Talk:Metrology

From Wikipedia, the free encyclopedia

This is a controversial topic, which may be under dispute.
Please read this talk page and discuss substantial changes here before making them.
Make sure you supply full citations when adding information to highly controversial articles.

This topic contains controversial issues, some of which have reached a consensus for approach and neutrality, and some of which may be disputed. Should you wish to make any substantial changes or additions;
  • Before making any such substantial changes, please carefully read the discussion-page dialogue.
  • During any such changes, please be careful to cite reputable sources supporting them, and when submitting your edit, please include an accurate and concise description in the "Edit summary" field-box.
  • After making any such changes, please also carefully describe the reason(s) for any such changes on the discussion-page.

(This message should only be placed on talk pages, please.)

Contents

[edit] Rewrite urgently needed

I don't know who wrote this article, but it is clearly not a metrologist! I will give it a shot in the near future, but will start with some comments:

Metrology IS the science of measurements. It may be divided into three levels: 1) Fundamental or scientific metrology, which concerns the establishment of systems of units of measurement, the definition of units, their realisation and the means of dissimination of traceability. Traceability further requires the quantification of uncertainy of measurements, almost a subfield in itself. 2) Applied or industrial metrology, which concerns the use of measurements and measuring instruments for manufacturing, trade, health care, environmental monitoring, etc. 3) Legal metrology, which concerns legislation on allowed units of measurement, requirements for measuring instruments and measurement methods used in society, and quantitative limits for e.g. alchohol in blood, toxic substances in air and water, radiation levels, etc.

The core concept in metrology is traceability, defined as "…the property of the result of a measurement or the value of a standard whereby it can be related to stated references, usually national or international standards, through an unbroken chain of comparisons, all having stated uncertainties." The level of traceability establishes the level of comparability of the measurement: whether a measurement needs to be compared to the previous one, or to a measurement performed anywhere in the whole world.

The Historical part is nonsense! Metrologists are neither historians nor anthropologists! Metrology is linked to science and trade. As soon as a culture needs or uses quantitative description of itself, its parts or its surroundings, measurements are performed, any measurement requires a unit, and metrology is established - it may not be on a strict scientific basis at first, but the basic concepts are there.

The historical development has many interesting aspects, and the basis is comparability. The Egyptian master builders responsible for the Pyramids had a strictly enforced metrology system. Their unit of length, the cubit, was traceable to the width of the palm of the Pharao. It was realised as a granite "Royal Cubit Master". Workers were required to take their working copies of the cubit to the Royal Architect for calibration and adjustment each full moon. Failure to do so was punishable by death! (See an account at http://www.ncsli.org/misc/cubit.cfm)

Metrology is closely linked to fundamental science. One of the first uses of new scientific discoveries is metrological, the laser, the Josephson effect, the quantised Hall effect, Bose-Einstein condensates, new science generates new ways of measurement, new methods of establishing units of measurement, and science itself requires better units of measurement to quantify the effects observed, to compare with other experiments or compare with theoretical results.

The basis of unit systems have changed over time, from units derived from the dimensions of the King to units derived from "invariants of nature" - early attempts such as the metre defined as a ten-millionth of the Earth quadrant, a definition which turned out to be too imprecise and the metre ended up as a platinum stick. Today, the efforts are put into establishing units based on assigning a definite numerical value to fundamental physical constants such as the speed of light and the permeability of vacuum. In a few years, the definitions for the kilogramme, the kelvin, the ampere and others will change to be based on Plancks constant, the Boltzman constant and the electron charge.

Metrology encompasses the science and technology employed to establish a definition such as "the meter is the length of the path traveled by light in vacuum during a time interval of 1/299,792,458 of a second", equivalent to stating that the speed of light IS 299,792,458 m/s, to realising this using frequency stabilised lasers, to laser interferometry establishing the length of gauge blocks, to the calibration of micrometers used by a car manufacturer and rulers used to measure the length of cloth you buy in the store.

