Talk:Free space/Archive 1

difference between Patent office and technical definition
I thought it would be obvious but " the atmosphere, the ocean, or the earth" are not free of matter. Whether or not something is a good approximation to free space depends on the wavelength, for example, the earth is far from free space for visible light. Salsb 15:01, 7 December 2005 (UTC)

Thanks ... I would guess also that the "vacuum" of space (which contains a small proportion of elements) could be a approximation to "free space". J. D. Redding
 * True, but a incredibly much better approximation for all wavelengths. In terms of density, comparing outer space to air is like comparing a neutron star to a medium vacuum! Salsb 16:42, 7 December 2005 (UTC)
 * I don't think I did make such a comparision. Anyways ... nothing is totally "free of matter" and, more importantly, this is in the article. Sincerely, J. D. Redding 17:03, 7 December 2005 (UTC)

Sources or substantial revision needed
The statement in the second paragraph of the definition article says that "The notion of free space does not correspond to the present-day understanding of what is called the vacuum state or the quantum vacuum". So there's the "partial vacuum" that people make in their laboratories, and the "ideal physical vacuum" (or whatever the proper term is) which is the limit of a partial vacuum as temperature and pressure go to zero. (So far this is widely accepted and uncontroversial.) But then defining "free space" to be the ideal physical vacuum without quantum fluctuations seems to me to be quite unusual and unconventional definition. Moreover, making the claim that the speed of light and vacuum permittivity are defined with respect to "free space" (and not the ideal physical vacuum) is a statement of fact which is unsourced, and which appears to contradict many reliable sources. For example, it's implicit in the SI definition of the meter that by measuring the speed of light in a better and better vacuum, you should get a better and better measurement of the meter standard. Therefore, they presumably intend the speed of light to be defined in the ideal physical vacuum, not free space. (If someone did a theoretical calculation that says that turning off quantum fluctuations would change the value of the speed of light by 20%, that would not mean that everyone in the world would have to order new meter-sticks and redo all their measurements!)


 * First, someone needs to find a source that uses the same definition as the one in this article, to show that this isn't the only place on earth which is using "free space" as synonymous with a vacuum with quantum fluctuations turned off. In fact, as best as I can tell, it seems like the term bare vacuum is much more common (at least in QFT literature), and I'd suggest that the article be renamed accordingly, unless sources say otherwise.


 * Second, the claim that the vacuum permittivity and speed of light are defined with to free space, and not the ideal physical vacuum, needs to be sourced or deleted. I'm skeptical, for the reason mentioned above.


 * Third, outer space and laboratories are not approximations to "free space" as defined here. They are approximations to the ideal physical vacuum. There's a big difference: For example, in "free space" the charge of the electron is not renormalized by virtual photons, and turns out to be infinitely large. Likewise, the masses of particles in "free space" are not renormalized by virtual processes, so they're the "bare masses" instead of physical masses, and these are also infinity, I think. Therefore, these sections should be moved to the vacuum article, or deleted, unless someone finds a source for these claims.

Alternatively, and perhaps better, the definition could be changed to define "free space" as the ideal physical vacuum. This would, however, entail some redundancy with the vacuum article, which could be minimized by appropriate editing and cross-referencing. What do people think is the better direction to go with the article? --Steve (talk) 23:55, 29 February 2008 (UTC)
 * Well, there is way too much here to tackle all at once. And the discussion is likely to go astray because of words like "vacuum" that have multiple definitions. I'll pick the third item as a starting point:
 * Third, outer space and laboratories are not approximations to "free space" as defined here. They are approximations to the ideal physical vacuum.
 * Let me start by suggesting the term "quantum vacuum" for the physically observable vacuum that actually occurs in nature. That definition seems to be in accord with the articles. So this convention on definitions replaces the term "ideal physical vacuum". Then outer space and laboratories are approximations to the quantum vacuum, only approximations because they cannot pump down to low enough pressures and cannot remove entirely things like background fields, neutrinos etc. I'd guess that agreement exists on that point?
 * With the quantum vacuum in mind, I agree with the first part of the statement, namely, Third, outer space and laboratories are not approximations to "free space" . So I'd guess that there is actually no debate here at all – we agree on point three. Brews ohare (talk) 17:13, 1 March 2008 (UTC)
 * Let's call the above Response to Point 3. Now let's go to Point 1:
 * using "free space" as synonymous with a vacuum with quantum fluctuations turned off. In fact, as best as I can tell, it seems like the term bare vacuum is much more common
 * Most authors simply refer to "free space" as a synonym for "empty space", and references 1-3 in the article take pains to distinguish that view from the quantum vacuum as defined in Response to Point 3. I like your term bare vacuum for this case. My view was that the common interpretation of free space was bare vacuum. I suspect that most EM books and even NIST itself is unclear whether or not free space is quantum vacuum or bare vacuum. In fact, I doubt that this distinction even occurs to them, never mind reaches discussion. I'd like a bit more on what you think should be done about this. A good choice might be to define free space as the same thing as quantum vacuum. What would be the ramifications? Brews ohare (talk) 18:57, 1 March 2008 (UTC)
 * And here is a whole book about it. Brews ohare (talk) 22:20, 1 March 2008 (UTC)
 * I revised this page; see how you like it. Besides free space, electric constant and magnetic constant have been slightly altered. Basically most EM articles refer to free space, so this is the key article to get straight. Brews ohare (talk) 23:01, 1 March 2008 (UTC)

Hi again! Thanks for responding. I haven't read your changes yet, but before I do I wanted to share the following bit of research that I just did.

Evidence that NIST means quantum vacuum, not bare vacuum, when defining speed of light and vacuum permittivity: If you look at p46, NIST says that in defining the meter via the speed of light in vacuum, "that in all cases any necessary corrections be applied to take account of actual conditions such as diffraction, gravitation or imperfection in the vacuum". It doesn't mention quantum fluctuations (and certainly doesn't advocate theoretical studies of speed-of-light renormalization), but nor does it explicitly say otherwise. But it's clear from the rest of the pamphlet what they mean by "vacuum". For example, if you flip to p51, it lists the "vacuum wavelengths" for certain atomic transitions. Without virtual particles, the values of atomic transitions are very different (since there's no virtual photons, the electron has infinite charge, etc.), so here they must be talking about quantum vacuum. On p62, it talks about weighing against the kilogram mass standard "in vacuum". Without virtual particles, the mass of each proton and electron is infinity, making such a comparison impossible. So I'd say there's substantial evidence that NIST means the quantum vacuum. [Just to be clear, as a matter of physics, I don't believe that quantum fluctuations affect the speed of light...but if some future theoretical advance indicated that it did, I believe that this wouldn't have any bearing on the NIST standard.]

