Talk:History of the metre

Start

I have started this page by copying the relevant section from the Speed of light article. The referenced need copying. Martin Hogbin (talk) 10:35, 30 July 2010 (UTC)

Premature

This article has been planted here without careful consideration of its content. Although an article on this topic is desirable, this very preliminary version requires a lot of work. That can be accomplished here, or the article can be resubmitted when various editors have prepared a more adequate presentation. Brews ohare (talk) 13:45, 30 July 2010 (UTC)

I had begun an article along these lines at this link. Brews ohare (talk) 14:35, 30 July 2010 (UTC)

I'm not looking. If you want to improve this article then work on it here. otherwise this article is likely to be deleted. --Michael C. Price ^talk 06:31, 31 July 2010 (UTC)

Not a POV fork

The statement below is copied to the deletion request page.

When I started this article it was intended to allow detailed discussion of this topic that, in my opinion, was not suitable for the Speed of light or even the Metre articles. I also hoped to move some of the detailed discussion and disagreement from the Speed of light article on this subject to a more appropriate place where those interested could discus the issues involved.

It was not my intention to start a POV fork and in particular it was not my intention to create an article in which an alternative or personal POV could be presented, or in which the Speed of light arbcom decision on tendentious editing could be circumvented.

Brews, having looked through this article after your edits, I notice that, in addition to adding much useful detail, you have taken the opportunity to add some of the contentious material that you tried add to the Speed of light article. I hope you will cooperate with other editors in ensuring that this page represents a consensus view in the subject. Martin Hogbin (talk) 10:17, 31 July 2010 (UTC)

Martin, I'm sorry to say this, but how could you have expected this article to evolve any differently than it is, given the cast of characters who you know so well? Now instead of one battleground, we have two. If there is any legitimate place for this article in Wikipedia, and I do not believe that there is, it must be confined to the specific reasons for the change in definition and its effect on use of the metre in labs. Any discussion of the effect on the speed of light, or the speed of light as a physical constant, belongs in the Speed of light article, and is a content fork here (regardless of the original intent). Likewise, the history of the accuracy and methodology of measuring the speed of light belongs only in the Speed of light article, and is a content fork (and unreasonable duplication of content that will lead to inconsistent treatment) here. UGH!!—Finell 20:52, 31 July 2010 (UTC)

Finell: It is not possible to enlarge upon a topic without introducing it. Brews ohare (talk) 05:12, 1 August 2010 (UTC)

Sorry if I have created a monster but the argument on the Speed of light article should now subside as the text on this subject is reduced to a sentence or two.

Regarding excessive duplication of the material that properly belongs in the SoL article it should be deleted from here. Martin Hogbin (talk) 10:41, 1 August 2010 (UTC)

Conversion factor?

Although one source (at least) uses the term 'conversion factor' for the speed of light, I do not think this statement is particularly useful, or indeed relevant to this article.

I think it would be better to say that the SoL is a dimensional physical constant and therefore its numerical value is determined by the choice of units. Because of the new definition of the meter, this dimensional constant now has a fixed numerical value in SI units. We might say something like it is now used a conversion factor between the SI units of time and length. Martin Hogbin (talk) 11:26, 31 July 2010 (UTC)

I'm confused by what you're saying - can you rephrase your last sentence? A missing "as" perhaps? --Michael C. Price ^talk 12:55, 31 July 2010 (UTC)

Yes.Martin Hogbin (talk) Martin Hogbin (talk) 16:01, 31 July 2010 (UTC)

Martin: I believe you have identified some semantical issues here that might be worth looking into. They may lead to a more complete exposition of the philosophy in Jespersen in his section “Relate all measurements to time”. His formulation of this objective of standards people involves these words you have selected: “And measuring length in terms of time is a prime example of how defining one unit in terms of another removes a constant of nature by turning c into a conversion factor whose value is fixed and arbitrary.” I believe Sydenham says “Thus, the speed of light as a numerical value, is not a fundamental constant”, which I take as a way to say that the numerical value could be defined to be any number whatsoever, and that value is to be distinguished from the 'speed of light' as a physical phenomenon. Maybe it would be a useful exercise to try to express this goal, that of reduction of the number of necessary standards and introduction of more defined values, using different terminology? Brews ohare (talk) 15:13, 31 July 2010 (UTC)

Yes, I think that would be good. Martin Hogbin (talk) 16:01, 31 July 2010 (UTC)

In this connection, the separation of the numerical value from the 'speed of light' as a physical phenomenon is related to the question of how one checks for the answers to questions about the ‘physical phenomenon’ (such as those that arise in quantum gravity, e.g.) when the ‘numerical value’ is a defined quantity beyond measurement. I am not trying to say that it cannot be done, but that an explanation of how it is done is useful. Brews ohare (talk) 15:26, 31 July 2010 (UTC)

I do not understand what you mean. The 'speed of light' as an abstract concept is just that, an abstract concept. To make it any more than that a system of units is required and if you choose to base one of those units in the speed of light then the SoL will clearly have a fixed value in that system of units.

I'd guess that questions bout the 'speed of light' as a physical phenomenon have to be phrased in terms of comparisons, rather than numerical values. For example, the isotropy of the 'speed of light' as a physical phenomenon is checked by looking at the ratio of its value in one direction with that in another. It is done, for example, by spinning a resonator and monitoring its resonance frequencies as a function of angle. Its constancy over time is about the ratio of today's value with yesteryear's. And so forth. Brews ohare (talk) 15:42, 31 July 2010 (UTC)

The questions you pose cannot be answered in any fundamental sense. Even the One-way speed of light cannot be measured. Martin Hogbin (talk) 16:00, 31 July 2010 (UTC)

Well, suppose the resonator experiment is done. For example Test of the isotropy of the speed of light using a continuously rotating optical resonator. If a variation is found, and one has scrupulously eliminated the obvious sources of error like the effect of rotational forces on the geometry, one is left with two choices: the metre is longer in some directions than in others (the SI units language) or the speed of light is anisotropic. I suppose that the theoretical understanding would decide whether the anisotropy was due to (say) quantum gravity effects upon the phenomenon ‘speed of light’ or some other cause. Brews ohare (talk) 16:37, 31 July 2010 (UTC)

That is right. This is really just the MMX only better. There is no way to tell if the apparatus is longer when it is pointing in one direction than when it is pointing in another of if the (two-way) speed of light is anisotropic. In the end it is just convention.

There is no need to consider quantum gravity, for which there is no workable theory, you can get enough unintuitive results from GR which is a well verified theory.

The use of quantum gravity is just a hypothetical with an associated WP link. Herrmann et al. cite 3 or 4 other such theories that employ local Lorentz violations. Brews ohare (talk) 17:53, 31 July 2010 (UTC)

Brews, I presume that you know the difference between an established theory of physics and stuff that you find on the internet? Martin Hogbin (talk) 20:08, 31 July 2010 (UTC)

If (hypothetically) quantum gravity explained quantitatively that the phenomenon ‘speed of light’ varied with angle, and the resonator experiment showed that same variation, then one has shown conclusively that the c in the SI units is simply a conversion factor an arbitrary number determined by the deliberations of the CGPM, only historically related to the actual phenomenon ‘speed of light’. Of course, from a logical standpoint that is already understood to be the case, but the experiment shows that the isotropy of the selected numerical value is not "accidentally" also imitating the behavior of the actual phenomenon ‘speed of light’. Brews ohare (talk) 17:00, 31 July 2010 (UTC)

I do not know what you mean by 'simply a conversion factor'. We can choose to use it as a conversion factor from seconds to metres if we wish. Martin Hogbin (talk) 17:24, 31 July 2010 (UTC)

I struck out this term and replaced with another. Brews ohare (talk) 17:48, 31 July 2010 (UTC)

No, what such an experiment would show is that the current definition of the metre is not unambiguous, leading to a similar situation as when it was discovered that the spectrum line of Krypton used to define the metre pre-1983 was assymmetric, and thus did not have a well defined wave length, even though that wavelength of that spectrum line had an exactly defined value. In this situation, strictly speaking the speed of light still has a defined exact value, but the length of a metre depends in what direction you measure.TimothyRias (talk) 18:52, 31 July 2010 (UTC)

Timothy: Yes I agree entirely. Brews ohare (talk) 19:29, 31 July 2010 (UTC)

Regarding the term conversion factor: Wheeler “a mere conversion factor...a factor that arose out of historical accident..with no deeper physical significance. ” Brews ohare (talk) 19:29, 31 July 2010 (UTC)

You keep saying the same thing, that the numerical value has no special significance. We all know that, because it is a dimensional constant. Martin Hogbin (talk) 20:08, 31 July 2010 (UTC)

Yup, and even some dimensionless constants are arbitary. E.g. 1000 metres to a kilometre.--Michael C. Price ^talk 21:38, 31 July 2010 (UTC)

Martin: Sorry if I wasn't clear; the point was that Wheeler uses the term conversion factor that you don't like to use. Brews ohare (talk) 23:06, 31 July 2010 (UTC)

How can the speed of light be monitored in a system where it has a defined value?