Dalle 20:44, 14 March 2006 (UTC)


It's true, this current Metrology article is a dishonour for en:Wikipedia.
With your endorsement, later on, perhaps, I can try to improve.
Meantime, please have a look to SMH.
-- Paul Martin 22:44, 5 May 2006 (UTC)
then puleeeze, someone who's knowledgable enough in metrology, enough to recognize the deficiencies in this article, please correct technical mistakes, bad writing, whatever. i am an electrical engineer, pretty knowledgable in classical physics and applied mathematics, and could make use of a good reference in metrology, but i do not feel competent to fix this article. i have made some changes to articles like ampere and kilogram and wrote the current natural units article as well as a major rewrite of Planck units (which has since been rewritten even more), but i can't do this article. so Dalle or Paul Martin, please can you guys, at least incrementally, fix what you know to be bad in this? r b-j 02:06, 16 August 2006 (UTC)


I am not an expert on metrology, but I am a junior metrologist. I feel that it is more important to establish a useable framework with which to work with than debate definitions. I feel that definitions are critical, but in the absence of a true wiki, it is futile to gloss over the minutiae of metrology when one does not understand the fundamentals.

I will attempt to put together the major sections, and hope that any errors or omissions will be pointed out and corrected along the way. At this point, the wiki would be more useful with standardizing and regulatory information than Roman units of measurement, (though I find that interesting) and propose that the wiki community clarify any misunderstandings as I move along.

Let me clarify one more thing - The truly technical aspect of metrology involves statistics of uncertainties. I am not a statistician. I would suggest that if anyone is looking for that kind of information, that the NIST/engineering handbook link be referenced, as I will not attempt statistical analysis.

Pancho-Villa --A ship in harbor is safe -- but that is not what ships are built for. -John A. Shedd 18:23, 1 September 2006 (UTC)

[edit] Removal of tags

I've removed the {{disputed}} and {{cleanup}} tags, because the individual who placed them on the main article didn't provide any reasons. Ken talk|contribs 22:06, August 27, 2005 (UTC)

[edit] reasons for disputing

"Metrology is the science and process of ensuring that a measurement meets specified degrees of both accuracy and precision."

  • Metrology is the study of measurements. Issues of accuracy and precision are a small subset of that study. This page doesn't touch on anything else and that is misleading.

"Accuracy is the "correctness" of a measurement; that is, how closely it matches the value being measured."

  • I disagree with that. The Accuracy of a measurement is determined by a statement of all the factors contributing error to a measurement and any correction factors applied.

Fuego007 21:13, 14 March 2006 (UTC) The accuracy of a measurement is the degree to which it deviates from the standard. If a reference of 1 Volt is applied to a voltmeter and that voltmeter indicates 1.05 V as an average of 10 readings, that voltmeter is said to be accurate to within 0.05 V. Precision is the degree to which the measurement can be repeated. If the lowest of the 10 readings is 1.03 V, and the highest is 1.07 V, the precision of the measurement is said to be +/- 0.02 V (or, in this case, +/- 2%). Since we know how far the meter deviates from the standard, by adding the uncertainty of the value of the standard (in this case a value of 1 Volt +/- (for the sake of argument) 0.000001 V) to the measurement error (0.05 V), we can determine that the instrument can measure 1 Volt with an accuracy of -0.049999 to -0.050001 V. If the resolution of the instrument is insufficient, the accuracy can be stated as simply -0.05 V. Since we took a series of readings, we were able to determine the degree to which our measurements were repeatable, which gives us the value of the precision of the measurement. In this scenario, all of our measurements deviate from the standard value in the same direction. The more accurate a measuring device is, the less likely that is to occur. Since, in this caes, we have a measured deviation of 0.05 V, we can use this value as a correction factor and apply the value of the precision of these measurements (+/- 0.02 V) to make an accurate measurement of an unknown Voltage source that is expected to be in the range of 1 Volt. Thus, if an unknown value measures 1.05 V on this meter we can state that the unknown is 1.0 +/- 0.02 Volts. The application of these measurementsare intended to take into account as many factors as possible, including sytemic, environmental, and human. I hope this explanation of the relationship between Accuracy and Precision has been of help in clarifying the matter. It comes from many years experience as a USAF Metrologist. Accuracy and Precision are at the heart of Metrology. I accept that Metrology is the science and study of weights and measures. If you remove accuracy and precision from the practice of metrology, you will have the same effect as if you removed the tires from your car. Fuego007 21:13, 14 March 2006 (UTC)

Dalle 16:29, 15 March 2006 (UTC) Fuego007, I'm sorry, you are mixing some of the basic things up. Accuracy and precision are qualitative terms, not quantitative. What you are describing, your 0.05 V is the "error of indication", the (partial) result of a calibration. The observed dispersion (e.g. in terms of the experimental standard deviation of readings) is one quantitative measure of the "precision" obtainable with the instrument described. Another measure that needs to be taken into account is the display resolution, in your example 0.01 V (at least that is what is implied in your statement). The 0.05 V is not a correction factor (which implies multiplication), the correction (the quantity you add to the reading to obtain the true value) would be -0.05 V. Sloppy language is not good in calibration work OR writing encyclopedias. Your calculation of the accuracy range does not make sense when you take the display resolution into account (adds another 0.005 V at each end).