I hope I get a chance later to comment more.--Steve (talk) 23:41, 1 March 2008 (UTC)
 * You make good points; the revisions just made in free space, magnetic constant and electric constant bring these articles to coincide with your views. Brews ohare (talk) 23:45, 1 March 2008 (UTC)
 * By the way, the point that the nature of the vacuum affects predictions of  atomic transition seems very significant to me - that should come up in the discussion of frequency standards somewhere -interested in looking at that? It seems worthwhile. Brews ohare (talk) 00:53, 2 March 2008 (UTC)

I have no objections to the article as currently written. Thanks for putting in the time to improve it! :-) --Steve (talk) 02:45, 3 March 2008 (UTC)

Brews ohare, you did a great job on this article. J. D. Redding 08:17, 17 April 2008 (UTC)

make alil more clear
moved from article
 * In another sense, it can be thought of the equalization of pressures. Such as a higher concentration of gas will equal out to a lower container or area of gas to achieve a nature of equality.  The same goes for free space, except free space is the bare bottom line of taking and not giving.  Its so abstract that its literally impossible to achieve equality of pressure, however, if one were to take all pressure in the universe and release it to all areas of free space and the such, a nature of equality would be obtained basically impossible.

J. D. Redding 23:37, 28 September 2008 (UTC)

D Tombe additions
I find the discussion of Maxwell's views of permittivity rather irrelevant in this context, which is about free space, not dielectric constants. In particular, the parameters of the SI system were not used by Maxwell. Perhaps the now outmoded names vacuum permittivity etc. have proved misleading. These constants now are defined values and not the observed or measured properties of anything real. That is why they now are called the magnetic constant, electric constant etc.  Brews ohare (talk) 06:57, 21 November 2008 (UTC)


 * Brews, Electric permittivity is essentially just the reciprocal of the dielectric constant. I suppose I could have got away with using the word 'permittivity', but then somebody might have tripped me up on that minor technicality.
 * At any rate, permittivity and permeability are both essential paramaters in Maxwell's equations, as has been stated in the main article. I see no harm in mentioning what those parameters meant to Maxwell. After all, he was the one that derived the equations.
 * In modern physics, we still use Archimedes' principle even though Archimedes was an ancient Greek. We still use Newton's laws of motion and Newton's law of gravitation, even though these are a few hundred years old. Nobody ever complains that they are old fashioned.
 * Maxwell's equations are much more recent. They date back only to the American Civil War. They still form the backbone to modern electromagnetism, so I can't see what the problem is about mentioning Maxwell's views on permittivity and permeability. It is a genuine point of interest for readers. David Tombe (talk) 05:29, 22 November 2008 (UTC)

Free space ≠ Quantum vacuum?
This is a continuation of the conversation earlier on this page. That conversation had ended, I thought, with the conclusion that "free space" was a synonym of "quantum vacuum". (By "quantum vacuum" I mean the extrapolated limit of a physical vacuum as temperature and pressure go to zero.) But the article now says otherwise.

In particular, the section "Realization of free space in a laboratory" seems to imply that the "operational definition of free space" is identical to the definition of the quantum vacuum. OTOH, the section "What is the vacuum?" says that they are different.

CIPM's definition of free space is, it seems to me, the same as the "operational definition of free space", which in turn is the same as the quantum vacuum. I think this is implicit in the quote from the article, and it's also implicit in the work of all the precision metrological physicists who are encouraged by CIPM and who use the "operational definition" of the speed of light for their measurements.

My view is that CIPM, NIST, and the physics community at large use "free space" as another word for "quantum vacuum". The values of ε0, μ0, c0 are defined by measurements in free space, not the other way around. It certainly can't be the case, as this article currently implies, that free space is defined by having the properties ε0, μ0, c0, while at the same time ε0, μ0, c0 are defined by being the properties of free space! That would be an empty and circular definition. Instead of that circular definition, I propose that free space is defined by the real world, systematically extrapolated to a particular limit. I don't see how any other (non-circular) definition of free space is possible. I hope the article can be changed to reflect this. --Steve (talk) 23:14, 2 February 2009 (UTC)


 * This topic is not simple. According to international accord, free space has various defined properties expressed by magnetic constant and electric constant, for example. One consequence of these material parameters is the principle of superposition: the field due to many charges is the sum of the fields due to each in isolation, regardless of the field strengths encountered.


 * Although experiment has yet to confirm it, quantum vacuum is predicted to have various properties such as field dependence at large fields (violates superposition), dichroism and so forth that are not consistent with material properties equal to μ0, ε0.


 * Thus it would appear that free space in conjunction with Maxwell's equations as we know them today ultimately may well not be consistent with quantum vacuum. However, it is not necessary that free space be a realizable (achievable) state.


 * The article adopts the view that free space is an unachievable reference state, not a realizable state. Real measurements on real media can be referred to this reference state of free space by following "best practices" (a BIPM term) for preparation of an approximation to free space and implementing the appropriate corrections to account for the imperfections of the realization. At present these corrections are made primarily for inability to achieve zero pressure, but in the future such corrections might include corrections to refer an approximate realization of quantum vacuum to the reference state of free space. For example, such corrections could include corrections for field dependence at large field strengths, for polarization dependence, or for size effects introduced by the the quantum vacuum in interaction with its container. Brews ohare (talk) 06:00, 4 February 2009 (UTC)


 * Hmm, first of all I'd say that at the enormous electric fields where there's known vacuum nonlinearity, there are zillions of photons present, so it's not really a close approximation of a quantum vacuum anymore. :-) More importantly, μ0, ε0 etc. cannot possibly be "defining properties" of "free space", because of the circular definition problem I describe above.


 * Would you agree with the statement: "Free space = the quantum vacuum in a very large volume and with small electric and magnetic fields"? --Steve (talk) 07:24, 4 February 2009 (UTC)

Steve: There is no circular definition problem here. One specs μ0, ε0 for ideal "free space" and then supplements this by a preparation procedure and a set of corrections to reach the reference state. One might argue over whether the corrections to the prepared state were accurate for a particular preparation methodology, but that is not circular. There is no logical or practical advantage in choosing some realizable "standard" state as a standard, like quantum vacuum: one still has to attempt to prepare the "standard" quantum vacuum and make the appropriate "best practices" corrections to the state actually used in order to refer your measurements on your particular sample to the "standard" quantum vacuum. Brews ohare (talk) 14:12, 4 February 2009 (UTC)

A subsidiary question is "what actually is meant by BIPM or NIST when they refer to the vacuum??" And beyond that, what does, say, Jackson, mean? The use of the term "definition" with regards to the properties of free space and avoidance of error bars, suggests an ideal reference state. Jackson also refers to the principle of superposition, again without implication of any approximation. It is arguable that "quantum vacuum" has none of these properties unless one imposes restrictions on fields etc. that automatically introduce error bars upon how accurately these limits allow the defined properties. Brews ohare (talk) 14:20, 4 February 2009 (UTC)