IMO this section has no place here. First, the title is ridiculously long. Second, the answer is just "it would appear as a recalibration".

Let's just stick to the topic of the article name, and not give everybody an excuse to delete the article.

--Michael C. Price ^talk 17:54, 31 July 2010 (UTC)

Michael: Clearly the rotating resonator experiment was done. It is cited by 50 other articles, indicating a sustained interest among the scientific community. Its espoused purpose is to determine how narrowly the isotropy can be limited. That will ultimately reflect upon a variety of conjectures about local Lorentz invariance. The value of c in SI units cannot be "recalibrated" to obtain anisotropy; instead it would be amended to include angle corrections. How can this discussion be out of place here? Brews ohare (talk) 18:04, 31 July 2010 (UTC)

So? The recalibration has an angular dependence. This article is not a collection of any and all related questions about the speed of light. Ignore this and you'll see the article deleted by Finnell and others. Wait and see. --Michael C. Price ^talk 18:11, 31 July 2010 (UTC)

As I predicted [1]. I'm going to delete the whole section. Possibly say something about recalibration. --Michael C. Price ^talk 19:49, 31 July 2010 (UTC)

Michael: So, title the subsection "Possible recalibrations?". The point here is the following: the special theory of relativity depends upon establishing various properties of the speed of light. For example see Frauser et al.. It seems to me to be a very probable question among readers as to how these tests can be carried out when the so-called "speed of light" is defined to be 299 792 458 metres per second. It is apparent that any such tests become reinterpreted as what you call "recalibrations" . Of course, any such recalibration requires some theoretical explanation, for example, does the recalibration mean the special theory is only approximate, even locally? It doesn't help clarity to avoid the term "speed of light" and bury it under the term "recalibration" as though it was some kind of minor adjustment. Brews ohare (talk) 18:39, 31 July 2010 (UTC)

Are there any sources that explicitly link the 1983 redefinition with possible anisotropy? If not then we must delete this section. Please just stay focussed on the article topic, OK? --Michael C. Price ^talk 18:46, 31 July 2010 (UTC)

There's plenty of garbage incorrect WP:SYNTH in the preceding sections as well. From "History": The speed of light, usually denoted c nowadays, has been subject to investigation for millennia. However the nature of c, as a universal physical constant, was only established by the development of special relativity in 1905. This left the value of c, as with all physical constants, as a subject of further refinement by measurement.; from "Consequences": The effect of this definition gives the speed of light in vacuum an exact value in metric units, namely 299,792,458 metres/second. This can be regarded as a conversion factor between the second and the metre. [...] Because the second is defined in terms of atomic transitions that can be measured accurately, the new definition, being a ratio of measured times, allows for a definition of the metre with greater accuracy in practical measurement than one based on a ratio of lengths determined using a fringe count of interference patterns.; in "Increased accuracy and redefinition of the metre": The introduction of many sources required a methodology for comparison. The wavelength of the sources could be compared, but using interferometry to measure wavelengths was subject to some serious errors.[Note 4] On the other hand the measurement of frequencies had become very good. As all believed that frequency and wavelength were related by the speed of propagation, it was evident that a comparison of the frequencies of sources was tantamount to a comparison of wavelengths, but more accurate. The only issue was to insure that the speed of propagation was the same in all such comparisons, which led to the adoption of the speed of light in vacuum as the standard of speed of propagation. In 1975, considering that similar measurements of c agreed with each other and their uncertainty was comparable to that in the realization of the metre [...]

All of the quoted phrases are at best misleading, at worst completely incorrect. Physchim62 (talk) 18:54, 31 July 2010 (UTC)

Okay, specifics please. Start from the top. What's your problem with However the nature of c, as a universal physical constant, was only established by the development of special relativity in 1905.--Michael C. Price ^talk 18:59, 31 July 2010 (UTC)

Michael: I am not tied to any of this stuff. Change it around if you like. At least some of it is inherited verbiage that I may not agree with myself. Brews ohare (talk) 19:09, 31 July 2010 (UTC)

My point is that there is so much that is simply wrong that it would be better to start again from scratch: my preference would be in the relevant section(s) of Metre, which certainly need improving. Let's try to get something out of this by improving an existing (and very important) article rather than creating a new one. What was "wrong" with the 1960 definition of the metre? How did we find out about it? What was done to correct it? Those are the three basic questions (for me, at least). Physchim62 (talk) 19:21, 31 July 2010 (UTC)

It would be helpful if a list of what specifically you think is "wrong" were provided. Vague allegations don't improve my clarity of mind, and probably would carry over to any revisions. Brews ohare (talk) 19:32, 31 July 2010 (UTC)

Physchim62, I'm not seeing what you say "is simply wrong". Again, specifics, please. And again, this isn't in metre for the same reason it isn't in speed of light: too much detail for them since it lies at the intersection of the two. I really fail to see what all the fuss is about. --Michael C. Price ^talk 19:42, 31 July 2010 (UTC)

Physchim62, this was the reason that I set up this page, so that Brews' rather idiosyncratic views on certain matters could be discussed here rather than causing a great stir on the SoL or Metre pages. That is still no reason to allow bad physics on this page though. Martin Hogbin (talk) 20:13, 31 July 2010 (UTC)

Martin: Although it may be fashionable to say so, I object to my views being called idiosyncratic; so far as I can see nothing I've said is unsourced or peculiar to myself. If you think differently, please itemize. Brews ohare (talk) 20:34, 31 July 2010 (UTC)

How can the properties of light be tested if c has a defined value?

The section below was removed as being off-topic. As I think the question in the header is perfectly natural and in fact has come up over and over again, I would like some explanation as to why this subject is off-topic. I'm inclined to think it is simply fear of the AfD that is the cause of this deletion. Brews ohare (talk) 20:41, 31 July 2010 (UTC)

The special theory of relativity is posited to apply locally even in the general theory of relativity, and the definition of the metre explicitly states that the ‘time-of-flight’ realizations of the metre must be restricted to distances short enough that general relativistic corrections are not significant. The special theory is based upon a number of postulates concerning properties of the speed of light. In the present SI units system where the speed of light has a defined value of c = 299,792,458 m/s exactly, how can these tests of relativity be pursued as measurement methods become more and more precise? And as measurements explore more deeply the properties of light, what is their effect upon the definition of the metre?

The constancy of the speed of light can be monitored by observing the back-and-forth time-of-flight of light over a fixed distance, for example. Naturally, this is an experimental measurement, and subject to three possible errors (at least): first, the accuracy with which the time can be measured, second the accuracy with which the length can be maintained at a fixed value, and third the accuracy with which the medium of propagation can be held constant. An uncertainty is whether the length of the path changes with time (for example, as the universe expands, there is evidence that distances change; see Metric expansion of space) and whether the time provided by the standard of time changes (see Fine structure constant). Much more mundane factors can change the fixed length, for example, a bar changes length with temperature, flexure (changes in bending because of changed gravitational field or movement of supports), and so forth. These effects are small, but the anticipated variation in the speed of light is really small, so very extreme care must be taken in looking at all the possible sources of uncertainty.

The isotropy of the speed of light can be monitored using a rotating resonator. Should it happen that such an experiment in the future demonstrated an orientation dependence of the resonant frequencies of the resonator, in the SI units that would translate into a metre that had different lengths in different directions, and the standard for the metre would have to be amended to specify corrections to measurements based upon their orientation. That would not mean that the value of c in SI units would have to be altered. However, the underlying cause of the anisotropy would have to be explained, and conceivably a theory such as quantum gravity (say) might explain that observed anisotropy as an anisotropy of the physical phenomenon ‘speed of light’.

