to be the same as the 'c' from the first two lines. However, that is not quite so. One could say: "In the International System of Units before 1983, the metre is defined so that ‘c’ has the approximate measured value of 299,792,458 metres per second."

Or, one could say: "In the International System of Units following 1983, the term ‘speed of light’ is used to refer to a conversion factor denoted by ‘c’ that has the exact value of 299,792,458 metres per second." One could elaborate upon that point, to be sure.

However, the key here is that the ‘c’ referred to in the 1983 definition is not the ‘c’ referred to in the first two lines of your intro. In 1983, length was replaced in the SI units by "time of transit" and conversion of such times is accomplished by an exactly defined conversion factor 299,792,458 metres per second. This conversion factor, being defined, clearly lies outside any kind of measurement.

One might argue that because the metre is defined elastically to make this so, you cannot argue about the value of the conversion factor. True. However, this elastic metre that makes ‘c’ unmeasurable and exact, also removes this meaning of ‘c’ from referring to the "fundamental physical constant" of the first two sentences, which most certainly is measurable, and most certainly is not known exactly. The SI conversion factor is not the numerical value of the "fundamental physical constant", even in SI units. It is just a conversion factor.

The fundamental speed of light is made measurable by introducing a length (like a wavelength) that is not elastic.

These points have been argued more carefully on this page, and sources supplied. Feel free to delve in further.

I find User:Abtract's leading sentences to be technically correct and yet convey all that you want to say unambiguously. Brews ohare (talk) 05:31, 29 August 2009 (UTC)

I've been unable to locate any recent sentences by Abtract that differ in any way that I can discern from the ones you're complaining about. Can you be more clear about what you would prefer to see that would "be technically correct and yet convey all that you want to say unambiguously"? Dicklyon (talk) 06:11, 29 August 2009 (UTC)

Dicklyon: The wording in Abstract's lead is

by the mid-20th century, the speed of light was known to be approximately 299,792,458 metres per second. For technical reasons in 1983, the metre was redefined using this value as a conversion factor. This had the effect of 'fixing' c at exactly 299,792,458 metres per second in the International System of Units.

Brews ohare (talk) 14:20, 29 August 2009 (UTC)

Say what? You won't find "my lead" anywhere. I copyedited and united two paragraphs in the lead. That's all. I didn't add a single word or concept of my own to it. Please don't interpret that to mean that I want to get bogged down in your personal obsessions about some purported problems with some particular definition of the speed of light. As I intimated in the first post to this section, you or anyone else should feel completely free to revert or change anything in my recent edits. That would certainly be preferable to polluting yet another section of this talk page with your useless "commentary". Tim Shuba (talk) 06:43, 29 August 2009 (UTC)

Tim Shuba Your suggestion that I dive into the article and rewrite it without discussion indicates little familiarity with what has been going on with this article. Likewise, you show no familiarity with the discussion on this Talk page. Calling my comment to you on the Talk page "pollution", rather than providing any response to it beyond pejorative characterization, is impolite and underscores your lack of understanding of decorum and of the use of a Talk page. Brews ohare (talk) 14:31, 29 August 2009 (UTC)

Brews ohare provides, in a direct response to me, a "commentary" in this section which has nothing, zero, nada, zilch to do with any edits I've made to this article or talk page, and expects me to respond and get involved in the tendentious editing that he has kept up for weeks? What a joke. Tim Shuba (talk) 15:58, 29 August 2009 (UTC)

Thanks for you contribution Tim. Although I think that we should say something about the aether and the MMX, the section that you removed had clearly been hijacked by crackpot science. Best to start again from scratch rather than argue endlessly about what exactly is wrong with each word.

Your work in improving the lead has been completely wasted. User:Abtract has decided that to rewrite it from scratch again. It is not that the new lead is that bad it is just that there is absolutely no point in trying to improve it as some other editor will then decide to rewrite it yet again.

I am beginning to think that trying to get this article back to being a FA by cooperative editing is a lost cause. There seem to be plenty of editors around who support this idea in principle but not may who are prepared to act in a way necessary to achieve this aim.

In my opinion a few more aggressive editing sprees by the right people would be most welcome. Martin Hogbin (talk) 10:56, 29 August 2009 (UTC)

In my opinion Abstract's lead should be adopted. If you differ, why not say why specifically, instead of urging your own form of words, which I find inaccurate for reasons explained at length here? It will not interfere with FA status, while your formulation will do so. Brews ohare (talk) 14:20, 29 August 2009 (UTC)

Approximate 300,000,000 m/s in lead paragaph?

Apparently we don't have as much consensus as it had appeared to start with the exact value in the opening paragraph. LouScheffer, Brews ohare, Abtract, and myself (Dicklyon) have all expressed a preference to open with 300,000,000 m/s and maybe also 186,000 mi/s approximation in the lead paragraph, and then introduce the complexity of a defined fixed value in a later paragraph, still in the lead. I don't think any of us are denying the fixed exact value, nor wanting to hide it, though we may have differing opinions on the deep philosophical implications of defining the speed of light and a system of units this way.