 * If there's a preparation procedure and set of corrections to reach free space, then I'd say that's the definition of free space. Free space is what you get from following that preparation procedure and set of corrections. If this is true (and I believe it is), then it can't also be true that free space is a hypothetical area of a hypothetical universe with different laws of physics such that Maxwell's equations are exact, there's no light-dispersion, etc. These are entirely different definitions, and you can't simultaneously insist on both. Clearly BIPM believes that you can follow the preparation procedure and set of corrections to measure a meter. So BIPM believes that "vacuum" is what you get by following that set of procedures. They even specify the set of procedures! They certainly don't say "Maxwell's equations are exact in a vacuum, whether you measure it that way or not." They also talk about atomic transition energies "in vacuum", and the nonlinear vacuum polarization has a finite effect on these transition energies, so obviously here they mean quantum vacuum. Jackson goes on and on about whether and to what extent it's true that Maxwell's equations hold exactly "in vacuum", so by "vacuum" he's also talking about what you get by following the preparation procedure and corrections, not part of a hypothetical universe where Maxwell's equations automatically hold exactly.


 * As far as I can tell, the only reason you're insisting that free space must automatically by definition satisfy Maxwell's equations is that BIPM, NIST, etc. say (for example) c0 is the speed of light in vacuum, ε0 is the permittivity of the vacuum, etc. You're stuck on the one word, "the", and based on this one word you've come up with the strange idea that Maxwell's equations must hold in free space, regardless of whether or not Maxwell's equations actually hold in the real world. And if Maxwell's equations don't hold in the real world, then free space is part of a hypothetical universe with different laws of physics.


 * I think it's more plausible that BIPM, NIST say "the" permittivity of vacuum, because as far as anyone can measure it's constant, and if it's discovered to be non-constant, they'll simply say "the permittivity of vacuum at (blah) field and (blah) frequency". :-) --Steve (talk) 16:47, 4 February 2009 (UTC)

Preparation procedures
''If there's a preparation procedure and set of corrections to reach free space, then I'd say that's the definition of free space. Free space is what you get from following that preparation procedure and set of corrections. '' I'd agree with this statement, although it does not imply that free space is a physically realizable state. So, as an example you have suggested, if one believes that measurements in a gas approach measurements in free space as pumping down continues, one can extrapolate to free space even though zero gas pressure is beyond one's technology.

One's belief in this pumping-down and correction procedure ultimately boils down to adopting a theory of the relation between gas pressure and permittivity, permeability (for example, constitutive relations). If one has a theory of quantum vacuum, it also will provide estimates of departure from free space, and so also can be corrected to refer to free space, even though free space has much simpler EM behavior than does quantum vacuum. Brews ohare (talk) 19:25, 4 February 2009 (UTC)


 * Great. I agree that free space is not physically realizable in practice, just like absolute zero. There are two differences that I see between correcting for gas pressure and correcting for "quantum-vacuum-effects". First, whatever theory you have for the proper extrapolation of gas pressure measurements, this theory will be testable. If you think the extrapolation will break down at some finite pressure, you can test that by creating a better vacuum pump capable of that pressure. On the other hand, a correction for quantum-vacuum-effects may well be untestable even in principle, because you can't ever build an apparatus that exists in a different universe with different non-quantum laws of physics. Second (and relatedly), it's not the case that Maxwell's equations are the basic laws of the universe, except that there are quantum corrections on top of that. On the contrary, we live in a fundamentally quantum universe, and it happens to be the case, we think, that under certain circumstances these true quantum laws approach Maxwell's equations. But we don't know that for sure, and even if there were a regime in which the true quantum laws approach Maxwell's equations, this regime might not be unique...for example, the limit of small field strengths, large volumes, and ultra-low frequencies might be describable with Maxwell's equations, and the limit of small field strengths, large volumes, and moderate frequencies might be describable with Maxwell's equations but with a different speed of light. Or maybe there's no regime in which the true laws of physics approach Maxwell's equations, because there's a fundamental anisotropy that's too small to be measured so far. Implying that you can come up with an experimental procedure that will get closer and closer to exact Maxwell's equations is a bold statement of physics, and that makes it very different from absolute zero, quantum vacuum, or other such. It is the case that an imperfect vacuum is a quantum vacuum with particles in it, and you can do a better and better job of purging the particles. It's not necessarily the case that the laws of physics are Maxwell's equations with quantum corrections, and you can do a better and better job of purging the quantum corrections.


 * I think of your definition sorta like a classical thermodynamics expert defining "absolute zero" to be the regime where all atomic motion ceases and each particle has a fixed, rigid position. Then someone points out that there's always quantum "motion" which cannot be stopped by lowering the temperature, and the expert responds by saying "OK, then I guess you can't get to absolute zero by just lowering the temperature, you need to come up with procedures to minimize quantum motion too!" :-) --Steve (talk) 21:46, 5 February 2009 (UTC)

Hi Steve: I don't have to be persuaded that Maxwell's equations may be approximate. I do debate the difference between pumping-down corrections and quantum vacuum corrections. It is not necessary to check out another universe to evaluate quantum vacuum effects. It is testable. Suppose, for instance that we calculate a polarization dependence of quantum vacuum. If our experimental apparatus is good enough, I am sure we can determine there is a different speed of light for different polarizations. That applies to field nonlinearities and dispersion too. These effects indicate not only a departure from c, but qualitative effects not described by a field and frequency independent scalar formulation of "vacuum". My point is that these attributes do not require any esoteric notions.

I am unsure whether you are an advocate of a "vacuum" that can be approached to within infinitesimal differences, and if free space with its ε0 and μ0 can't serve that purpose, maybe quantum vacuum can? Or, are you suggesting that is my hang-up? I'd say that, if you have a theory with parameter λ in it, and recover free space when λ→0, what does it matter whether we can identify a sequence of realizable states with ever smaller λ's? The adjustment of real measurements to a "kosher" measurement in free space is done by subtracting the λ corrections from the measurements, no??

The use of the free space standard is not a statement about what some physical ideal is. Its just an origin of coordinates. Brews ohare (talk) 22:59, 5 February 2009 (UTC)


 * For the quantum vacuum testability, of course you're right that a departure from Maxwell's equations can be tested. I guess what I mean is, for example, if c is found to differ between high and low frequencies in quantum vacuum, you can't experimentally test whether the low-frequency c is the free-space value while the high-frequency c is sullied by quantum corrections, or whether it's the other way around.