This subsection can be beefed up with more experimental references and book discussions, but it hardly seems off-topic to ask how the properties of the “speed of light” are to be discovered when c has an exact value. Brews ohare (talk) 20:49, 31 July 2010 (UTC)

Find a source that makes that link. It seems you still fail to understand basic Wikipedia policy. --Michael C. Price ^talk 21:27, 31 July 2010 (UTC)

What link are you looking for? Are you simply asking for a verbatim reference here as a time-honored scheme to resist material you have a dislike for? Brews ohare (talk) 21:34, 31 July 2010 (UTC) BTW, WP policy is not to knock down perfectly valid exposition on the pretext of requiring verbatim confirmation of obviously relevant points. Brews ohare (talk) 21:36, 31 July 2010 (UTC) Which things do you doubt?

• That continued tests of special relativity are being made?
• That they involve the speed of light?
• That a defined value of c in the SI system is compatible with such tests?
• That a reasonable source of confusion is that c has a defined numerical value, and yet the properties of the speed of light still can be measured?

Or is it your stance that you agree with all these points (your answer is that they are not questions, but facts) but that WP policy ties your hands and (sadly) requires a source that assembles them all in one place? Brews ohare (talk) 21:40, 31 July 2010 (UTC)

I've already answered your questions; you're just not listening (clue: focus, relevance, sources). Obviously you didn't spend your time in the cooler reading the Wiki policy pages (which I mostly agree with, BTW). --Michael C. Price ^talk 21:55, 31 July 2010 (UTC)

My opinion is that these issues don't fit well in an article about "the 1983 redefinition of the metre". But to address the issue here, if new physics is discovered that makes the original definition of the metre ambiguous, then that simple means that the definition will be amended. Compare his with the way we measure time. We know that moving clocks tick at a different rate than stationary clocks. This is a relevant factor in GPS and this is corrected for.

But because there are no such issues for the metre right now, you can't make an issue about specific things like potential violations of Lorentz invariance. I could just as well invoke other issues, like mixing of photons with axions, the fact that according to QED, light actually does not travel at c in external electromagnetic fields, that you can have dark matter particles with a very small magnetic moment that will cause a small aniotropy in the speed of light (because the Solar system moves at about 220 km/s in the galactic frame) etc. etc. You can find references for all these facts, but none of these consider this as a problem for the definition of the metre. Count Iblis (talk) 22:03, 31 July 2010 (UTC)

Count Iblis: You bring up a number of potential issues that I was unaware of. I agree that none of these are presently a problem for the metre, and that anyway, however they turn out, the metre will not change except to better specify what the ideal circumstances are under which the metre will be observed. As you know that would be only to add to a laundry list of requirements that grows longer all the time. The point of this subsection, although it has yet to be recognized that way, is not to suggest that there are real serious problems requiring immediate attention. The point of the subsection is that having a fixed defined value for c does not preclude investigation of the properties of the phenomena that is the speed of travel of light. That should be an obvious point, and it also should be obvious that a defined value for c will lead to confusion about whether these experiments can actually be done. The reader may suspect, as did Eddington, that the use of a net with 5" mesh will lead to the assessment that no fish exist that are less than 5" in size. To spell that analogy out, the definition of c as an exact value well may lead to the impression that any test of whether light is isotropic, dispersionless, etc. is placed beyond test by that definition. Brews ohare (talk) 22:20, 31 July 2010 (UTC)

What is placed beyond test is the defined value of c, not any experimental investigations. Brews ohare (talk) 22:28, 31 July 2010 (UTC)

As Wheeler says: “What will be the effect of future, still better, measuring technique? ...will that improvement in precision change the speed of light? No. ...we have to expect that the speed of light will remain at the decreed figure of 299,792,458 meters per second...” Brews ohare (talk) 23:11, 31 July 2010 (UTC)

Reply to M Price: Although these pages often are full of rhetorical devices, my questions above that you did not answer were posed in earnest as an attempt to engage you in dialog. I am not playing a debating game, and do not appreciate being fobbed off with some airy rebuttal that has no content. Brews ohare (talk) 22:31, 31 July 2010 (UTC)

I'm not interested in engaging in dialogue with someone who just doesn't listen. --Michael C. Price ^talk 04:15, 1 August 2010 (UTC)

Brews, I am not sure what you are trying to say here but you seem to be confusing and misunderstanding certain concepts. If you are saying, 'Does the new definition of the metre change reality?, the answer is 'No, of course it does not'. What it changes is the way we have currently chosen to measure some aspects of reality.

We can always do experiments and we must always be prepared to get results that we do not expect but how those results might be interpreted is to some degree a matter of choice. Forget quantum gravity, there is no theory of that subject that makes any testable predictions, according to the very well established theory of general relativity the current definition of the meter could already lead to some strange results if used inappropriately. Luckily the people at CGPM have though of this and have defined the metre as a unit of proper length. Have a read of this [2] Martin Hogbin (talk) 17:24, 1 August 2010 (UTC)

Does the definition preclude some measurements?

I wrote a new subsection that I hope is clearer about my concerns. Brews ohare (talk) 00:16, 1 August 2010 (UTC)

I have removed the entire section per WP:OR. This is 99.9% original research. The only thing that is not, is the footnote "The isotropy of the speed of light can be monitored using a rotating resonator". I don't think that deserves a section of its own. DVdm (talk) 09:22, 1 August 2010 (UTC)

DVdm: Although you view this section as original research, you haven't explained why you think this section is original research. The purpose and claim of the section is simply to say that the 1983 decision does not prevent any experiments, even though it does result in units where the speed of light is a defined constant. Do you disagree with that claim?

It also suggests that this point could be a source of confusion, and attempts to make clear just how such confusion can be avoided. You may disagree that this point can be confusing, but that is your own personal clarity on the matter, and needn't apply to everyone. Do you expect a source that suggests that this can be a confusing point? How about Wilkie's article cited in the intro? Why do you imagine so much is written about this definition?

Can you explain further what is OR in this section, and why you object to it? Brews ohare (talk) 10:19, 1 August 2010 (UTC)

There is nothing to be explained beyond what is said in WP:OR. DVdm (talk) 10:26, 1 August 2010 (UTC)

DVdm: I've asked for clarification because I do not understand your viewpoint. It would be helpful if you would explain your stance, rather than simply making an assertion. Perhaps the addition of some sources on issues you debate, or some rewording would fix matters. For reference in your reply, the subsection is provided below. Brews ohare (talk) 10:33, 1 August 2010 (UTC)

Does the definition preclude some measurements?

The special theory of relativity is based upon a number of postulates concerning properties of the speed of light. In the present SI units system where the speed of light has a defined value of c = 299,792,458 m/s exactly, can these postulates continue to be tested as experimental technique improves? For example, does the definition of c as an exact value mean that any test of whether light is isotropic, dispersionless, etc. is placed beyond test, by definition? The answer is: ‘No. What is placed beyond test is the defined value of c, not any experimental investigations.’

For example, consider the hypothetical observation of anisotropy. In the SI system of units, the anisotropy would take the form of the metre having different lengths in different directions. Of course, a standard of length cannot be allowed to be uncertain in this way, so the effect of this hypothetical observation would be a change in the definition of the metre, to add a directional correction. At the same time, however, the explanation of this anisotropy would be attempted by improvements in theory, and quite possibly the successful explanation would be that the physical phenomenon of the speed of travel of light was anisotropic. The physical phenomenon of the speed of travel of light must be kept separate in our thinking from the numerical value of c in the SI system of units. (This distinction might be a good reason to use the symbol c₀ for the SI conversion factor instead of c, as is recommended by the CGPM.)

In sum, tests of the special theory of relativity, for example, are not impeded by an exact definition of c in the SI system of units. The experiments show up as tests of the adequacy of the definition of the metre. Any necessary changes in this definition due to future experimental results will be accompanied eventually by theoretical explanations, and these theories may indeed invoke as yet undiscovered properties of the speed of travel of light.

Brews, I repeat: there is nothing to be explained beyond what is said in WP:OR. If you don't understand what is explained there, you should not be editing articles here. DVdm (talk) 10:43, 1 August 2010 (UTC)

DVdm: What I understand is that you cannot make any specific connections between WP:OR and the text above, and so refuse to do so. Brews ohare (talk) 11:25, 1 August 2010 (UTC)

I second DVdm's view. Brews, stop behaving like a baby that has to be spoonfed everything. Read WP:OR and inwardly digest, for heaven's sake. --Michael C. Price ^talk 11:31, 1 August 2010 (UTC)

With this it has become clear to me that Brews ohare should not be allowed to edit articles on Wikipedia. After all this time he acts as if he still has wp:NOCLUE about one of the most basic policies of Wikipedia. I am speechless. DVdm (talk) 11:36, 1 August 2010 (UTC)

I'd take these comments as violations of one of the WP:Pillars, namely WP:Civil, an unfortunate response to a simple request to connect some dots. If you are more able to understand WP:OR than I, go ahead and apply it to the above text. Point out what needs to be sourced, and is not. Brews ohare (talk) 12:06, 1 August 2010 (UTC)

References

RfC: Is the following proposed subsection WP:OR?

