Sometimes very active discussions about peripheral issues overwhelm a thread, so this is a permanent home for those conversations.
I’ve opened a new “Sandbox” thread as a post as the new “ignore commenter” plug-in only works on threads started as posts.
Jock,
If the SAM, an indirect method, doesn’t actually measure distance, then why do you accept that the AVO indirectly measures current?
I cannot figure out how that applies here. I would think any savvy ‘professional’ using your SAM would take latitude into account – the tidal effects that you mentioned might be a step too far — so your suggestion that they use the equation for the circumference of a circle is an invitation to error.
Of course, if you were trying to calculate the arc length on a fictional sphere, then using the circumference would be okay (but unnecessary), but you would be calculating something quite different from your YSM…
I am enjoying your ongoing impression of phoodoo, though. It’s epic. What happens to the accuracy of your voltmeter if it (the meter) gets warm? LOL.
keiths,
AVOs measure current somewhat more directly than they measure voltage, as you so helpfully pointed out. They are all indirect, and how indirect matters.
Good, that clarifies some things, Jock. I was using “finding” because I didn’t understand that the “two different things” you were referring to were “measurement” vs “calculation”, but now I do, I think.
Some tell me if this is accurate in your view. Both methods will arrive (I need some general term) at a number representing the distance from A to B, which is an arc length, but there is a qualitative difference in how they got that number: one is measuring and the other is calculating. They are examining the same property of the same object, the length of arc AB, but the methods they use are not the same. Finding a value of a property by measurement is different than calculating a value for that same property.
How’s this sound?
aleta,
Yes, that is pretty good, subject to one minor caveat. They are purporting to be ascertaining the same property, and we’ve cross-calibrated them so that under normal circumstances they seem to line up, but we could be quite deluded about that, and other factors could throw that calibration way out of whack.
Static-pitot tubes come to mind.
keiths:
Jock:
The SAM, which assumes sphericity, is distinct from the location-adjusted SAM, which does not. Remember, Neil and Flint have been emphasizing that the SAM produces different results depending on location, and they’ve tried to argue on that basis that the SAM and the YSM are incommensurable.
In addition, though you don’t realize it, the radius-based method you’ve described also fails to account for latitude. You are talking about a radius (singular) when in fact the radius will vary not only between arcs but within them as well. Consider my Johannesburg-to-Halifax route. No way is the radius constant for that entire distance.
So to summarize, C = 2πr is applicable to the (non-location-adjusted) SAM, and knowing the circumference is therefore just as good as knowing the radius.
What happens to the accuracy of the YSM if your stainless steel yardstick gets warm? What happens to the accuracy of your laser rangefinder if the air density decreases due to higher temperature?
There is this thing called “measurement error”, and it applies to all methods of measurement. You presumably agree that the YSM still measures distance despite being subject to an error due to thermal expansion, so why doesn’t the voltmeter measure voltage despite being subject to an error due to temperature variation?
Jock:
Excellent! This is major progress.
Nine hours ago you were denying that AVOs measure voltage:
Now you are conceding that they do measure voltage (and presumably resistance too). That’s absolutely correct.
Regarding indirectness, I can heartily agree that an AVO measures voltage and resistance more indirectly than it measures current.
OK. Having established that AVOs measure A, V, and O despite doing so indirectly, the next question is this: If the AVO’s indirectness doesn’t disqualify it from measuring those things, then why does the SAM’s indirectness disqualify it from measuring distance? What is the relevant difference between the SAM and the AVO in this regard?
Good. The cross-calibration of methods is important. I’m assuming rational people, not deluded ones, so I don’t think that is an issue. And of course either method has possible errors, and one method or the other may be more prone to errors, or just have someone make a bad error:
But in general, we have some support that what I wrote represents your view:
“Both methods will arrive (I need some general term) at a number representing the distance from A to B, which is an arc length, but there is a qualitative difference in how they got that number: one is measuring and the other is calculating. They are examining the same property of the same object, the length of arc AB, but the methods they use are not the same. Finding a value of a property by measurement is different than calculating a value for that same property.”