 * It's my belief that we don't know for sure whether or not there's a theory with a parameter λ in it, such that we recover free space when λ→0, and if there is, whether this theory is unique. It's also my belief that BIPM is strongly implying through their writing that we can identify a sequence of realizable states with ever smaller λ's. This is something that I imagine a standards organization would want, for practical reasons. --Steve (talk) 02:26, 6 February 2009 (UTC)

Comparison to kilogram
I think if we're trying to dissect the words of BIPM, a useful comparison is the kilogram. There's a block of metal in a case in Europe called the "IPK", and it's defined to have a mass of exactly one kilogram (after you wash it off). Now everyone knows that after washing off the IPK, its mass won't be exactly the same each time, it'll have one extra picogram of dust or whatever. But it's nevertheless exactly one kilogram each time.

So: When BIPM says "the mass of IPK is always 1 kilogram", they're not implying "we know for sure that the mass of IPK is exactly constant".

Likewise: When BIPM says that "ε0 is the permittivity of free space", I think we shouldn't infer that BIPM is implying "we know for sure that the permittivity of free space is exactly constant". --Steve (talk) 02:16, 6 February 2009 (UTC)


 * Well, it's not easy to determine what BIPM means or NIST either. My attempts to get some specifics from them have been met by pretty much a regurgitation of their public documentation that doesn't address the issues.


 * I don't agree that the kilogram is a parallel. The kilogram is an example of "old" standards and has all the problems of knowing whether it is being properly maintained, and how to do comparisons. It is these issues that have led to the adoption of the free space type of standard that does not require a physical embodiment.


 * If you define the material parameters of the standard, my view is that it is ipso facto not a real medium. As you say, it can be a practical advantage to have a sequence of preparation steps to extrapolate to the standard. However, that is not a requirement. As you know, the present approach relies heavily on the accepted theory of polarization of a gas that is traced directly to the number of molecules etc. that it contains, and so supports the notion that no constituent particles means free space material parameters. Obviously, this idea does not account for quantum fluctuations. The BIPM and NIST take the attitude that they will cross this bridge when necessary.


 * The polarization of the constituent molecules also is tied to a theoretical model, tested by other measurements with some error bars. So measurement without extrapolation also is possible by making a count of the number and type of constituents.


 * My view is that the present free space article does not contradict anything said by BIPM, is consistent with physical theory, and is logically self-consistent. Maybe its aesthetics don't agree with you, but how does a hodge podge of mixed up definitions and procedures suit you? Brews ohare (talk) 02:49, 6 February 2009 (UTC)

Free space has exact (not measured) values of ε, μ, and c
Hi Steve: I don't know if we can settle this matter. It takes cool heads, and a strong discipline to avoid rhetorical remarks. I don't think either of us qualify.

Let's begin by agreeing that BIPM and NIST go further than leaving matters to what the meaning of "the" is. For instance, ε0 is defined as a numerical value, and has its standard uncertainty listed as exact. Likewise for μ0 and c0. I don't know how a measurement of any property can have zero error bars. So I'd say the implication is that the properties of free space are not measured.

By way of contrast, assuming development of accurate techniques, ε, μ, and c of quantum vacuum certainly can be measured, and will have error bars. In fact, quantum vacuum will most likely exhibit dispersion, polarization and field dependence, impossible to represent by a singe set of scalar material properties, regardless of their numerical values.

How do you deal with this adoption of defined exact, scalar values? Perhaps the answer lies in the next topic, discussed below? Brews ohare (talk) 18:16, 4 February 2009 (UTC)


 * Hehe, I'm optimistic, I think we can work this out. :-) I agree, no need for rhetoric.


 * BIPM and NIST decided to put error bars on the definition of meter, amp, coulomb, etc. They could equally well have defined a meter, coulomb, etc. in terms of some other standard (say, a meter = 10^10 hydrogen atom diameters, a coulomb = 10^19 electron charges, etc.), in which case there would have to be error bars on ε0, μ0, and c0 and no error bars on the definition of the meter, amp, coulomb, etc. This decision, where the error bars belong, was based on what set of definitions would be most reproducible and convenient for physicists to use, and I don't think we should read into it any further than that. :-) --Steve (talk) 01:06, 5 February 2009 (UTC)


 * Better example: The charge of an electron is a measurable quantity. In atomic units, it's exactly -1. These statements are not contradictory. :-) --Steve (talk) 05:25, 5 February 2009 (UTC)

In summary, you suggest that by defining the permittivity and permeability of free space, in effect you are doing the same thing as some dimensional analysis? I'm supposing that using units where c=1 is the same idea? If this is what is going on, then quantum vacuum has nothing to do with it at all. Brews ohare (talk) 07:24, 5 February 2009 (UTC)


 * I don't think that's quite what I'm saying. I'm saying that there are bunch of different measurable lengths (c0*1 second, or a hydrogen-atom diameter, or the length of a stick in a glass case in France, etc.) The SI people need to pick one of them as the basis for the definition of the "meter". Whatever length standard they pick as the basis for the definition of the meter, from then on is going to be an exact number of meters, while all the other measurable lengths will continue to have error bars. When the SI people do that, they haven't changed any actual physics. All of these are still equally-empirically-measurable quantities. Nevertheless, one of them won't have error bars, because that's the one that they happened to choose as the definition of the meter. They chose c0*1 second, so now c0*1 second doesn't have error bars. If they had chosen a hydrogen-atom diameter as the basis for the definition of a meter, then the hydrogen-atom diameter would no longer have error bars, but the speed of light in vacuum would. That's why the exactitude of c0 isn't a statement about physics, it's a statement about the what the SI committee happened to choose for their unit definitions. --Steve (talk) 21:26, 5 February 2009 (UTC)

The standard of free space is more than choice of a length: it specifies a dispersionless, isotropic, polarization-independent case. That may transcend any realizable medium, regardless of the choice of measuring sticks. Brews ohare (talk) 23:10, 5 February 2009 (UTC)


 * Yes, this is the standard of free space as you understand it, and not as I do. Getting back to the original point of this section, you said: "I don't know how a measurement of any property can have zero error bars." Well, do you understand now? For example, the hyperfine transition radiation frequency of caesium is exactly 9,192,631,770 Hz, but is nevertheless a measurement of a property. --Steve (talk) 06:18, 6 February 2009 (UTC)

Informal RfC for 'Speed of light' article
There is a dispute concerning the wording of the 'Light as electromagnetic radiation' section of the 'Speed of light' article. Editors are requested to give their opinions on the 'Speed of light' talk page. We decided to ask on related article talk pages rather than go for a full RfC so that we would get editors with a knowledge of and interest in the subject. Martin Hogbin (talk) 18:08, 6 February 2009 (UTC)

Comparison to second
Is the caesium-atom based second a parallel? When the BIPM says that the transition radiation frequency is 9,192,631,770 Hz, is the BIPM saying "We know 100% for sure that caesium transition radiation frequency doesn't change from year to year" (as some weird proposed physics theories would have it)? Or is this in the category of "old" standards? --Steve (talk) 06:26, 6 February 2009 (UTC)


 * Hi Steve: I'd say the standard for the second is a case unto itself. It differs from the kg because, unlike the kg, any competent metrology outfit is supposed to be able to make their own. It differs from free space in that it uses an actual realizable source (although a look at atomic clock and its links to NRC will curl your hair). It also has the interesting property that there is no theory capable of predicting the transition frequency: it's entirely empirical. And it differs historically in that time after time, test after test, some new correction is needed: T=0 corrections, gravitational time dilation corrections, who knows what is next? It probably will be replaced again and again.