Does WP:OR apply to the subsection appended to this request on Talk:Redefinition of the metre in 1983, and if so what needs to be sourced? Brews ohare (talk) 12:42, 1 August 2010 (UTC)

The paragraphs follow below:

Does the definition preclude some measurements?

Does the definition of c as an exact value mean that any test of whether light is isotropic, dispersionless, etc. is placed beyond test, by definition? The answer is: “No. What is placed beyond test is the defined value of c, not any experimental investigations.” For example, the special theory of relativity is based upon a number of postulates concerning properties of the speed of light. In the present SI units system where the speed of light has a defined value of c = 299,792,458 m/s exactly, nonetheless these postulates can continue to be tested as experimental technique improves.

To illustrate how the properties of light may continue to be examined within the 1983 decision, consider the hypothetical observation of anisotropy in the propagation of light. In the SI system of units, anisotropy would take the form of the metre having different lengths in different directions. Of course, a standard of length cannot be allowed to be uncertain, so the effect of this hypothetical anisotropy would be to change the definition of the metre, by adding a directional correction. At the same time, however, the explanation of this anisotropy would be attempted by improvements in theory, and one might conjecture that the successful explanation could involve an anisotropic propagation of light. In any event, the fundamental physical phenomenon of the propagation of light must be kept separate in our thinking from the numerical value of c in the SI system of units. (Separation of these notions might be a good reason to use the symbol c₀ for the SI conversion factor instead of c, a practice recommended by the CGPM.)

In sum, tests of the special theory of relativity, for example, are not impeded by an exact definition of c in the SI system of units. The experiments take the form of tests of the adequacy of the definition of the metre. If there are any necessary changes in this definition due to future experimental results, they will be accompanied eventually by theoretical explanations, and these theories may indeed invoke, if need be, as yet undiscovered fundamental properties of the propagation of light.

Observation

The above paragraphs say nothing controversial. Brews ohare (talk) 12:42, 1 August 2010 (UTC)

Comments on RfC

• Please add comments here with a leading asterisk

• Yes, it's OR. Please read WP:OR again if it's not clear to you: it's a core policy of WP and you need to be familiar with it.--JohnBlackburne^words_deeds 13:31, 1 August 2010 (UTC)

What needs sourcing? Brews ohare (talk)

the whole of it: you can't just take a few choice facts from different sources and build an argument around it. That's OR.--JohnBlackburne^words_deeds 14:30, 1 August 2010 (UTC)

John: Your description of these paragraphs is incorrect. They present a very simple fact, the ability to carry out all the experiments that one used to do, despite the SI defined value for c, and illustrates that fact with the isotropy experiments cited, which are ongoing today. There is nothing more to it than that. If you read more into it, perhaps you can point out what wording is ambiguous or liable to misinterpretation? Brews ohare (talk) 15:19, 1 August 2010 (UTC)

If there is to be any objection to these paragraphs, it should be that it expresses something obvious. However, that appears not to be the case. Brews ohare (talk) 15:25, 1 August 2010 (UTC)

Having no guidance form the critics, I have sourced just about every sentence. Brews ohare (talk) 16:08, 1 August 2010 (UTC)

• Yes, well actually I am not sure if OR is the right term. The paragraph in question is a rambling essay mixing a number of things that are correct but do not need saying with a naive view of physics that means very little. Martin Hogbin (talk) 16:48, 1 August 2010 (UTC)

The OR arises from the synthesis, i.e. the drawing together valid sources to write I'm not sure exactly what but something not given in the sources. That it's poorly written and makes little sense, rather than drawing clear conclusions with a clear POV, makes no difference. The point of WP:OR is that anything written by editors which draws conclusions other than those given in the sources is OR, precisely because without the conclusions being in reliable sources there's no way to be sure if it's Nobel prize worthy insight or juvenile ramblings. That's why we insist on sources, and any reasoning not given in sources is OR.--JohnBlackburne^words_deeds 19:51, 1 August 2010 (UTC)

Again, John, what is the synthesis you speak of? Vague generalities are great for political speeches intended to seem to say something, but are not useful here. Brews ohare (talk) 20:18, 1 August 2010 (UTC)

Martin: Well, at least you are on the right track in saying OR doesn't fit the objections. However, I don't understand the characterization as "rambling". The lead sentence says precisely what is to be discussed. The second paragraph gives a pertinent example with references to the literature. and the last paragraph is a succinct summary. These paragraphs do not say "a number of things" but one thing (experiments still can be done even though c has a defined value), with an example, and a summary. If it seems rambling (or as Blackburne says, "unclear") I'd submit that it is because you expect to find something there that isn't (maybe some diatribe about modern physics?), and so get lost trying to find the arguments you expect, in place of reading the arguments actually made. Brews ohare (talk) 20:29, 1 August 2010 (UTC)

So your point is that experiments can still be done, after the speed of light is fixed by definition? Yes, of course they can. I do not think anyone is challenging that. Martin Hogbin (talk) 21:32, 1 August 2010 (UTC)

Yes, Martin. As I said, the segment is obvious. However, there does exist confusion on this point, as the segment says. Brews ohare (talk) 23:44, 1 August 2010 (UTC)

To make the point clear, it helps to notice that because c is fixed, as umpity sources say, any refinement in measurement of the properties of the propagation of light are reflected in the metre. So, in particular, anisotropy in propagation (if it occurred) would show up in anisotropy in length (also cited). Brews ohare (talk) 23:48, 1 August 2010 (UTC)

What would an experiment that appeared to show anisotropy in the speed of light tell us? That depends very much on what the experiment was. If, for example, it was an experiment to compare the speed of light round a closed path in one direction with its speed in the other direction, it might tell us that the apparatus was in a rotating frame. If it compared the speed of light in an upward direction with that in a downward direction it would tell us that there is gravity on Earth and that GR describes it well. If an interferometric measurement in space showed an anomaly in the speed of light it might tell us that we have detected gravity waves, another verification of GR. None of these have much to do with the definition of the metre, which is a unit of proper length.

If, on the other hand, an anomaly in the speed of light that could not be explained by our current theories of physics was detected experimentally then something would have to change. A new theory maybe or perhaps just a modification to an old one. Exactly how this might affect the current definition of the metre would depend on what was discovered. Maybe the frequency of the light would be specified, maybe some kind of correction for some new effect would be needed, maybe the speed of light standard would be abandoned in favour of something else, maybe our whole concept of distance would need to be revised. Who can tell? Speculation about the anisotropy of the speed of light and its effect on the definition of the metre is completely pointless. Martin Hogbin (talk) 14:12, 2 August 2010 (UTC)

Martin: You've made some good points about measurements and theory. But I'd like to say as gently as possible that this is not what the subsection is about. The object is to explain to the naïve that just because c₀ has a fixed value beyond measurement does not mean that the fundamental properties of light are beyond study. An example is provided in the second paragraph of very particular studies done relevant to isotropy. The point of the example is not to laud the experiment, but to show that experiments about light are still ongoing, and that if one wishes to describe them in SI units they are phrased as searching for differences in lengths in different directions, that is, a positive outcome would lead to an anisotropic metre unless the definition were amended to incorporate a directional correction. I think you have become drawn into a discussion of the types of experiments and their possible interpretations, and that is not what the example is selected to illustrate. Brews ohare (talk) 15:05, 2 August 2010 (UTC)

If all you want to say is that the current definition has no power to make light travel at the speed we tell it to that is so obviously true that it does not really need saying in any great detail. I see three classes of possible future experiment being possible possible, those that refine the delineation of the meter, those that are outside the scope of the meter definition, and those that will change everything. Which need a detailed explanation here? Martin Hogbin (talk) 15:27, 2 August 2010 (UTC)

An interesting formulation. I find the literature to so easily confuse the "exact speed of light" with c₀ that something needs to be said, even if it is obvious to you. Your three categories of measurements may not be mutually exclusive. If you want to talk about that, I'd suggest another subsection. This one has only a very simple goal. Brews ohare (talk) 16:33, 2 August 2010 (UTC)

What i do not understand is which of my three categories do you want to give details of in the article. Or do you claim there is another category? Martin Hogbin (talk) 17:15, 2 August 2010 (UTC)

Hi Martin: That is an interesting question. You are thinking harder about this than I have.