Note that instruments such as voltmeters and other indicating instruments simply provide a reading - you apply a known voltage, and some number appear in the display. The objective of a calibration is to relate whatever reading they show, to the value of the physical quantity applied to the input. But, this whole discussion should appear under Calibration. Dalle 16:29, 15 March 2006 (UTC)

"Precision, on the other hand, is how detailed a measurement is."

  • I disagree. The precision of a measure is a statement of the range of error.
    • Not true, precision is defined (VIM) as "closeness of agreement between quantity values obtained by replicate measurements of a quantity, under specified conditions" - hence it can be described in terms of repeatability, reproducibility or similar. Dalle 16:29, 15 March 2006 (UTC)
Perhaps this subject is slumbering but here goes anyway. Dalle's definition of precision is fine up to a point. My reservations concern the lack of reference to significant digits. For example if an item is measured in millimetres but does not produce absolutely consistent results as observed by all observers, at any time, under standard conditions, then perhaps we are exceeding the accuracy of the measuring unit and should remeasure the item using centimetres, which will produce completely consistent results. Geologician 09:01, 4 July 2006 (UTC)

In order to repair this page those errors should be corrected and the discusssion of metrology opened up to include everything that is measured weighed and judged including the historical precedents.

Regarding accuracy vs precision: That good old standby the OED would have us believe they're synonymous. Now, in technical fields, I've always seen them used in exactly the way this article states them, but there you have it.
Regarding differing definitions of precision: Your definition and the one in the article are equivalent. Saying that my digital caliper offers 4 digits of precision over the range 0 to 6 inches is equivalent to saying that it's range of error is +/- 0.0005 inches. Although I like your wording better.
Please don't be argumentative just for the sake of arguing, which is how you're currently coming across. Realize that on Wikipedia, just as in email, the mental image the reader forms is the person.
Ken talk|contribs 03:20, August 29, 2005 (UTC)
Metrology is more than just the calibration of standards. Accuracy and precision are two different things and neither of them is equivalent to metrology. Accuracy has more to do with noting things which introduce error and correcting for them. For example the tension on a measuring tape, the distance it spans unsupported, the coefficient of expansion of the material the tape is made of, in some cases wind velocity, tidal influences and the rotation of the earth are factored into an analysis of the accuracy of a measurement. Precision has more to do with identifying the range of accuracy in a calculation before and after correction. Metrology is the study of measures and includes all their historic changes and the reasons for them. In many cases the standards of measure are changed but not the system, in other cases the system is changed but many of the standards are preserved.Rktect 12:55, 23 September 2005 (UTC)

[edit] Uncertainty Analysis and Certification

Friends, I appreciate all of the inputs on this topic I have read thus far. However, this article still does not effectively define "metrology" as I know it. One way I think we can make this a stronger and more efficient definition is to cite sources for some of our inputs. For example, while the previous discussion this group had on "accuracy" and "precision" were close to the mark, the competing arguements lacked the integrety of peer review. Accuracy, as defined by NIST in its Handdbook NIST/Sematech handbook for Engineering Statistics is as follows: Accuracy: “…a qualitative term referring to the closeness of agreement between a measurement made on an object and its true (target or reference) value.” (NIST/SEMATEC, undated. Precision, as defined by the National Conference of Standards Laboratories is simply: Precision: "The resolution to which the value of a quantity can be specified." (NCSL, 1994)"

However, neither of these terms really get to the roots of metrology. By the way, I have taken the liberty of ammendmening the article to reflect the internationally accepted definition of metrology which is simply: The science of measurement. Note, "measurement" and "measures" are two different concepts. In truth, the goal of metrology is to minimize the "uncertainty" of measurements by using measures (standards) of known quality and applying statistical annalysis to compensate for sources of possible "errors." Consequently, we metrologists have coined the phrase of "uncertainty analysis" as being the primary tool for performing reliable metrology.