 * If it becomes possible to test these clocks astronomically, it seems likely to me that it will be found that just like on Earth, every environment leads to a different second. Atoms are not environment insensitive, as was the original notion. Maybe Caesium in outer space has a Lamb shift different from Earth. It is very likely that Caesium in free space (were that feasible) is a completely different kettle of fish. Having no theory, it is impossible to correct the second to refer to free space. Instead, corrections are based on comparisons between clocks and applied to gain reproducibility and repeatability.


 * When the BIPM says that the transition radiation frequency is 9,192,631,770 Hz, is the BIPM saying "We know 100% for sure that caesium transition radiation frequency doesn't change from year to year" Actually the BIPM does seemingly say that; however, it means that if you build a clock in such and such a way and implement such and such corrections (for example: This note was intended to make it clear that the definition of the SI second is based on a caesium atom unperturbed by black body radiation), we will accept (for now) that you have got a transition frequency of 9,192,631,770 Hz. The NRC attaches a "relative uncertainty" of 10−15 (possibly a linewidth/center frequency) and is suggesting use of an Sr88 ion trap instead. It would be interesting if clocks yesterday could be compared to clocks today to see if there is some relativistic drift or something. For the latest traffic updates, stay tuned. Brews ohare (talk) 12:22, 6 February 2009 (UTC)


 * Maybe you would like to undertake a comparison of this approach to that defining the properties of free space? It appears that the definition of the second includes a whole procedure plus a constantly updated list of corrections. The definition is thus a fluid thing, with a huge and essential and evolving attached context. To the extent that any clock has a linewidth, the spec of the frequency cannot be translated to an exact second in practice, even ignoring the other problems that affect where the line is centered. Brews ohare (talk) 12:59, 6 February 2009 (UTC)


 * I've added some links and references to second. Brews ohare (talk) 14:11, 7 February 2009 (UTC)

Thinking about the standards for the second vs. free space, one standard is a set of procedures for arriving at a mechanism that produces a certain frequency; the other is a set of procedures that lead to an extrapolation to a certain ideal medium. Both are subject to errors of both a random and a systematic nature, and both types of error can be discovered using multiple determinations with the same system and by comparing with other systems that should give the same results. Random errors lead to error bars, while clustering of measurements into different groups suggest possible systematic errors in preparation of the mechanism or extraction of the results. So the Cesium ion used as a standard can be achieved in principle, but its realization can never be verified beyond some error bar. On the other hand free space never can be realized, and what is more, the extrapolation to free space cannot be verified beyond some error bar. Brews ohare (talk) 20:31, 7 February 2009 (UTC)

Quantum vacuum as a standard vs. free space
Quantum vacuum: A theoretical εij(E, B,ω), μij(E, B,ω) for some ideally prepared quantum vacuum might be used for a reference state. This new reference state might be an unrealizable state that is approached only asymptotically by experimental realizations; for example, by pumping down a gas to lower and lower pressures and taking precautions about shielding the vacuum and correcting for the containment dependence of vacuum fluctuations, among other things. Adopting Maxwell's equation for EM behavior of this hypothetical medium, the behavior of EM radiation in this medium can be related to εij(E, B,ω), μij(E, B,ω). (Somewhat beside the point, here's two papers to think about: Polarizable-Vacuum (PV) representation of general relativity, Dielectric Theory of the Vacuum).

Free space: The properties of an ideal, linear, isotropic, dispersion-free and polarization-independent medium can be defined by a particular permittivity (say ε0) and permeability (say μ0). Adopting Maxwell's equation for EM behavior of this hypothetical medium, relations for Z0 and c0 can be found relating these properties directly to ε0, μ0.

The present position of BIPM as far as I can tell is that "free space" is interesting because the theory of EM behavior in all real media to within present experimental accuracy is determined from constitutive relations that can be measured experimentally and in some cases calculated theoretically. The theory and experiment both suggest that the ε & μ of real media trend toward those of "free space" if an appropriate limiting process is employed, for example, pumping down a gas to lower and lower pressures.

These ideas are adequate given the present accuracy of such measurements. However, as technique improves, simply pumping down will not lead to a unique reference state without other precautions (e.g., extrapolating to zero field as well as zero pressure) and crude extrapolation will result in different measured values unless additional precautions are observed, simply because quantum vacuum is more complicated than free space.

In the future, the practicality of retaining free space vs. switching to quantum vacuum as a reference will have to be assessed. If it is not too difficult to do, it may still be practical to retain free space as a standard and simply state real measurements as a departure from this standard, rather than as a departure from quantum vacuum. That correction could be a theoretical adjustment based upon the calculated departure of the asymptotic quantum vacuum from free space. Brews ohare (talk) 15:22, 7 February 2009 (UTC)


 * First, there are infinitely many different ideal, linear, isotropic, dispersion-free and polarization-independent media. Each would yield a meter with a different length. The low-frequency properties of the quantum vacuum might approach one free-space-type medium, and the high-frequency properties might a approach an entirely different one. How do you decide which is the actual "free space" that BIPM is talking about? I think that this is the biggest problem with your definition.


 * Second, the evidence from the "second" and "kilogram" is that BIPM is happy to define a unit of measurement based on something that may or may not be, strictly speaking, unique and constant, but is constant according to the best modern experiments. And when it stops being constant according to the best modern experiments, they add additional precautions and specifications into the definition, as they have many times in the past (mandating relativistic corrections, etc.). What's your basis for thinking BIPM's statement "c0 is the speed of light in vacuum" is so different from those other cases, "1 second is the length of time for (blah) cesium-hyperfine-radiation cycles" or "1 kilogram is the mass of the washed IPK"? --Steve (talk) 16:53, 9 February 2009 (UTC)