My stance was just that choosing a unit of speed didn't mean that light traveled at that speed (a point you regard as trivial), particularly if the speed of light depends on a variety of parameters that we don't know about and therefore have omitted from the definition of the metre. Therefore, an experiment using a particular realization of the speed of light doesn't necessarily realize the standard unit because these hidden variables are not controlled for. I interpreted that as showing up as failure of the metre to be uniquely defined as we discover these hidden variables, because extra conditions would have to be attached to its definition to insure that the realization would produce the unit. Obvious examples are isotropy, dispersion, and less obviously, the volume of the universe.

Help me out with other kinds of examples. Brews ohare (talk) 17:34, 2 August 2010 (UTC)

I have started a section below to discuss this subject further. Martin Hogbin (talk) 17:52, 2 August 2010 (UTC)

More discussion

• The problem with this section is that it is trying to answer a question without referring to a source that asks and treats the said question. Instead it tries to answer the question by arguing based on facts drawn from other sources. This technique is very usual for scientific writing (you might even go as far as to say that it is the cornerstone of scientific writing.), but is unacceptable on wikipedia. The reason is that in scientific writing there is reason to assume that the author is competent in interpreting his sources (and there is peer review to check that he is), moreover in the case of scientific journals the readership can also be expected to be able to interpret the quoted sources. In the case of wikipedia neither is the case, there is no way to know if an editor is a specialist in the field, or just some cranky teenager, nor can it be expected of the general reader to interpret the given sources. As such, articles should keep interpretation of sources to a minimum, and more importantly directly cite sources for their conclusions. Brews, I repeatedly noticed in your writings that this point about wikipedia policy has not quite come across with you. Many times have I seen you stick a whole battery of refs on a statement, none of which directly dealt with the statement to be sourced, but where to statement should (according to you) follow from this collective of sources. I really do hope, you start to understand what this policy means and more importantly why it exists in first place. It might prevent a lot of hardship and drama and save you a lot of energy. This is not meant as a personal attack, rather it is meant as very well meant advice.TimothyRias (talk) 09:30, 2 August 2010 (UTC)

Hi Timothy: I do appreciate your taking the time to explain matters carefully. Let me see whether this analysis applies to the segment above, by pulling it apart to identify whether indeed this criticism applies. To do that, let me state the various points and examine the evidence brought to bear.

• Does the definition of c as an exact value mean that any test of whether light is isotropic, dispersionless, etc. is placed beyond test, by definition?

Is this question asked and answered by a source? Yes, it is. I haven't dredged up all such sources. One is E Richard Cohen, quoted in the piece. The question also is asked (but never answered) by Wilkie “Does the new standard mean that we will be unable to detect [various variations in the speed of light]” One example of this question is that of isotropy, treated in the second paragraph. The Michelson-Morely experiment as done today using the analysis of relativity is another way to test relativity itself, and two sources are cited referring to measurements of isotropy using a rotating resonator by means of detecting dimensional changes in the cavity via a shift in frequency. They clearly ask and answer that particular version of the question. So I'd say the question doesn't originate with me, and neither do the answers. So your first concern is put aside, do you agree?

• The summary says, in effect, that questions about propagation of light are turned into questions about the definition of the metre by the new definition. That point has been cited in the first paragraph, referring to four sources: Adams, Rindler, Wheeler and Wilkie all of whom point out that refinements in c don't affect c₀. It is established in the example of isotropy by reference to Brillet and Hall, but also applies to later work using rotating resonators like the cited Herrmann et al. These authors specifically refer to the changed length of the resonator as a function of orientation. Evidently, the idea here is that the length didn't really change, but the frequency shifts indicating an apparent change, which they actually think is a test of the isotropy of light propagation. Other interpretations are that the metric of space is anisotropic. In any case, the point is that tests can be made, and if they proved to establish a real effect, the metre would be different in different directions. I believe this point to be adequately sourced.
• The object of this section is not to preach to the choir, but to explain to the naive that just because c₀ has a fixed value beyond measurement does not mean that the fundamental properties of light are beyond study. Do you disagree with the thesis?

I hope you might engage in some detail here as to how you would present this simple matter? Brews ohare (talk) 12:59, 2 August 2010 (UTC)

Experiments with a defined speed of light

I cannot see any present or future experiments that would not fall into one of these categories.

Those that refine the realisation of the metre

These would include the many things currently being done in metrology laboratories, such as developing more stable light sources, more accurate clocks, better vacuums etc. All we need to say about these is that they will enable us to realise the meter more precisely.

These efforts are not all-inclusive, as measurements of isotropy (for example) also could result in refining the realization of the metre. So also could measurements of dispersion, nonlinearity (intensity) effects, etc. They would lead to additional requirements upon the conditions under which the standard speed of light would be realized. Equivalently, if the standard orientation, or wavelength, or field strengths were not realized, the metre would be subject to corresponding corrections to obtain the standard metre. We already see the beginnings of such restrictions in the present requirements upon gravitational conditions and language about proper lengths. Brews ohare (talk) 19:27, 2 August 2010 (UTC)

No, current theories of physics tell us that the speed of light in an inertial frame is isotropic. If any experiment were to find it not to be it would come under the third category here.

Regarding GR corrections all that can be done is to try to reduce these to a minimum. They are well understood. Martin Hogbin (talk) 20:20, 2 August 2010 (UTC)

Those that are outside the scope of the definition of the meter

The meter is defined as a unit of proper length. It thus only really exists in an inertial frame free from gravitational waves etc. We require another unobtainable environment like the perfect vacuum.

In a real world environment, things are, in principle, more complicated due to the curvature of spacetime by gravity. How, for example, would we measure the height of a flagpole in meters? What exactly to we even mean by length in this case, it is not uniquely defined quantity in GR. The point to be made here is that there are experiments which could be done which might show an anisotropy in the speed of light, but this would be due to failing to meet the conditions under which the meter is defined. The physics of the experiments is well understood but irrelevant to the definition of the metre.

What can we say about these experiments. I suggest very little, except to say that in GR all bets are off, and refer the reader to a good book or WP article on the subject.

Those which will change everything

If it is found that, even in an inertial frame, the speed of light is anisotropic, or that the speed of light varies with frequency, as predicted by some theories, then something would have to change. The current definition of the meter might be refined, or completely re-thought in such circumstances.

Perhaps complete revision would be necessary, but the easiest first step would be a simple correction factor. I mention this in Part 1 above. Brews ohare (talk) 19:27, 2 August 2010 (UTC)

I have no idea what will be discovered next and neither have you.

What can we say about this? Not much except that experiments will continue and if they find something unexpected we will have to change the rules accordingly. Martin Hogbin (talk) 18:31, 2 August 2010 (UTC)

I agree that there is a lot of depth involved in looking at the origins of such effects. I don't think they have to be settled in order to make the point that defining c₀ as a definite numerical value doesn't mean that experiments about the speed of light can't be made. The basic problem with the whole thing is that the CGPM should never have called the unit the ‘speed of light’. If they had called it the ‘SI standard speed’ and said that it could be realized as the speed of light in vacuum all this confusion would die. Brews ohare (talk) 19:27, 2 August 2010 (UTC)

There is no unit called the speed of light. Nobody but you seems to be confused by this. It is not a standard speed it is the seed of light, the meter is a standard distance. Martin Hogbin (talk) 20:24, 2 August 2010 (UTC)

OK, Martin. That is not a point that needs to be settled here. The point of the subsection is now clear to you. It is not as deep as you imagined it to be. It simply discusses a possible confusion over whether measurements involving light can be made despite c₀ having a defined value. The point is made, the example of isotropy is given, and a summary provided. Sources are given throughout. So, the big question: how would you change this proposed sub-section? Brews ohare (talk) 05:26, 3 August 2010 (UTC)

I cannot find anything about isotropy or other experiments in the article now so this particular discussion seems irrelevant for the moment. Martin Hogbin (talk) 10:36, 3 August 2010 (UTC)

Martin: You seem to have forgotten that this is a discussion of the RfC, not the article. In any case, Blackburne has filed suit here and here, so I will no longer discuss this matter. Brews ohare (talk) 19:34, 3 August 2010 (UTC)

What is the speed of light now?

This article variously refers to c as, a derived quantity, as standard speed, and a unit. It is a dimensional constant with a value fixed by convention in the SI system of units. Martin Hogbin (talk) 10:39, 3 August 2010 (UTC)

Well done!