One thing I suggest we all remember is that metrology is not a derivative of "science". Metrology is a precurser to science. If anything, metrology allowed the sciences to grow and gain acceptability. Metrology has always been a tool of commerce. It was started well over 4,000 years ago in ancient Summaria. In fact, a well know phrase amoung metrologists is: "Mathmatics is the language of science and metrology is the language of commerce." Statistics and Probability theory are the rules or syntacs of metrology.

I urge you all to review the literature below to help get some ideas of how we can make this article a little more reliable and give it some strength. Thanks for your inputs to this discussion which I am sure will follow soon

[edit] Suggested Reading

NIST/Sematech Engineering Statistics Handbook

Bureau International des Poids et Mesures. (2000), The International System of Units (SI)

Bureau International des Poids et Mesures. (2000), The Convention of the Meter

Melville, D.J. (2001). Sumerian metrological numeration systems, Mesopotamian Mathematics

Office of Weights and Measures

--David.c.h 10:03, 11 October 2005 (UTC)

[edit] Major changes and tags added

Friends: OK, here is what I did today:

First paragraph, the debate over the definition of "metrology" hopefully was put to rest by citing international authorities.

Second paragraph was tagged for clean up. While the discourse on the Time and Space continuum was interesting, I believe it to be irrelevant to the subject this early in the article. Perhaps it would fit in a subsection titled "History or metrology." However, I have some serious disputes with the author’s fundamental assumptions. Hopefully we can work it out amiably.

I have a serious dispute with the author of the third paragraph. The concept of standards and measurement processes is far from a “small subset” of metrology. Standards and measurement processes is metrology. The proper use of standards and interpreting measurement results is the backbone of metrology theory. I believe this paragraph should be deleted in its entirety. However, I am more than willing to discuss the issue with the author to see if we can’t find an acceptable compromise of some sort. If not, we will probably require some form of mediation by the administrator.

Still in the third paragraph, it is true that the accuracy ratio for the standard to the unit under test has traditionally been 10:1. With this wide a ratio it dramatically speeds certification of the test with minimal work. Unfortunately, with today’s modern technology, maintaining a 10:1 ratio is becoming increasingly hard to accomplish. NIST, NCSL, BIPM are now recommending a four to one (4:1) ratio as desirable. However, even with that, 2:1 ratios and a lot of uncertainty analysis is becoming more commonplace. Manufacturers are now delivering production equipment almost as accurate as some of our national standards. In some cases, companies like HP are now creating and maintaining the national standards for physical attributes NIST can not afford to develop.

I next added some more recommended reading on the subject of metrology. I hope you will all find them appropriate. In addition, I added a couple of new external links for your review.

Finally, I reinstated this article to the category of “Science.” As shown by my modifications to the first paragraph, the rest of the world considers metrology as a separate scientific study. This recognition is also formalized in many universities around the world that offer science degrees in metrology.

I look forward to further discussion with all of you on this subject.--David.c.h 08:47, 12 October 2005 (UTC)

[edit] Metrology rates category status

I believe that there are enough links to this article that it has earned subcategory status. I believe that it should be a subcategory of Science because it is a seperatly recognized scientific field of study. I also believe it should be a subcategory of Manufacturing as metrology is intregal to commerce. Metrology in the USA is controlled through the Department of Commerce though the NIST. I will leave this article as is but utilize the first paragraph to start the Metrology subcategory. I may have to ultimately remove some of the categories for this article though to prevent looping. Your inputs on how to best effect this change are appreciated.--David.c.h 00:17, 13 October 2005 (UTC)

[edit] Reason for changes

First, I must apologize. I accidently submitted these changes the first time when I was logged out. I didn't know Wikipedia had a time out feature. The input directly before this one was mine and is the same.

[edit] My reasons for change

This article is in serious need of help. An awful lot of uncoordinated, irrelevant, and unsubstantiated information is contained here. I would suggest that all contributors make better use of citations for their inputs if they expect universal buyoff. I would like to refer everyone to Wikipedia:Cite sources. I am sure following some of the simple guidelines in it will dramatically aid in developing this article further. Here are some of my changes for today.