Jock, to aleta:
Calibration error is just one more form of measurement error, and it applies just as surely to the AVO as it does to the SAM. If calibration error disqualifies the SAM from measuring distance, as you seem to claim, then why doesn’t it disqualify the AVO from measuring A, V, and O?
As with AVOs and the SAM, pitot-static systems are subject to calibration error. That’s why the POH (Pilot’s Operating Handbook) for your aircraft will contain a table relating indicated airspeed (IAS) to calibrated airspeed (CAS). Calibration error doesn’t mean that the system isn’t measuring airspeed, just as calibration error doesn’t mean that AVOs aren’t measuring A, V, and O.
Incidentally, airspeed calibration error can be dramatically reduced via the use of a very long pitot tube. With a long tube, the opening is well away from the aircraft and thus less subject to pressure variations due to airflow over the wings and fuselage.
This is especially important on research aircraft. Check out the pitot tube on this baby:
You can read about that plane here:
Flight Testing NASA’s Prandtl-D Research Aircraft
I’d like to now move to direct vs indirect measurement, Jock, and see if you agree with my explanation of the following situation.
It is common to measure weight based how much the weight depresses springs that are under the weighing platform, which is then translated into a number that we read on a scale. The weight is not measured directly, but indirectly through the intermediary device. Most measurement (I think you have stated this), is of this indirect type. I’ve claimed that really length is the only property that can be directly measured*, and such properties as weight, temperature, time etc. can only be measured with a device that responds to the desired property** and then affects a scale that we can read. Is such indirect measurement still a measurement as opposed to a calculation?
Again, I’m interested if you agree with these points, Jock, or wish to make some clarifying or dissenting points.
Footnotes
* We could measure area directly by filling a space with squares or a volume with cubes, but we don’t: we measure sides and calculate. Therefore I don’t think you would say it was correct to say we “measured” the area of a room.
** For instance a spring depresses or stretches when a weight is put on it, a bimetal coil rotates a pointer when it is heated, the gears of a grandfather clock mark steady intervals of time, etc. In all of these the property under consideration causes an effect which can be correlated with a scale, from which we read the desired measurement.
As a generality, we measure, we assume, and we calculate. Almost all so-called ‘measurements’ are no such thing. This confuses some people, but it is a really important distinction, because those assumptions and even those equations may be wrong.
Your example is an excellent one. Every chemistry/biochemistry lab in the world has a number of “chemical balances” which work as you describe: the weight of the substance on the balance compresses a spring, this compression is measured, and a value is reported out.
Strangely, the “balance” reports out values that are denominated in grams which are units of mass, not weight. There’s a hidden assumption rocking here.
Of course, these instruments are called “chemical balances” because their predecessors were in fact balances (think Justitia) which do in fact measure mass against truth standards.
Bathroom ‘scales’ and similar ‘balances’ measure compression, which is assumed to be proportional to force (weight), which is assumed to have a specific relationship to mass (small g). They will misread badly on the moon, whereas an old-school Justitia ‘balance’ will do just fine.
The method matters.
Land on a strange planet, and you will be able to use celestial navigation to find your way back to your spaceship, without ever measuring a single distance.
DNA writes, ” Almost all so-called ‘measurements’ are no such thing. This confuses some people, but it is a really important distinction, because those assumptions and even those equations may be wrong.”
I really don’t think this confuses people who work with these quantities. Of course some assumptions and equations might be wrong, but for the most part our measuring/assuming/calculating procedures have been corroborated over time, and people know what levels of accuracy are suitable for what practical purposes.
I think I understand enough to bow out of this conversation now. Thanks.
How did we jump from what are we measuring to how accurately are we measuring?
For example: does arc become something different depending on how it is measured?
Petrushka asks, “For example: does arc become something different depending on how it is measured?”