The expansion and move have, in my opinion, created an excellent and informative article. Well done to all involved. Martin Hogbin (talk) 11:38, 22 August 2010 (UTC)

Is this true?

The article currently contains the following line:

"The shortcomings of the krypton standard were demonstrated by the measurement of the wavelength of the light from a methane-stabilized helium–neon laser (λ ≈ 3.39 µm)."

My understanding is, that this is not completely accurate. The short comings of the Kr-86 standard were already determine by WRC Rowley and J. Hamon in 1963, who following the new definition of the metre 1960 preformed a thorough investigation of the Kr-86 line (at the request of the BIPM). Their article is in French, so I cannot directly verify what they reported, but Barger and Hall cite them in their article when talking about the line asymmetry.TimothyRias (talk) 14:15, 30 August 2010 (UTC)

The statement is true, but may not be precise ;) I probably don't have access to the Rowley and Hamon paper, but feel free to Wikimail me if you have a copy in PDF format and I'll have a look at it. Physchim62 (talk) 18:35, 30 August 2010 (UTC)

I haven't been able to find a digital copy of the Rowley and Hamon article. Just other articles referring to it. It seems some digging may be in order. I'll collect some of the data I've dug up in a new section.TimothyRias (talk) 08:38, 31 August 2010 (UTC)

Relevant data for history of the metre

I've been doing some digging wrt to the history of the metre. I'll post some of it here so we can decide on how it should be reflected in the article.TimothyRias (talk) 09:34, 31 August 2010 (UTC)

• In 1963 WRC Rowley and J Hamon discovered the assymmetry in the Kr86 line used to define the metre. This is mentioned in the Evenson et al article on page 1347, where they refer to "W. R. C. Rowley and J. Hamon, Rev. Opt., Theor. Instrum. 42, 519 (1963)."

• In 1970 the Comite Consultatif pour la Definition du Metre (CCDM) discussed the use of stabilized lasers as (secondary). The possibility of a new definition of the metre based on stabilized lasers was discussed, but it was found to early to make a specific recommendation. The commite did recommend that the laboratories continue to improve stabilized laser AND improved measurements of the speed of light. Terrien, J. (1971). "News from the Bureau International des Poids et Mesures". Metrologia. 7: 43. Bibcode:1971Metro...7...43T. doi:10.1088/0026-1394/7/1/008. According, to the Evenson et al. ref at same meeting the reproducibility limit of the Kr86 line due to the assymmetry was 1×10⁻⁸. (Footnote 17 of that article).

Different regimes of length and the metre

The subject of the metre seems somewhat incomplete without some comment upon how lengths are determined in practice, especially in greatly different length regimes.

Evidently the time-of-transit definition can be used directly in some cases to find a length in metres by taking this length as (time-of-transit) × c₀ and correcting c₀ for the medium in which the transit occurs. However, that doesn't always work, and the time-of-transit approach has to be replaced by, for example, light interferometry or by electron scattering. The general subject is called traceability and refers to a kind of "analytic continuation" of measurements using one technique to calibrate measurements using another technique using an overlap regime where both methods can be used. The most important of these transitions is from use of time-of-flight techniques to use of various wavelengths, first of light, and second of electrons. One example is the use of crystal lattice spacings to calibrate electron beam measurements with X-ray measurements. A much more complicated and controversial regime than the spacing of atomic planes is the measurement of objects physically small in every dimension, where 3D effects on different surfaces interact.

Something of these issues in extending the notion of "metre" to different lengths should be included here. Brews ohare (talk) 20:14, 21 November 2011 (UTC)

In this connection, it might be mentioned in the article that the second being 9,192,631,770 periods T of a certain atomic transition, combined with the relation between period and frequency f=1/T, means a wavelength in metres of a transition corresponding to a frequency f is λ=c₀T = c₀/f. This matter is spelled out on the BIPM website. This connection allows a measurement in wavelengths to be converted immediately to metres, with a conversion error related primarily to two factors: (i) the error in determining the frequency, and (ii) the errors introduced by the departure of the medium from classical vacuum. This observation is important in establishing interferometric measurements of lengths in wavelengths of light as proxies for time-of-flight measurements, enabling the use of interferometers for length measurements in metres, usually a much more accurate approach in practice than the time-of-flight definition. Brews ohare (talk) 21:30, 21 November 2011 (UTC)

Welcome back to WP Brews. You still seem to be trying to make some obscure point about the current definition of the metre. I hope you are not intending to repeat your endless arguments on this subject that eventually resulted in your being banned. Martin Hogbin (talk) 10:06, 22 November 2011 (UTC)

Hi Martin: Thanks for the welcome. No, I'm not looking to repeat past history. The above observation is not about the definition of the metre, but about applying the definition to various length scales. The difficulty is that one has to switch measurement techniques as the scale changes, or as greater accuracy is demanded, because each technique has its own realm. So, the suggestion is to point out the issue and how this is accomplished, the process called metrological traceability by the BIPM. Brews ohare (talk) 19:33, 22 November 2011 (UTC)

That is a valid point but would it not be better in an article on metrology. This article is about the the definition of the metre itself rather than the way that tiny fractions of the metre or large numbers of metres are measured or realised. Martin Hogbin (talk) 23:02, 22 November 2011 (UTC)

Martin: If one wanted to make a full-scale discussion I'd agree about that. Here in this article something briefer would be appropriate, just to point out the issues of metrological traceability without going into all the nitty gritty. Brews ohare (talk) 23:40, 22 November 2011 (UTC)

I too don't see what this has to do with the history of the metre, rather than metrology, although it does not seem appropriate for that either which is a broad overview, covering many different aspects of the topic: for example it mentions the metre but in far less detail than here or at metre.--JohnBlackburne^words_deeds 00:13, 23 November 2011 (UTC)

←outdent The history of the metre includes the development of its application to new length regimes, for example, the evolution occurring in measurement in metres of very small structures in integrated circuits. These applications involve specific forms of metrological traceability peculiar to the metre.

Of course, these developments have more than an historical aspect, and details include technical matters about their limitations, their accuracy, their instrumentation and so forth, not appropriate to an historical article like the present one. Although these details are a subtopic of metrology in general, I'd agree with Blackburne that they do not belong in the article metrology, which is a very broad subject. I believe the appropriate place for greater detail is the article metre because that concerns metrological traceability specific to the metre.

Perhaps a subsection on metrological traceability in the article metre would allow the reference to these matters in the present history article to be very brief with a link to that subsection in metre? What do you all think? Brews ohare (talk) 17:24, 23 November 2011 (UTC)

That is really the history of the realisation of the metre. I have no strong opinion on whther that would bet be discussed in a section of this article or in another one. Martin Hogbin (talk) 13:13, 26 November 2011 (UTC)

I'm working on a new article maybe to be called length measurement. Perhaps you would like to comment upon it. The draft article is located here and the talk page link is here. Brews ohare (talk) 00:25, 26 November 2011 (UTC)

Exact location of Méchain's meridional measurement on Montjuïc hill?

Hi, this and a number of related articles currently state that Mechain's meridional measurement took place at a fortress at the top of the Montjuïc hill near Barcelona. The corresponding pictures link to the Castle of Montjuïc, but call it "Fortress Montjuïc". The Montjuïc article, however, states that the hill was the site of several fortifications, of which only the latest (the Castle of Montjuïc), remains today.

Our articles on the metre assume "fortress" and "castle" to be the same, whilst the dedicated article on the castle does not even mention any meridional measurements at all.

In order to verify and improve the information given in the articles, do the historical sources specify the exact location of measurement in better details, f.e. by naming a specific building or place? --Matthiaspaul (talk) 01:12, 21 October 2015 (UTC)

Delambre specified the location as the tower of the fortress of Montjuic ('tour du fort de Montjouy'), in his 1807 report about the measurement of the meridian arc. ('Base du système métrique decimal ..') Ceinturion (talk) 14:24, 21 October 2015 (UTC)

Great, and thanks for looking this up! So, unless there were two towers (unlikely), we are correctly linking to the Castle of Montjuïc and photos showing its tower. --Matthiaspaul (talk) 23:17, 21 October 2015 (UTC)

I also found a photography of a commemorative plaque on this tower:

Charles Inigo (talk) 04:36, 28 September 2023 (UTC)

Incorrect entry in the "history of definitions" table

The first entry of the table is evidently wrong, as it says that the measurement by Delambre and Mechain was complete in 1795. In fact the measurements by Mechain were not finished until 1798.