I understand that Martin Hogbin and some others are firmly set on wanting the exact value in the lead paragraph; but I'm not sure I understand why. Can we have some discussion on this again, without all the angst, but directed toward what makes a better wikipedia article? If each person with an opinion could limit themselves this weekend to one not-too-big well-thought-out position paragraph, we might start to collect some points of view and begin to understand each other. Please hold off a day or two on responding to each other and starting to argue the points, so that we can have the luxury of seeing who stands where on a simple issue for a change. Dicklyon (talk) 16:42, 29 August 2009 (UTC)

• As before, I've no objection to mentioning approximate values, as long as the exact value is also mentioned. Philosophicallly speaking the problems seem to be caused by different interpretations of the word distance. If you define distance between two points as the time it takes for light to travel (measured in seconds) then speed of light defined as d (s) /t (s) = 1 with no dimensions. If you define distance as the number of metres then speed of light defined as d (m) / t(s) = 299.... m/s exactly. Charvest (talk) 16:57, 29 August 2009 (UTC)

• Approximate in first para followed by 'exact' later in the lead. Imho wp:lead is clear in not advising too much detail too early. When that detail (nine significant figures and the word 'exact') may sidetrack readers before the context is known, I believe it should not be shown. I strongly favour not showing the word 'exact' until the history/context paragraph (of the lead) and I weakly favour using 300... in the first para because it is more user friendly (easier to read and all that most viewers need/want to know) whilst still being accurate (to three s.f.). This approach is aimed solely at helping the reader. Abtract (talk) 17:24, 29 August 2009 (UTC)

• We have just had a poll on this very subject (18 August), the votes were:

• A. Show only the approximate speed 1 Vote

• B. Show the approximate speed, followed by the exact speed 2 Votes

• C. Show the exact speed after the current first sentence 8 Votes Martin Hogbin (talk) 17:52, 29 August 2009 (UTC)

• Thoughts move on ... why not just give us your views? Abtract (talk) 17:59, 29 August 2009 (UTC)

• The discussion in this article is hard to follow .

Not being an expert i do know .

The speed of light can be measured to 10^-11.

Time can be measured to 10^-11.

Distance or length can not be measured with that accuracy except for the Kaisers foot when he was still alive.

What is the problem with by definition m=l/t ?. Wdl1961 (talk) 18:08, 29 August 2009 (UTC) p.s. Pi has been calculated to at least a million decimal places so you are way behind .

• I support the use of an approximate value at the outset, and an exact value, properly introduced, as in Abstract's Abtract |lead, later on. My reasoning is that early introduction of a nine or ten place number called "exact" for a speed of light that intuitively (and over centuries of history) would be thought of as a measurement, with a measurement error bar, along with the seemingly solipsist statement that the metre is "defined" to make it exact, is all just too much for a reader to absorb without more extended discussion. Arguing on the Talk page that a more extended discussion does appear in a later section "Speed of light by definition" is not too helpful, as that section is not mentioned in the intro. There is the further difficulty that the present formulation in the intro is in flat contradiction with the more correct presentation in the later "Speed of light by definition" section. Brews ohare (talk) 18:30, 29 August 2009 (UTC)
  • The statement in the lead section is in no way in contradication with the longer section that appears later. The problem is that Brews simply cannot accept that there is no problem with fixing the speed of light in SI units. S/he seems to feel that it is better to fix in terms of the length of some arbitrary bit of metal. This argument has gone on over many pages, and should stop forthwith. Brews, if you don't shut up on the question, I will ask for you to be banned from all pages relating to the speed of light in any way. Consider yourself warned. Physchim62 (talk) 19:47, 29 August 2009 (UTC)
    • Physchim62 might do well to reflect on the fact that threats rarely work and often rebound on the threatener ... time spent reaching the best wording, on the other hand, is rarely wasted. Abtract (talk) 20:15, 29 August 2009 (UTC)

1. "Fundamentals of Physics" Halliday, Resnick and Walker. edition 8E 2008
2. "Fundamentals of Physics" Halliday, Resnick and Walker. edition 8E 2008
3. Uzan, J-P; Leclercq, B (2008). The Natural Laws of the Universe: Understanding Fundamental Constants. Springer. pp. 43–44. ISBN 0387734546.
4. Greene, G (2003). The Elegant Universe. WW Norton & Co. pp. 55–56. ISBN 0393058581.
5. Davies, PCW (1979). The Forces of Nature. Cambridge University Press. pp. 127–128. ISBN 052122523X.
6. Duke, PJ (2000). "Electromagnetic waves in free space – no electric charges or currents". Synchrotron Radiation: Production and Properties. Oxford University Press. p. 53. ISBN 0198517580.
7. Schwinger, JS (2002) . "Gravitational waves". Einstein's Legacy: The Unity of Space and Time (Reprint ed.). Courier Dover. p. 223. ISBN 0486419746. {{cite book}}: External link in |chapterurl= (help); Unknown parameter |chapterurl= ignored (|chapter-url= suggested) (help)
8. Smolin, L (2007). The Trouble with Physics: The Rise of String Theory, the Fall of a Science, and What Comes Next. Houghton Mifflin Harcourt. p. 215. ISBN 061891868X.
9. Cite error: The named reference Boyes was invoked but never defined (see the help page).

• Misplaced Pages former featured articles
• Featured articles that have appeared on the main page
• Featured articles that have appeared on the main page once
• B-Class physics articles
• Top-importance physics articles
• B-Class physics articles of Top-importance
• B-Class relativity articles
• Relativity articles