No, it doesn’t
Then Jock went on to explain that by “same thing” he meant the method of determining the length of the arc were not the same things: he wrote.
“It’s right there in the verbs that you carefully used: Joe measures the length of his string, whereas Bill “calculates what he thinks is the length of the arc AB. … The second method measures an angle and calculates an arc length.
Joe believes the measured length of his string is the length of the arc, and Bill believes the result of his method is also the length of the arc. Joe and Bill don’t disagree about they are trying to do – determine the length of arc AB; they disagree about what can be properly called a measurement.
Jock:
It isn’t confusing at all. Just goofy, and obviously wrong. Mercury thermometers really do measure temperature here on planet Earth, where up is not down and left is not right, despite the fact that they infer temperature indirectly from the expansion and contraction of a quantity of mercury.
Save yourself some stress, Jock. You got yourself into a bind with a faulty argument about what the SAM measures because you hadn’t worked out the absurd implications of what you were arguing. Some of your fellow commenters pointed out the problems. What’s the big deal? You made a mistake. Anyone who’s spent time at TSZ has seen you do it before and knows that it isn’t uncommon.
Why not just acknowledge your error and move on? As I’ve said, I’m willing to bet that prior to this discussion, you wouldn’t have emitted the slightest peep of protest if someone had claimed that thermometers measure temperature. It’s common knowledge, and there’s nothing problematic, controversial, or dishonest about saying so. Nothing. To normal speakers of English, the word ‘measure’ describes what thermometers do with respect to temperature, and I’m virtually certain that it means the same thing to you. It’s just that you don’t want to admit that in the context of the present discussion.
Why not simply concede that yes, measuring instruments and methods do in fact measure what they’re intended to, though most of them do so indirectly? Aleta and I would agree, and I’d be surprised if anyone didn’t.
You’ve already conceded that AVO meters measure both current and voltage. If those are measurements despite being indirect, why not extend that to other instruments and methods, such as mercury thermometers or the SAM? What justifies your resistance? Is it principled, or are you just trying to avoid admitting your mistake?
People take all of that into account when designing instruments and methods of measurement. That’s why they cross-check their instruments and methods against others which measure the same thing. This is routine.
Time to hop in your spaceship and travel back to planet Earth, where thermometers measure temperature, airspeed indicators indicate airspeed, and the SAM measures arc lengths from point to point on the surface.
As far as I can tell, the problem here is semantic. If the disagreement is about exactly how direct or indirect a measurement is, whether either one is or is not a “measurement” is indeed goofy. “Measurement” is the linguistic term we use to indicate our intent.
The important questions have to do with the accuracy and precision of any measurement technique, and therefore the size and distribution of the error. As I understand it, the more indirect the technique and the more externalities it rests on, the wider the error bars. So we find it useful or necessary to specify these things, like “standard temperature and pressure”, or like “at earth surface gravity” or like “in the presence of a sufficient supply of reagent X (like oxygen)”. And of course we generally presume appropriate calibration to some useful scale.
So in this statement, the word “here” is ambiguous. There can be significant differences in readings depending on where “here” is on earth.
Flint writes, “As far as I can tell, the problem here is semantic. If the disagreement is about exactly how direct or indirect a measurement is, whether either one is or is not a “measurement” is indeed goofy. “Measurement” is the linguistic term we use to indicate our intent.”
I agree. Well said.
Flint writes, “The important questions have to do with the accuracy and precision of any measurement technique, and therefore the size and distribution of the error. As I understand it, the more indirect the technique and the more externalities it rests on, the wider the error bars.”
I agree strongly with the first sentence. The second sentence seems like it might be true, but my guess is that there are exceptions, and that in fact there are so many factors that you have to look at things on a case-by-case basis. Someone measuring the length of a city block with nothing but a foot ruler has a direct technique and no externalities, but the error involved in moving the ruler 400 times or so is huge.
I don’t think one can make strong generalities here.