Definitions of the metre since 1795

Basis of definition	Date	Absolute uncertainty	Relative uncertainty
1⁄10,000,000 part of one half of a meridian, measurement by Delambre and Méchain	1795	0.5–0.1 mm	10−4
First prototype Mètre des Archives platinum bar standard	1799	0.05–0.01 mm	10−5

Cardarelli, which is given as the source for the entire table, is a questionable source because of unclarities.(link) In this "history of definitions" table, it is unclear how the uncertainty was defined, and why the uncertainty changed. When comparing it to data from better sources, it seems that in this table, before 1800, uncertainty represents uncertainty in the meridian measurement, and after 1800 it represents something else: differences between replicas and the primary standard, the Metre des Archives.

The provisional metre was based on the meridian measurement by Lacaille in 1740. In 1793 Borda estimated the uncertainty of that meridian measurement was 0.03 percent, and the same uncertainty applied to the provisional metre.[3] In 1798 the meridian measurement by Delambre and Mechain was completed. The final metre was declared to be equal to 443.296 lignes, implying an uncertainty of 0.0001 %. A few years later Delambre reviewed the errors and concluded the uncertainty was 0.01 %, according to Ken Alder.[4] These uncertainties represented how much the metre deviated from 1⁄10,000,000 part of the quadrant.

In the 19th century replicas were occasionaly compared to the Metre des Archives, by means of a special micrometer screw gauge. In 1806 the difference was less than 2 µm (or 0.0002%), the uncertainty of the micrometer. In 1864 the difference was 3 µm (or 0.0003%). The uncertainty of the improved micrometer was 0.3 µm.[5]

Therefore, the beginning of the table should be replaced by something like this:

Definitions of the metre since 1793

Basis of definition	Date	Absolute uncertainty	Relative uncertainty
Provisional metre, based on meridian arc measurement by Lacaille in 1740	1793	0.1 mm	10−4
Final metre, based on meridian arc measurement by Delambre and Méchain	1798	0.1 mm	10−4
First prototype Mètre des Archives platinum bar standard	1799	2 µm	10−6

Ceinturion (talk) 13:41, 22 October 2016 (UTC)

References

Some matters of form and composition

As a matter of form, the first sentence of the lead for an article titled foo should say, "A foo is a bar, baz, quux...". Hence: "The history of the metre is the history of <some events> over <some time period>." Now we know what we're going to write about, and the reader knows what he's going to read about. Then the second sentence should probably tell the reader what a metre is in everyday terms. It does say it's a unit of length. But if the article were titled "History of the toise", it too is a unit of length, very familiar (at least in France), but we'd scratch our heads because we don't know what it is. This is an encyclopedia, and it is supposed to be a careful description of what things are. If readers already knew what things are, they wouldn't need us to tell them. A grade schooler in the U.S. probably doesn't know what a metre is. So the second sentence should be something like, "A metre is the metric system unit of length, approximately the height of an average adult man's hip line." We could use any everyday object for comparison, but body proportions are familiar to everyone, even a child. The rest of the sentences in that paragraph also have a defined structure and composition-defined content.

The article should begin with text, not a table. The direct predecessor of the metre was the toise defined in the Carolingian system as a body measure. The metre was originally defined in terms of the toise. Pendulums are irrelevant, so the section Universal measures is toast. The very first provisional metre in 1794-5 was defined according to a survey of earth's meridian by Lacaille in 1740. The metre was never defined as 1/10,000,000 of the meridian. The definition of the metre in French law in 1795 was exactly 443.44 lignes; in French law in 1799, the metre was defined as 443.296 lignes. Both figures represented lengths of fabricated platinum bars, and both were known at the time (or very soon afterward) to be in error as fractions of the meridian, though they didn't know by how much.

The section International prototype meter is a little long but ok. We don't need to know about any organizations or their meetings in order to understand this object. Getting rid of those would shorten it quite a bit. This is an article about history, not administration.

The sections Krypton standard and Speed of light standard are pretty hard science, and this is an article about history (did I say that before?). It is about one of the most familiar everyday items, like a baseball, so the article should be toned in everyday language and terms. I don't think I should need to understand things like "zero nuclear spin" and "laser interferometry" to read an article on baseball. A degree in physics is necessary to substantively understand these sections. The average adult probably doesn't have a good idea of what cadmium and krypton are, certainly not why they're suited to the purpose here. They're wikilinked, but as long as the text stays with the bounds of the article topic, we shouldn't need to keep jumping outside the article. Footnotes describing these two is probably a better way to go. Something like, "krypton is an inert gas similar to neon that occurs in trace amounts naturally, and is used in certain kinds of fluorescent tubes". Even so, spectral analysis is something most adults won't know anything about. History is a kind of narrative and these two sections are "bumpy". I think they can be rewritten to make them shorter and more accessible. Technical terms should be replaced with vernacular. Any "hard" science should be moved to sideboxes or footnotes so the text is readable.

Sbalfour (talk) 22:57, 19 January 2018 (UTC)

Duplication of text and content

The existence of this article is quixotic, and duplicative.

The section Universal measures is about pendulums, which are irrelevant.

The sections Meridional definition and Metre des archives were largely cut-and-pasted from other articles, esp. an earlier rendition of History of the metric system. Parts of other sections while not cut-and-pasted, duplicate facts and content in other articles. What content here wouldn't be a welcome addition to History of the metric system? There are also large history sections in International system of units and Metric system articles. They're all one and the same. I estimate by count of paragraphs, sentences and phrases/facts found in other articles, that over 2/3 of this article is duplication.

I suppose there'll be, or someone will deign to create it, History of the Kilogram, and History of the second, etc. It's hubris, a snake eating its own tail. It's a kind of internal copyvio, because editors will go to existing articles to cut-and-paste for the new one. That should tell them straight away that the article should not be created, and whatever effort they wish to pursue should go into an existing article. The history of the meter is an integral and inseparable part of history of the metric system. So I propose the following:

There are 7 structuring level 2 sections here:

1 History of definition

2 Universal measure

3 Meridional definition

4 Mètre des Archives

5 International prototype metre

6 Krypton standard

7 Speed of light standard

The first "section" is just a table, not really a section at all. It could actually appear in an article just about anywhere and is probably better placed as a kind of appendix rather than introduction. Universal measure is about pendulums, which never became any part of any measurement system. That section can simply be deleted. Meridional definition and Metre des archives are wholly duplicate, and can be deleted. The first two paragraphs of International prototype metre are also cut-and-pasted with a little fudge in the last sentence or two; what's left of it can be merged into History of the metric system.

That leaves the last two sections, 1/3 of whose content is duplicate, so in aggregate they will shrink substantially when properly merged. Both sections are highly technical, and will need to be rewritten for accessibility as noted above. Speed of light standard is mostly current science, and will merge into Speed of light, a science article. The Krypton standard is more history, so will merge into History of the metric system. If it's not rewritten, it will fit better in a science article, Metric system#Realization of units section.

Sbalfour (talk) 22:57, 19 January 2018 (UTC)

It has been proposed that the section 3 Meridional definition be split out into another article titled Meridian arc of Delambre and Méchain. So why not merge the sections Meridional definition and Mètre des Archives into a new section titled "The metre linked to the figure of the Earth". Charles Inigo (talk) 04:47, 28 September 2023 (UTC)

Told

We are told that the Scientific Revolution started with Copernicus in 1543. Copernicus's system was the same as that of Aristarchus, who was much earlier. — Preceding unsigned comment added by 82.15.21.214 (talk) 12:19, 1 March 2019 (UTC)

We are told that the history of the meter started in 1543. It was not mentioned for some time after that. — Preceding unsigned comment added by 82.15.21.214 (talk) 12:24, 1 March 2019 (UTC)

It is implied that the decimal system was introduced after 1543. It was used earlier. — Preceding unsigned comment added by 82.15.21.214 (talk) 12:30, 1 March 2019 (UTC)

It is implied that royal decree or physical prototypes are inferior to the meter. Actually, many definitions of the meter use much the same things. A platinum bar could seen as a physical prototype. — Preceding unsigned comment added by 82.15.21.214 (talk) 12:55, 1 March 2019 (UTC)

The decree of the General Conference is not much different to a royal decree. — Preceding unsigned comment added by 82.15.21.214 (talk) 13:12, 1 March 2019 (UTC)

The sentence "The history of the metre starts with the scientific revolution that began with Nicolaus Copernicus's work in 1543" could be change to "The history of the metre starts with the scientific revolution, which started in the 16th century". The reader can follow the link to scientific revolution to find the exact who and when details. It was the scientific revolution which made the first systematic attempt to base things on unchanging facts rather than vague beliefs or imprecise things like somebody's foot/arm/thumb. It was still a decree but at least it was a lot more precise and didn't change according to who measured it and didn't change according to which person's body part it was measured from. Ie, it was meant to be objective.