 * Hi Steve:I agree that there are many media; at the moment you don't have to decide which is the "right one" because technique is too primitive to distinguish among them. So there is no problem yet.
 * I think the BIPM definition is exactly as stated in your second paragraph. However, it is a definition, whatever problems it may lead to in the future. I think we agree on that? So maybe I don't see any difference in the two standards other than free space involves an extrapolation procedure and the second does not.
 * So the problem we have is that I would say there is zero dispersion, anisotropy, nonlinearity etc. in free space, because that is the definition. You apparently think the definition is wherever you end up after doing the extrapolation. I'd say you never can establish the extrapolation actually has reached free space (there are error bars), so it remains a tantalizing unobtainable objective, with its tantalizing properties, not of this Universe.
 * And, of course, if you push things by looking a absorption lines in deep red-shifted objects, you may discover that no resemblance to "free space" can be found in astronomy, only approximations.
 * Likewise, you can never establish your clock measures a second, because you always have a linewidth, not to mention some systematic errors in where the line center is, the subject of an ever-lengthening list of "corrections".  Brews ohare (talk) 04:50, 10 February 2009 (UTC)

Jackson thinks c=299792458 in the quantum vacuum
Jackson page 3: "Essential to electrodynamics is the speed of light in vacuum, given in SI units by c=(ε0μ0)-1/2. As discussed in the Appendix, the meter is now defined in terms of the second (based on cesium-133) and the speed of light (c=299792458 m/s, exactly). These definitions assume that the speed of light is a universal constant, consistent with evidence (see Section 11.2.C) indicating that to a high accuracy the speed of light in vacuum is independent of frequency from very low frequencies to at least 1024 Hz."

A couple pages later, as you know, he discusses at length the deviations from the superposition principle in "vacuum", so clearly he interprets "vacuum" as "quantum vacuum".

Jackson, like me, seems to takes the point of view that the statement "the speed of light in vacuum is 299792458" to be not a statement that the vacuum is a weird non-physical entity where Maxwell's equations hold automatically. Instead, he views the statement as containing an "assumption" about the properties of the quantum vacuum; an assumption supported by all the experimental evidence so far. If he were talking about "free space", he wouldn't have to bring up the evidence of the measured constancy of the speed of light, and there would be no "assumption" whatsoever in the SI definition.

So can we agree that Jackson understands the term "vacuum" as quantum vacuum", and understands the statement "c=299792458 m/s, exactly" as a statement about the quantum vacuum? --Steve (talk) 16:39, 9 February 2009 (UTC)


 * Steve: Inasmuch as Jackson states unequivocally that superposition holds in what might be called "free space" (see also: Cottingham, Kenyon, Ahmanov and many more), and that superposition (at least theoretically) is not a property of quantum vacuum (εij(E, B,ω) ≠ ε0, etc.), I would object strenuously to saying Jackson believes these two are the same thing. I don't have my copy of Jackson with me, so I cannot corroborate that he uses the word "vacuum" interchangeably for both concepts. However, if he does that, there is no need to propagate a failure to maintain a distinction where there clearly is a distinction to be made. I'd suggest maybe Jackson suffers from sloppy English, using "vacuum" with multiple meanings and depending upon the reader to use context to tell which meaning is meant. Brews ohare (talk) 03:15, 10 February 2009 (UTC)


 * I'm not sure where you're getting that Jackson unequivocally states that superposition holds. Here's excerpts from Section 1.3, "Linear Superposition", pages 9-13:


 * The Maxwell equations in vacuum are linear in the fields E and B. ...There are, of course circumstances where nonlinear effects occur--in magnetic materials, in crystals responding to intense laser beams.... But here we are concerned with fields in vacuum or the microscopic fields inside atoms and nuclei.


 * What evidence do we have to support the idea of linear superposition? At the macroscopic level, all sorts of experiments test linear superposition at the level of 0.1% accuracy--groups of charges and currents produce electric and magnetic forces calculable by linear superposition, transformers perform as expected, standing waves are observed on transmission lines--the reader can make a list. In optics, slit systems show diffraction patterns.... At the macroscopic and even at the atomic level, linear superposition is remarkably valid.


 * It is in the subatomic domain that departures from linear superposition can be legitimately sought. ...In attempting to avoid infinite self-energies of point particles, it is natural to speculate that some sort of saturation occurs, that field strengths have some upper bound....One well-known example is the theory of Born and Infeld. The vacuum is given electric and magnetic permeabilities, [equation where epsilon and mu are functions of B and E.] ...There is no evidence of this kind of classical nonlinearity. The quantum mechanics of many-electron atoms is described to high precision by normal quantum theory with the interactions between...electrons and electrons given by a linear superposition of pairwise potentials. ...At the present time there is no evidence for any classical nonlinear behavior of vacuum fields at short distances.


 * There is a quantum-mechanical nonlinearity of electromagnetic fields that arises because [talks for a while about photon-photon scattering, including an eqn for field-dependent epsilon and mu].


 * In analogy with the polarization P=(D-E)/4pi, we speak of the field-dependent terms in [above equation] as vacuum polarization effects.....


 * The final conclusion about linear superposition of fields in vacuum [italics are in the original] is that in the classical domain of sizes and attainable field strengths there is abundant evidence for the validity of linear superposition and no evidence against it. In the atomic and subatomic domain there are small quantum-mechanical nonlinear effects whole origins are in the coupling between charged particles and the electromagnetic field. They modify the interactions between charged particles and cause interactions between electromagnetic fields even if physical particles are absent.


 * Reading this, I think it's 100% clear that Jackson does not think the superposition principle is true "in vacuum" by definition. Can we agree on this? Maybe later in the book, having already spent four pages on the topic, he casually invokes superposition without further discussion. I imagine that's what you're referring to. But I think we should take that in context, and understand that the four-page section is his full opinion, and he can casually invoke superposition in the context of applying Maxwell's equations, knowing that the readers have already read that four-page section and understand how to interpret it. :-)


 * Also, back to the previous quote: You think he has two definitions of "vacuum". Which one do you believe he's using when he says the definition "c=299792458 m/s, exactly...assume[s] that the speed of light is a universal constant, consistent with evidence"? If he's discussing free space, it would be incorrect for him to have said that the definition "assumes the speed of light is a universal constant, consistent with evidence"...If it's a constant by definition, then there's no assumption and no need for evidence. If, OTOH, he's discussing the quantum vacuum, then apparently Jackson thinks the SI definition of the meter refers to the quantum vacuum, not free space. So which is it? :-)


 * As for your book references, Cottingham calls superposition an "experimental law". If something is true by definition, then it's not an "experimental law", right? Ahmanov also explicitly says superposition is not exactly true "in vacuum", but is true in Maxwell's equations, and "we shall consider it to be valid for electromagnetic waves in vacuum" (presumably because the book won't be considering ultra-intense fields). Of course there will be plenty of examples of books where the whole book is clearly within the Maxwell's-equations approximation to electromagnetism, and then they'll say that superposition is true, but I don't think we should read too much into it...At the end of the day, the books always say that Maxwell's equations are approximate empirical laws, and it's redundant to repeat that clarification after every equation. --Steve (talk) 08:04, 19 February 2009 (UTC)

Another way to put it
Here's another way to put my problem with your definition.