  1. I added major section for Introductory information
  2. I added subsection for “historical development.” See Wikipedia:Guide to layout
  3. I added a subsection for the “Mechanisms of metrology.”
  4. I addes a subsection for “References” and moved cited material from “Recommended reading into it. See Wikipedia:Cite sources
  5. I renemed the “Recommended reading” section to “Further reading” to put in line with Wiki guidelines in Wikipedia:Cite sources
  6. I added a few tags.

Wikify tag: I believe following the guideline of Wikipedia:Guide to layout will make this an easier to read product and add strength to it.

Section Fact dispute tag: in the new “historical development” section. I contest the author’s assumption that the space time continuum has anything to do with metrology. Metrology can be accurately dated back to 4,000 years BC to ancient Mesopotamia. The earliest use of units of measurement can be found on clay tablets from ancient Uruk. I refer our contributors to the following articles:

These tablets represent the oldest know writings of mankind. Uruk is believed to be the birthplace of literacy and mathematics. Most of the tablets were, in actuality, ancient bills of lading. They were probably used between cities when shipping goods to help ensure there was no pilferage along the way. These tablets consisted of symbols representing weights and measures of products. They were tools of commerce, not science. They were the precursors to written languages and mathematics.

Dispute about tag: The author contends that “standards” and “measurement” are only a “small subset” of the field of “measures.” I strongly disagree. Metrology is about “measurement science,” not measures. The very act of making a measurement is to compare one thing against another. However those two things disagree with each other requires one to be a base (standard) unit and the other to be a differential (test) unit. For example: stick "A" is longer than stick "B." This statement makes stick "B" the standard of measure and stick "A" the differential (test) unit. If we were to say; on the other hand, that stick "B" is shorter than stick "A", than their roles would have to be reversed. "A" becomes the standard and "B" becomes the test. No “measure” can be absent a standard.

Globalize tag: Finally, many points in this article fall outside the realm of world wide agreement on the term “Metrology” I have cited my authorities in paragraph one. I urge all other contributors to cite their authorities likewise.

I welcome discussion on this topic.--David.c.h 03:10, 16 October 2005 (UTC)

[edit] Changes

I came to this article while doing wikification. I wikified the article and removed tag. I also moved introductory material to the introduction section, which was empty. I moved the globalize tag to the top of the article because the complaint on talk (see above) doesn't give a specific section as being the problem. Finally, I removed the section dispute tag. The person who placed the tag gave no reason for doing so in the edit summary or on the talk page and it is not obvious what the reason for the tag is. Also, the location of the tag made it ambiguous as to what section the person was referring to. The other dispute tag remains on the article. This concludes my work on the article, as I do not know much about the technical aspects of metrology. Good luck, Kjkolb 12:51, 5 April 2006 (UTC)

Much of the mess is left over from someone who was pushing his personal theory of the interconnectedness of all non-Metric systems of measurement. --Carnildo 02:34, 6 April 2006 (UTC)


The historical development section appears to be a demonization of metrology, and it's "controversial" nature. Further, I find it lacking in that the section develops a view of standardization being political and not scientific, when in fact metrology has an agenda of standardization. While there has been some disagreements, I think that the historical dev section should at minimum cover the groundwork to founding of NIST, SI, and other standards agencies.

Also, I really dislike that De Sanctis quote, is it really necessary to blatantly demonize metrology? Not NPOV in my opinion.

Agree, Dalle 20:46, 4 September 2006 (UTC)

Also, metrology has not been "historically" discrete units of anything, but rather comparisons to an arbitrary standard. Modern science has given us intrinsic and discrete qualifications of our discrete units. I really must put forth the Egyptian King's cubit example here. The king's arm does not qualify as a discrete unit, does it? What if he's gained weight? What if he's laying down? Standing up? Stretching? What's the consensus here?

--A ship in harbor is safe -- but that is not what ships are built for. Pancho-Villa -John A. Shedd 21:58, 1 September 2006 (UTC)

The units was often realised by a material artefact kept by some proper authority. A new King could then easily decide to change the unit. Dalle 20:46, 4 September 2006 (UTC)


One more thing - Can we all agree that the functional authorities on metrology are NIST, SI, and equivalent EU agencies? I don't have to 'prove' that, do I? I am using the U.S. agencies simply because this is en.wikipedia, and not fr.wiki, es.wiki, or whatever. I obviously share the English wikipedia with my British counter-parts, so you'll forgive me if I forget to mention the U.K. equivalents, but feel free to add those too. Metrology is international in nature, so I feel I'm saying the truth when I say it doesn't really matter which standards agency one is talking about, as they all are "standardized," so to speak (no pun intended)