I once spent some hours with the people who were surveying the boundary of my 80 acres of partially wooded land. They took many measurements of distance with a chain and angles with a transit, and used trig to calculate the straight line distances of the property (since they had to keep going around trees and couldn’t measure in a consistent straight line), and establish the corners. When they returned to the starting point they were only off by six inches. They had high quality equipment and knew what they were doing. Even though they used calculations and didn’t measure the boundaries directly they turned out to be very accurate. It was impressive.
Flint:
Since the disagreement is over what the word “measure” means, it is indeed a semantic dispute. That doesn’t make it insignificant or irrelevant, however. Semantic disagreements often matter.
If a kid swears to his mom that he didn’t knock over that expensive vase, but home security camera footage shows him doing exactly that, he isn’t free to claim that he was telling the truth since “knock over” means something different to him than it does to everyone else.
Likewise, Jock can’t reasonably claim that thermometers don’t measure temperature when they obviously do, according to the accepted usage of the word “measure” in the English language.
It isn’t the question that’s goofy. It’s Jock’s answer.
The relevance to the larger discussion is that Jock (along with Neil and you) claims that the SAM and the YSM don’t measure the same thing. If they don’t measure the same thing, they are incommensurable, and Neil wants to use that purported incommensurability to suggest that distance doesn’t “come from nature”.
The truth is that the SAM and the YSM do measure the same thing and are commensurable, so Neil’s argument doesn’t make it out of the starting gate.
keiths:
Flint:
I’m saying that here on earth, thermometers measure temperature. Apparently they stop measuring temperature at some point on the long journey to Planet Jock.
Besides the odd attempt to commandeer the word ‘measure’, another goofy aspect of Jock’s position is that he is all over the map on the question of whether indirectness disqualifies something from being classified as a measurement.
He has argued that because it is indirect, the SAM doesn’t actually measure distance, yet he has also said that “almost all science these days involves indirect measurement”. He has stated that an AVO does indirectly measure current, but not voltage and resistance, yet nine hours later he is telling us that yes, AVOs do indirectly measure voltage as well as current.
He is clearly making this up as he goes, and it’s leading to all sorts of silliness.
Of course. It was ever thus..
And instruments don’t measure things. People measure things. And please lets have more explanations how AVOs work.
🙄
Alan:
Haha. “Thermometers don’t measure temperature. People do!” earns you a spot on the next rocketship to Planet Jock.
One counterexample among hundreds:
Robot Adds New Twist to NIST Antenna Measurements and Calibrations
keiths,
Semantics. Always the sematics.
Does anyone ever seek your input in a conversation more than once, keiths?
Alan:
Says a guy who thinks “thermometers measure temperature” is a false statement. Semantics, indeed.
Lol.
Here’s Merriam-Webster:
Gotta love their serendipitous choice of a thermometer for their example.
That single definition puts to rest both Jock’s indirectness objection and Alan’s silly “instruments don’t measure things, people measure things” claim.
In case Jock is tempted to argue that the above applies to instruments that measure indirectly but not to the SAM, here’s Wikipedia:
The SAM clearly qualifies.
The SAM and the YSM both measure the arc length and are therefore commensurable. Neil’s argument fails.
Can we move on now? We need to get back to Flint’s worry that reality might not exist. 😄
I’m reminded of the dispute as to whether the victim was killed by the bullet (as the coroner argues), or by the gun (as liberals argue) or by the shooter (as conservatives and lawyers argue). Which, oh which, could it be?
Where did you find any worry? I’m not worried, but I can understand why you unconsciously projected that into what I wrote. For you, it would be a major worry. But I am content knowing that whether or not reality exists is something we can never know, and that it doesn’t matter. What matters is the utility (and continuing extension and refinement) of our models. Again, if it soothes your worries to believe our models describe an objective underlying but ultimately unknowable reality, I think that’s great. Go for it.