I have a "friend" who believe that the effect of the quantum vacuum is to slow down light by a factor of exactly 2. Any measurement made in our quantum universe is uniformly affected by this. For example, as you pump out a real-world vacuum better and better, the true index of refraction goes from 2.1 to 2.001 to 2.0000001, approaching the eventual limit of 2 (relative to free space which doesn't have this quantum-vacuum effect). Textbooks and tables will have to be revised, of course. And according to the free-space definition of the metre, every metre stick is actually two metres long.

What would you say to my friend? Do you agree that, if his theoretical calculations are correct, then a metre is actually twice as long as we thought? Can you propose any experiment that can prove him wrong? --Steve (talk) 19:09, 15 July 2009 (UTC)

Dicklyon's change to lead
The statement added by Dicklyon:
 * This vacuum can be considered either as a "classical vacuum" or as a "quantum vacuum", depending on what properties are of interest.[1]

does not apply to this article. Free space as used throughout texts on EM usually does not refer to "quantum vacuum", but to a fictitious medium with μ0 & ε0. Originally, empty space and free space were the same thing, but that is not true today.

The source cited to support this statement, Weiglhofer in fact makes the above statement about the word "vacuum", and vacuum normally is a much broader concept. He parenthetically adds free space in the sense of empty space, but he really does not use this term in a technical sense in the book. A bit later this source goes on to discuss "classical vacuum", which is synonymous with "free space" as used in EM texts, and which is the subject of this article, and there says the "classical vacuum should not be called a medium after all, ...[but is] more appropriately designated as a reference medium.

Consequently, I have removed this statement and added classical vacuum with a reference to this source. Brews ohare (talk) 13:18, 5 September 2009 (UTC)


 * You put a different page of the same reference – a page that doesn't mention free space. I had looked for a source that would indicate that free space and classical vacuum were alternative terms for the same thing, but that's not what I found; so I put what I found.  If you change it, you should find a source that supports the change.  Or add a statement that of the two alternatives, this article is about one.  Dicklyon (talk) 05:13, 6 September 2009 (UTC)


 * Dick: This source on the whole uses "vacuum" in the BIPM sense of that word as a synonym for "free space" I'd say, not a more general sense of "vacuum". For instance, on p. 4 talks about "vacuum" having the "free space light velocity c, and on p. 169 uses the old terms "free-space permittivity" for εo and "free-space permeability" for μo. That usage of free space is just what the article is about.


 * Page 28, which you linked, introduces a general notion of "vacuum" and then backs away from it. It says the following:


 * "The vacuum (or free space) itself is like Janus... with quantum and classical faces. In the days before quantum mechanics ... The vacuum has become the most complex of mediums..."


 * However, he proceeds to distance himself from this view of "vacuum" saying
 * "unless one is specifically interested in QED ... it is the classical vacuum that provides...the backdrop of our work."


 * Thus, the general term "vacuum" is relegated to the background, and by and large "vacuum" is used throughout in the BIPM sense, which is the same as "free space" or "classical vacuum".


 * The source does use the term classical vacuum and I switched the page link to refer to the extended section in the text by that name. I included "classical vacuum" as a synonym in the lead. In the section on classical vacuum, the author discusses classical vacuum in terms of εo and μo which are the same parameters used to describe "free space" in the usual sense of EM. If the rest of the text already did not say so, that shows the concepts are the same. So it isn't necessary to choose between "free space" and "classical vacuum", if that is your meaning.


 * Throughout the WP articles free space is referred to specifically in connection with the meaning classical vacuum with the exceptions noted in the subsection on "US Patent Office".


 * Here are some sources that connect "free space" and "classical vacuum":
 * 1 23


 * The use of "vacuum" in the BIPM sense, that is, as synonymous with "classical vacuum", appears to be very common.


 * I don't see a reason to rewrite this article to cover the same ground as vacuum, do you?


 * Are you happy with what has been done here? Brews ohare (talk) 16:25, 6 September 2009 (UTC)


 * The page I had cited was quite explicit about the relationship between the concepts free space and vacuum and the two different conceptions of vacuum. The page you cited does not mention free space.  I'm unclear on why you didn't like the page I cited; it would be easy enough to say that of the two conception of free space or vacuum, this article focuses on the classical one.  I didn't find any source that explicitly identifies "free space" with "classical vacuum".  Dicklyon (talk) 04:52, 7 September 2009 (UTC)

Dick: Forgive me for using this format to respond. We seem to be prone to misunderstanding each other, so I've adopted this format to force me to deal with each of your points in detail. You say:
 * The page I had cited was quite explicit about the relationship between the concepts free space and vacuum and the two different conceptions of vacuum.

I tried to point out by referring to passages in this text that the source you chose begins with a broader usage of vacuum but rapidly narrows down to classical vacuum, and that usage of free space as synonymous with classical vacuum is common throughout the book. I've given the wording showing this on the page you linked, and a number of other page references indicating this is the case. I don't think the book suggests that the usage of "free space" in a broad sense of "vacuum" is the author's ultimate usage, but rather he uses it primarily in the sense of "classical vacuum". Did you look at that?

You say:
 * The page you cited does not mention free space. I'm unclear on why you didn't like the page I cited;

I switched to his discussion of "classical vacuum" in this text, because that is the subject of this article, and because he makes the important point that it is a reference state.

The page you cited begins with a description of why the author is going to discuss classical vacuum instead of vacuum in general. He uses "free space" in this introduction first as "vacuum", and then restricts it to "classical vacuum", which is the sense he uses predominantly in the book. I don't see why this author's choices should dictate the structure of the WP article.

You say:
 * it would be easy enough to say that of the two conception of free space or vacuum, this article focuses on the classical one.

It seems odd to me to take an article that is about free space in the sense of classical vacuum and take the lead sentence, which normally defines the topic of the article, to bring in a broad definition of "vacuum" that is not the subject. As you know this article does later carefully bring up the notion of the uses of the term "vacuum", and in my mind such digression properly belongs in the body of the article, not in the lead defining sentence. Do you disagree?

The present lead sentence very clearly states what this article is about. If a person wants to look at vacuum more generally, the correct article is vacuum. Do you agree?

You say:
 * I didn't find any source that explicitly identifies "free space" with "classical vacuum".

I provided links above to three sources that use free space synonymously with classical vacuum. Did you look at them? Your source also uses free space as "classical vacuum" for the most part, as explained already.

If one leaves electromagnetism, the term "classical vacuum" does not refer directly to its EM properties. See 1; 2 3 4. I haven't found a discussion in this field of EM behavior of classical vacuum except, possibly this, which I have not pursued.