(Well, maybe a little)

--Pancho-Villa "A ship in harbor is safe -- but that is not what ships are built for." -John A. Shedd 22:10, 1 September 2006 (UTC)

For the SI the "top" is the Metre Convention, the member state metrology institutes, e.g. NIST. SI units voted upon at the CGPM from proposals put forth by the CIPM, presented by the consultative committees, developed by the science staff of the metrology institutes. Dalle 20:46, 4 September 2006 (UTC)


Wow, this is thick stuff. I am taking a short break, as it is a long weekend. I am open to commentary on the edited sections. The historical dev section was the most inaccurate, so I fixed most of that first, and changed the intro a bit. I dislike referring to early metrology as proto-metrology, but it most definitely was not the same as modern metrology. Modern metrology is predicated on scientific definitions and laws of physics, which Sumerians, Egyptians, and Hebrews most certainly did not have.

Later, I plan to add a section that elaborates on the definitions of the SI units, and some physical constants that are used for standardization. Also, elaboration on the differences between primary standards, secondary standards, reference standards, and transfer standards will be needed. *_* No rest for the wicked you know!

Pancho-Villa "A ship in harbor is safe -- but that is not what ships are built for." -John A. Shedd 02:18, 2 September 2006 (UTC)

I suggest to keept the metrology article mostly on the concepts and mechanisms of metrology, and elaborate on the SI in the SI article or the units of measurement article. Dalle 20:46, 4 September 2006 (UTC)

Here are my sources. I will figure out how to footnote them shortly: http://www.bipm.org/en/convention/wmd/2004/ http://physics.nist.gov/cuu/Units/background.html <--(Note that checking the BIPM site references you to NIST)

[edit] Changes

Good that the revision of the metrology article has begun. I have taken the liberty of expanding the definition a bit, introducing the three distinct sub-fields. I have also corrected some inaccuracies concerning the organisation of SI. And added a few things here and there.

May I suggest that contributors download and read the vocabulary documents referenced and add text in compliance with the definition of terms in there. Dalle 20:46, 4 September 2006 (UTC)

[edit] Reconstruction of article

I hope that this is the beginning of a real collaboration on this.

Good efforts on rewriting Dalle and Pancho-Villa, the article is now in a much better state, I wanted to do something like this a while back but was wary of jumping in (and also never got round to it). To make the article a bit more international I would suggest that the references to NIST and ANSI be taken out as well as all the references to NMIs in the external links and something like a List of National Measurement Institutions be created where all this can reside. In the meantime I will try to do what I can to improve the article and other related ones. JMiall 22:40, 4 September 2006 (UTC)
There is presently some overlap between the metrology article and the articles on units of measurement, SI, the information in the articles on the meter and kilogram. I suggest to move the information specific to SI to those articles and keep the metrology article more clean. However, on the historical side, it is useful to get the point across concerning the introduction of the scientific approach to metrology (rational system of units).Dalle 19:30, 5 September 2006 (UTC)
Where are we on the tags attached to this article and the talk page? Are we still controversial? Dalle 19:31, 5 September 2006 (UTC)
I don't think that the article is currently at all controversial. As far as I was concerned the main controversial bit was the 'one small subset' section which isn't there any more. JMiall 21:43, 5 September 2006 (UTC)

Chugging along - Detailed the application of metrology to industry, as it is really about validation of measurements. A co-worker asked where she could go to get a degree in metrology, and I/we were in the slightly difficult position of explaining that there wasn't really a metrology degree, per se. Anyways, metrology is about ensuring quality in measurement systems, so I felt it necessary that a metrologist is really just kind of an equipment QA/validation technician. To be sure, metrologists have specialized knowledge, but in practice, it's a bit hard to clarify where my instrumentation and electronics training stops, and metrology begins.

I thought people might like to see some clarification on ISO standards, so I stuck those in there, along with others. This could probably become a separate all-inclusive list, if we cared to get thorough about it (though I don't).

The definition of a unit - well, we define it in terms of something right? I agree that SI details should be left to CIPM, but I feel the metrology article would be well served by explaining to visitors that the units we use in metrology are fixed, constant, and precise. Perhaps some elaboration on defining units?