OK, I buy that. My thinking was, the more steps between the source (what’s being measured) and the target (the person interpreting the measurement), the more opportunity for error. But I think you’re right that if a direct measurement requires 400 steps, that’s lots of opportunity for error.
Alan:
keiths:
Flint:
Right. Same principle.
When we say that X causes Y, we aren’t usually saying that X entirely on its own causes Y. The bullet on its own wouldn’t have killed the victim if it hadn’t been loaded into a gun whose trigger was pulled by the shooter. Those were all necessary parts of the causal chain.
Similarly, the gun wouldn’t have killed the victim if it hadn’t been loaded or if the shooter hadn’t pulled the trigger, and the shooter wouldn’t have succeeding in killing the victim if the gun hadn’t been loaded or had malfunctioned for some other reason.
It’s perfectly reasonable to say that the death was caused by a bullet to the victim’s brain, or by the discharging of a firearm, or by the action of the shooter. Likewise, it’s perfectly reasonable to say that thermometers measure temperature and that people measure temperature using thermometers. Person and instrument are both part of the causal chain (and the person can sometimes be omitted, as my robotics example illustrates).
Things can get murky, though. The formation of the earth was as much a part of the causal chain as the shooter, the gun, and the bullet, but we would hesitate to say that the formation of the earth caused the death of the victim. The question of what can reasonably be substituted for X in the statement “X caused Y” is an interesting one.
Flint:
What “soothes” me is to have theories and hypotheses that actually make sense of the world. I find it soothing that the theory of evolution explains the presence of complex forms of life on earth — much more soothing than saying “that’s just the way it is”. The presence of complex life cries out for an explanation, and science provides one.
It’s no different with the consistency of measurements (or a thousand other phenomena). You can throw up your hands and say “that’s just the way it is”, or you can seek an explanation. The hypothesis of an external reality nicely explains why when I measure the length of this particular bolt, I consistently get answers around 1 3/4 inches, but when I measure a different bolt, I get a different result.
You have pointed to the non-sphericity of the earth as an explanation of why the SAM needed an upgrade. But as I commented earlier:
There’s also this:
Flint:
keiths:
I think this is wrong.
Digital instruments are pretty indirect.
Certainly more indirect than a mercury thermometer, or an analog voltmeter, or a measuring tape. I have a laser rangefinder that is more accurate and demonstrably more consistent than most people using a tape.
I would argue that the method is relatively unimportant, except at the bleeding edge of technology.
Seems like yesterday that we were arguing about the age of the earth. And stuff along those lines.
It’s all an illusion, man.
*exhales*
petrushka:
Don’t bogart that joint, man.
*takes deep toke*
The funny thing is, I actually thought about philosophical idealism while making my arguments, but I concluded that it didn’t make any difference. If matter doesn’t exist, only ideas, then reality consists of ideas, but it’s still reality, we still interact with it, and it’s still the reason why measurements are consistent.
Even if those ideas exist solely within my own mind, and reality is therefore an elaborate hallucination, it doesn’t mean that reality doesn’t exist. It just means that reality is the hallucination. I still interact with it, and it’s still the reason why measurements are consistent.
If reality doesn’t exist — neither in the form of matter, nor of ideas — then I can’t see why our measurements should produce consistent results, why experiments teach us new things and prompt us to improve our models, and so on. That’s why I think Flint’s position isn’t tenable.
There is something out there that corresponds to length, even if it’s ultimately different from how we perceive it.
Well, my notion of a causal chain is a sequence of events where each event led to the next. But the example I provided had no sequence stated or implied. It was an illustration of three different views of the same event. And in life, those different views aren’t theoretical or trivial, they are sources of major and important disagreements leading to conflicting policies.
Now, my model of “reality” as a useful philosophical construct and your model of “reality” as something actually out there independent of any philosophy, are (fortunately) not the stuff of truly deadly differences in legal policies. Indeed, I can’t find any practical difference between these models. I think similarly about whether a “number” is a symbol embedded in a philosophical structure, or whether it is something “actually out there”, provided we don’t make the kind of mistakes we’ve seen on this thread.