So my view is that the present lead introduces the subject clearly and makes good use of your reference. Do you agree? Brews ohare (talk) 05:41, 7 September 2009 (UTC)


 * Brews, the problem I was trying to solve is very simple: the lead says "free space is a concept of electromagnetic theory, corresponding to a theoretically perfect vacuum and sometimes referred to as the vacuum of free space, or as classical vacuum", yet I don't find a source that clearly identifies that terms "free space" and "classical vacuum" as referring to the same thing. You are inferring that from the usage you find in various sources, but such inference is not the same as WP:V (it' more like WP:SYNTH).  You cite a source that says free space means vacuum, explains the two faces of vacuum, and then goes on to talk about classical vacuum; I'm not saying to let that dictate the structure of the article, but it wouldn't hurt to do similarly, or to cite a source that more specifically supports your POV that "free space" means "classical vacuum".  It looks like you could easily do so with the first two of three you linked above. Dicklyon (talk) 17:35, 7 September 2009 (UTC)

Dick: So you take it that "classical vacuum" and "free space" are different. I'd say as general terms that is correct. However, in EM the two terms overlap and mean exactly the same thing. As evidence, please notice the following:


 * 1) Your source in the section I linked talks about "classical vacuum" as a medium characterized for EM purposes by εo and μo. These parameters describe entirely the EM behavior of a medium, as this source details. They are exactly the same parameters with the same notation used for "free space" within EM and elsewhere in this text. Moreover, until recently, these parameters were called the permittivity of free space and and the permeability of free space, and are called so in this text.
 * 2) If you insist, I can provide you with huge numbers of texts that deal with "free psace" using εo and μo.
 * 3) I then can provide you with huge numbers of texts that deal with classical vacuum using the same parameters.
 * 4) I have already provided you with three references that use both terms in the same breath.

Do you need more than that?

Is this a point of persuading you, or of meeting some kind of WP standard? If the latter, inasmuch as there is no disagreement over what is going on, and the reader can see exactly the points made above, why must the point be dwelt upon? I have added a reference to Lakhtakia & Messier in the hopes that makes you happy. Brews ohare (talk) 18:17, 7 September 2009 (UTC)


 * It's about making your editing satisfy WP:V; especially when you change stuff that's sourced, you need to provide a source to back up the change. Dicklyon (talk) 21:25, 7 September 2009 (UTC)

So now with citation of Lakhtakia & Messier all is well? Brews ohare (talk) 02:08, 8 September 2009 (UTC)


 * Much better; between them the cited sources now support what's asserted there. I took out again the bit about "very large fields near a point source", since that detail is more relevant to the 1/r field thing in general than to superposition, and since none of the three cited sources included any hint of such a distraction from the point. Dicklyon (talk) 04:06, 8 September 2009 (UTC)

195.47.212.108 revisions
I believe most of these edits should be reversed, and I have done that. No basis has been laid, and links are to a page about to be deleted (self-consistent EM constants). Brews ohare (talk) 13:24, 15 October 2009 (UTC)

POV and fake citation links
I would like to draw attention to the fact that this page has various PV statements on it masquerading as scientific consensus opinion.

or example, there is the frequent assertion that:

"The concept of free space is an abstraction from nature, a baseline or reference state, that is unattainable in practice,"

which is referenced to (indirectly) a couple of philosophy textbooks or directly (ref[33]) to a page that is in fact an attack on wikipedia called "The rise of the latrines".

It appears that the author of that page is a proponent of a strange pseudoscientific idea called Aetherometry, and is a person who has long been causing problems on Wikipedia to do with editing and original research/opinion masquerading as science in articles they wrote; see []

I am tagging this page and I would advise people to take the "philosophy" posted on here with a large pinch of salt.

Liquidcentre (talk) 15:52, 24 March 2010 (UTC)

I looked over the page and there doesn't seem anything currently here that's particularly contentious or pseudoscience. In particular, free space is a theoretical abstraction that is fairly obviously not practically realizable, so that statement seems fine. I do agree that questionable references should be checked. Are there other contentious statements here to consider? --Scientryst (talk) 21:50, 24 March 2010 (UTC)


 * Yes, that claim in itself is OK, but the way it's intended and the way it comes across in context is not. The real contentious idea permeating this article is that "free space" is an imaginary place where the laws of quantum mechanics do not apply. Quantum mechanics is not a "correction" that needs to be applied to get from "free space" to the real world. The fact that we inhabit a quantum-mechanical universe is not one of the reasons that free space is unattainable in practice. It's not logical for the definition of real-world quantities ("meter" and "ampere") to depend on an imaginary universe with entirely, fundamentally different laws of physics. In my opinion this article should be split into "classical vacuum", describing what a vacuum might look like if the laws of classical electromagnetism were exact instead of approximate, and "quantum vacuum" or "perfect vacuum" describing the limit of a real-world vacuum as pressure and temperature and gravitational field approach zero. Most importantly, it would be made clear that the definitions of SI units like meter and ampere are related to the quantum vacuum not the classical one, because the classical one is a hypothetical entity that has nothing to do with the universe we live in.


 * The problem is that the person who wrote most of the article, User:brews ohare, is not a crackpot or idiot, and this article is wrong (IMO) but sort of self-consistent, plausible, and referenced. Moreover, it is hard to find any books that explicitly say that the meter is not defined by the classical vacuum (for example) because no one would ever think of such a weird idea in the first place (IMO). Also, I consider it rude to delete his work while he is unhappily banned from editing wikipedia physics articles and responding to criticisms. I'm just leaving the article as it is, and I'm happy that it has templates so that the reader may read the article and references and judge for themselves. --Steve (talk) 00:21, 25 March 2010 (UTC)


 * Hi Steve: My notion is that "free space" is an entirely classical concept, and distinct from the vacuum of quantum field theory or more modern theories. For example, things like nonlinearity have been calculated for these modern vacuums, and free space is defined to satisfy liner superposition, even near point charges where fields become infinite. Thus, the difference in our views may not be about "what is free space" but about "what is the role of free space". It may be that "free space" is a bit of an anachronism. However, I don't think NIST has quite caught up with the situation, which doesn't bother them because all the differences are beyond their ability to measure things. Brews ohare (talk) 17:00, 20 July 2010 (UTC)


 * Steve, as to "I consider it rude to delete his work while he is unhappily banned from editing wikipedia physics articles and responding to criticisms," I think that's backwards. It was completely pointless for editors to try to improve this article while Brews was participating.  Now is the time to decide what to do about it, clean it up, unbloat it, get rid of it, or whatever.  The topic ban and block are to allow things to get back to normal in physics articles, even if normal is flawed in the opinion of some.  Dicklyon (talk) 05:21, 3 November 2010 (UTC)

Outer space
No one in astrophysics refers to any part of outer space as "free space". I removed this bit of synthetic original research. ScienceApologist (talk) 15:10, 3 November 2010 (UTC)