Pancho-Villa "A ship in harbor is safe -- but that is not what ships are built for." -John A. Shedd 14:57, 6 September 2006 (UTC)


Pancho, you may have misunderstood my remark, I suggest to describe the details you mentioned, not the the CIPM, but to the article on SI... Concerning units: it is "easy" to define a unit, take for example the present definition of the ampere, but this definition is useless until you can realise it in an actual experiment, and assign a value to some current generator. This will be an experimental procedure, prone to experimental error which will set a limit to the ability to measure current in all subsequent experiments. So your realised unit is not fixed, constant, but precise to the level of the evaluation of the uncertainty of you realisation. But the present definition leaves it up to you to do better, in fact, what the present definition says is simply that the permeability of vacuum is exactly 4 π 1e-7 H/m. So you can use Maxwells laws on electromagnetism and invent a clever setup that generates current calculable from other measurement quantities and permeability. That's how scientific metrology works.

I believe that an industrial metrologist has a lot more work than the QA stuff you describe. There are constantly comng new instruments to market, and the metrologist must devise calibration procedures, etc. technical aspect, if not scientific, to ensure the proper use of measuring instruments. A QA person may wave some standard in front of a lab guy, but it is the tech/metrologist who must figure how to implement, technically, requirements for traceability, estimate uncertainty, conformity, fit-for-purpose, etc. Take a look at what the NCSLI is doing - they're industrial metrologists.

Dalle 16:02, 7 September 2006 (UTC)

Dalle: Yes, yes, and yes, you are pretty much right.

I'm no industrial metrologist, so I know that what you are telling me must probably be the case. In any case, it really varies. My last cal job had hundreds of pieces of equipment due any given month. Not so in my current post.

I know see what you mean by way of unnits, but I wonder how best to put that forth in the article. I know that some of this is more electronic POV than industrial POV, so I task an industrial metrologist to NPOV this from a metrology POV. Obviously, the different fields of metrology, calibration, and instrumentation all share in the metrology burden, so I will not attempt too much more on this article, as I'm not qualified to. It's scope is much bigger than my background. Thanks all, and gl, I'll still check back!

Pancho-Villa "A ship in harbor is safe -- but that is not what ships are built for." -John A. Shedd 05:32, 8 September 2006 (UTC)

[edit] Missing section of Metrology - Human Metrology

I do not wish to add to the debate over the 'precise' or 'accurate' definition of metrology, it is sufficient enough for me that Metrology is the Science of Measurement.

I would submit that historically metrology started with the first numbering system put to general use, counting was the first true measurement, and its reference standard is zero.

I would like to see added the field of Human Metrology, the collection and analysis of subjective human response to various stimuli. Although these types of measurements do not always resolve to standard, calibrated units, or reference standards, they are measurements none the less. There are many standard procedures, methodologies, and statistical models for analyzing human response data (Paired comparison, godel squares, rank-order, etc.). I am a metrologist (certified by the U.S. Dept Commerce), and deal with both objective and subjective types of measurements, often trying to correlate the two.

Human Metrology does not easily fit in the afore mentioned 3 categories of metrology due to highly subjective nature, lack of calibrated standards, etc. but in fact many standards and methods are based on Human Metrology experiments and observations (CIE, Acoustics, etc.). After all, if metrology is for commerce, and commerce is for humans, you may be able to measure and reproduce a particular property very accurately, but it won't matter if no one 'likes' it. To that end I would dispute the statement 'no measure can be absent a standard', in the context of subjective measurements, if standard [unit] or [reference] standard was the intended meaning of standard.

I would like to see more detail around this field, particularly common methods, and tests, and units of measure that are defined by human metrology studies, etc. Perhaps starting with the methods for collecting and analyzing the original 1931 CIE Observer data, etc.

The planned references to units of measure, would be a great addition. I think it would aid in navigation if there was a listing of links to specific sub fields of metrology (radiometry, spectrophotometry, etc.) and then the typical units of measure, etc.

I am sorry that I do not have several sources to back my arguments, they are simply based on logical deductions. I was actually looking to Wikipedia for more info on Human Metrology, because there is an incredible lack of it.

15.251.169.69 00:57, 15 December 2006 (UTC)chayden@hp.com

Retrieved from "http://en.wikipedia.org../../../m/e/t/Talk%7EMetrology_2463.html"