And I can’t see why a view of reality as a conceptual model must necessarily produce inconsistent results. My mental model of reality both requires consistent results, and rules out paradoxes. It also implies the possibility of perpetual refinement and extension – and the possibility of being wrong!
Flint:
As is mine. The causal chain in this case looks something like this:
person forms intention to shoot -> pulls trigger -> gun fires, propelling bullet -> bullet penetrates victim’s skull -> bullet damages brain tissue -> victim dies
It definitely implied a sequence, because we all understand the world (oops, there’s that pesky reality again) well enough to know how those causes fit together. For instance, we know that people don’t die first, retroactively causing bullets to be sucked into their brains, which bullets, upon being sucked, cause the gun to fire, which retroactively causes the person to pull the trigger, retroactively causing him to form the intention to shoot.
Yes, and those views can easily be reconciled when you understand that the events are all part of the causal chain.
Was the victim killed by the shooter? Yes, because given a loaded gun in his hand, aimed at the victim, the shooter’s pulling of the trigger set a causal chain in motion that ended in the death of the victim. Had he not pulled the trigger, this chain would not have been set in motion, and the victim would not have died.
Did the gun kill the victim? Yes, because the gun converted the trigger pull into a strike on the firing pin of the cartridge. Had this not happened — for example, if the gun had malfunctioned, or if a gun wasn’t available in the first place — then the causal chain would not have been completed, and the victim would not have died.
Did the bullet kill the victim? Yes, by causing fatal brain damage. Had there been no bullet — in other words, had the cartridge been a blank — then the causal chain would not have been completed, and the victim would not have died.
All three views are correct. They are not mutually exclusive.
The fact that the three views can be reconciled doesn’t mean that the disagreements and policy disputes go away. While everyone might agree that the causal chain (and others like it) should be broken, they could still disagree on which link to remove. A gun rights advocate might argue that we should try to prevent people from aiming guns at innocent victims and pulling the triggers, say by deterrence or through mental health interventions. A gun control advocate might argue that reducing the availability of guns is the right way to keep many of these causal chains from being completed. A third (odd and highly impractical) person might argue against deterrence, mental health interventions, and gun control, and focus instead on the bullets — say by requiring everyone to wear heavy duty helmets so that even if the rest of the causal chain is completed, the final step of bullet entering brain and causing fatal damage is averted.
Those three people could agree on the reconciliation of their causal views via an understanding of the causal chain, yet continue to disagree vehemently about the appropriate policies to implement in response.
Another approach could be to change the question being asked. Instead of asking what caused the victim’s death, we might bypass that altogether and simply ask what could have prevented it, evaluating potential policy changes on that basis.
That approach might make it less likely for the discussion to be derailed by Alanesque “guns don’t kill people, people kill people”-type arguments.
Flint:
There are quite a few problems with your ideas regarding reality, and I look forward to discussing them with you. It would be helpful if you could start us off by answering the questions I posed to you earlier. They’re quite relevant. I’ll repost them here for your convenience, and I’ll number them to facilitate discussion.
1. If reality doesn’t exist, what did I stub my toe on?
2. If reality doesn’t exist, why do repeated measurements of this wood scrap give answers that are all close to 14.9 inches, while repeated measurements of this other scrap give answers that are all close to 5.2 inches? Those numbers aren’t built into my model. Where do they come from, if not from reality?
3. If reality doesn’t exist, why do we have to update our models in light of new discoveries? Why doesn’t every observation match what is already predicted by our models?
4. When I run an experiment, what am I interacting with, if not reality?
5. If reality doesn’t exist, what exactly are we modeling? If there’s nothing to model, how can we speak of improving our models?
I have more questions to ask, but I’ll pause here for now.
Déjà-vu ?
Comments maybe worth a glance.
ETA (for completeness)
Thus I refute Johnson