The article doesn't claim that the metre itself started in 1543 - only that the process which led to the metre started circa 1543.

The article doesn't claim that decimals were invented after 1543 - only that the proposed unit of length would use the (already existing) decimal system.  Stepho  talk  19:27, 1 March 2019 (UTC)

Recent edits on "seconds pendulum"

I think these are trying to say that a seconds pendulum of 993.934 millimetres replaced the physical International prototype metre - but that doesn't seem to be the case. Hence my revert. — Preceding unsigned comment added by Snori (talk • contribs) 11:19, 17 June 2019 (UTC)

That paragraph and two similar paragraphs which replaced it were placed at the end of the wrong section. I moved them to the end of the Mètre des Archives section where gravimetry, the figure of the Earth, and de Ibero himself are already extensively discussed. — Joe Kress (talk) 23:19, 17 June 2019 (UTC)

Pendulums and geodesy

Charles Inigo (talk · contribs) recently added a lot of material about pendulums and geodesy. For most readers, this is a mere sideline to the history of the metre. It distracts the reader from the history while he is trying to make sense of the technically detailed pendulum and geodesy information. This would be better served by linking to the appropriate articles instead of trying to push it all into this article.  Stepho  talk  00:53, 13 April 2020 (UTC)

-Thank you for your contribution. I took it in account when adding informations on the two first international scientific associations following mention of the use of the metre in Great Britain. Charles Inigo (talk) 08:43, 13 April 2020 (UTC)

Combinations of time and space as measures go back to antiquity, but in modern wikipedia the round offs disquise the system.

When did we begin using twice as many seconds in a century as inches in the circumference of the earth at the equator to set a standard?

We might note that subdivisions go back to the use of body measures wherein a hand is 100 mm and 5 hands make a cubit. In Egypt a foot is 300 mm. Feet and cubits begin to be used to measure a days sail in terms of stadions by the Greeks and Stadiums by the Romans and then 75 Roman miles or miliari to a degree. A Greek stadion measures 185 meters as does a Roman Stadium, and both are counted 8 to a mile and 75 miles to a degree from Britain to China in the Geography.2604:6000:1513:4FFD:D8FF:20A9:7618:15DC (talk) 11:38, 8 August 2020 (UTC)

What does this line mean?

What does the line 'and that it could complement meridian arc measurements in determining the figure of the Earth' mean? - Para 2, Lines 3-4 — Preceding unsigned comment added by Dhairya Patel0 (talk • contribs) 14:31, 6 December 2020 (UTC)

Scientific revolution began with Copernicus work. Galileo discovered gravitational acceleration explaining the fall of bodies at the surface of the Earth. He also observed the regularity of the period of swing of the pendulum and that this period depended on the length of the pendulum.

Kepler's laws of planetary motion served both to the discovery of Newton's law of universal gravitation and to the determination of the distance from Earth to the Sun by Giovanni Domenico Cassini. They both also used a determination of the size of the Earth then considered as a sphere by Jean Picard.

Christian Huygens found out the centrifugal force which explained variations of gravitational acceleration depending on latitude. He also discovered that the seconds pendulum length was a mean to measure gravitational acceleration. Earth proved to be an oblate spheroid. According to Alexis Clairaut, variations of gravitational acceleration were a mean to determine the figure of the Earth, whose crucial parameter was the flattening of the Earth. When the length of the metre was defined in 1799, the flattening of the Earth was assumed to be 1/334.

In 1841, Friedrich Wilhelm Bessel using the method of least squares calculated from several arc measurements a new value for the flattening of the Earth, which he determinated as 1/299.15. He also devised a new instrument for measuring gravitational acceleration which was first used in Switzerland since 1864 by Emile Plantamour, director of Geneva Observatory. Charles Sanders Peirce and Isaac-Charles Élisée Cellérier (8.01.1818 – 2.10.1889), a Genevan mathematician soon independently discovered a mathematical formula to correct systematic errors of this device which had been noticed by Plantamour and Adolphe Hirsch, director of Neuchâtel Observatory. This allowed, Friedrich Robert Helmert to determine a remarkably accurate value of 1/298,3 for the flattening of the Earth in 1901 more than 25 years after the metre had been adopted as an international scientific unit for the convenience of continental European geodesists following Guillaume Henri Dufour, Ferdinand Rudolph Hassler and Carlos Ibáñez e Ibáñez de Ibero example. Charles Inigo (talk) 06:37, 28 September 2023 (UTC)

Translation

I identified a problem of translation of the source n°5. I translated the corresponding passage from the Wikipédia article Histoire du mètre in French along side with additional material. I have not yet inserted the sources which can be found in the article in French. Charles Inigo (talk) 09:34, 15 October 2023 (UTC)

Airy points: tagged as dubious

I've tagged the following sentence as dubious:

The support requirements represent the Airy points of the prototype—the points, separated by 4⁄7 of the total length of the bar, at which the bending or droop of the bar is minimised.

The reference provided, Phelps 1966, does not back this up, instead stating "the Airy points are symmetrically spaced and separated by the distance ${\textstyle {\frac {l}{\sqrt {3}}}}$." (Note ${\textstyle {\frac {1}{\sqrt {3}}}\approx 0.577}$.) It also mentions "[Bessel showed that] if two supports were used they should be separated by 0.5863 of the length of the bar if the shortening of the bar due to flexure is to be a minimum." Neither of these statements back up the sentence in question.

Airy_points#Bessel_points notes "[support points separated by 571mm] would be the Bessel points of a beam 1020 mm long", suggesting the International Prototype Metre bar (with markings exactly 1 metre apart) is supported at its Bessel points, and should be 1020 mm long. However, we don't currently have a good source confirming this.

Lastly, the provenance of the 4⁄7 claim is as follows: it was added (as 5⁄9) in 22:58, 17 August 2010, followed by a correction to 4⁄7 in 09:59, 26 September 2014. Neither figure matches the bar length proportions of the Airy or Bessel points. Preimage (talk) 15:55, 26 August 2024 (UTC)

Good catch. It looks as if the whole sentence should go. I looked at the report of the 7th CGPM in the hope of finding a hint as to why 571 mm was chosen, but couldn't see any (but my French is not good). NebY (talk) 16:34, 26 August 2024 (UTC)

Thanks. I can now confirm the bar is 102 cm long and is supported at its Bessel points, as per the following three references:

Direction générale des Entreprises. "Histoire du mètre" [History of the meter]. metrologie.entreprises.gouv.fr (in French). Le prototype du mètre est une règle, dont la section a la forme d'un X à talons, comportant une ligne axiale (fibre neutre) ni tendue ni comprimée (selon la théorie des moments de flexion) quand la règle est légèrement fléchie, et qui conserve la même longueur de 102 centimètres, sur laquelle deux traits transversaux marquent les extrémités de l'unité. [The prototype of the meter is a ruler, whose section has the shape of an X with heels, comprising an axial line (neutral fiber) neither stretched nor compressed (according to the theory of bending moments) when the ruler is slightly bent, and which retains the same length of 102 centimeters, on which two transverse lines mark the ends of the unit.]

"National Bureau of Standards Replica Meter Standard". Smithsonian Institution. This aluminum bar, with an X-shaped cross-section, is a replica of the platinum international meter prototype housed in Paris and used as a standard for the metric system from 1889 to 1960. ... Like an actual meter standard, the bar is 102 centimeters long and there are marks 1 centimeter from each end on this side to show the precise length of a meter.

Page, Chester H.; Vigoureux, Paul, eds. (1975). The International Bureau of Weights and Measures, 1875-1975 : translation of the BIPM centennial volume (PDF). U.S. Dept. of Commerce, National Bureau of Standards. p. 67. The distance specified between the support rollers is chosen to minimize the shortening of the standard by flexure under its own weight.

If we're going to mention how long the bar is, its cross-sectional dimensions may also be of interest to readers: 16 mm x 16 mm, as per the following reference:

Gupta, S. V. (2020). Units of measurement: history, fundamentals and redefining the SI base units (2nd ed). Springer. p. 108.

Preimage (talk) 18:34, 26 August 2024 (UTC)

References