and to how many decimal places?

[fumei]

Just read some text that mentioned a "precise estimation".

Hmmmmm.

I can understand an estimation being as precise as possible, but a precise estimation?

However, I found one definition to be - "conforming strictly to rule or proper form". So if a protocol can only return an estimation, if one follows, strictly, the forms and rules of the procedure, can it really be called precise?

Gerry
My paintings and sculpture

 

[Gooser]

Damnit, where's the unsubmit button? I meant to uncheck 'Signature'.

 

[CajunCenturion]

==> As in, the degree of refinement with which a measurement stated.
That directly correlates to the precision of the equipment being used to take the measurement. When counting steps, I can't be precise to a quarter of an inch, because that process does not afford that degree of precision. When using the range finder, I can be considerably more precise.

I think the term accurate is being mis-applied in your example. I think in that example you're talking about acceptable tolerance. If your specifications are too exact to be met, then you have no choice but to relax your specifications, and when you relax your specs (reduce precision), it's not your accuracy that's increasing (in fact, the accurate length is still the same), it's the acceptable tolerance range that's increasing. I do completely agree that if you reduce your precision so that the acceptable tolerance range falls outside of operating requirements, then you need to increase your precision, or tighen the specs.

--------------
Good Luck
To get the most from your Tek-Tips experience, please read FAQ181-2886
As a circle of light increases so does the circumference of darkness around it. - Albert Einstein

 

[Gooser]

Centurion: I guess in the case of accuracy I was looking before the OR:

...to a standard...

As in, the degree of conformity of a measure to a standard; the standard being the degree of precision.

I'm not saying you're wrong, just trying to illustrate my point-of-view.


v/r

Gooser

Why do today
that which may not need to be done tomorrow --me

 

[fumei]

CC - then, as the estimate is the result would it not be correct to say there is a difference is accuracy?

The result (the stated altitude - "this spot is 16,200 feet above sea level...oh, darn there is that Tibetan soldier firing at me with that matchlock again. You blokes with the Maxim...could you do me a favour? Ah, thank you...) is derived, I assume with some calculation based on the input of the barometer reading.

Regardless of the input, the process, the calculation will be the same - whether it is windy, or not.

You read X inches on the barometer, you get a derived Y in altitude.

If X is different on DayB, because there is no wind, the SAME calculation is used.

DayA - 16,200 feet - windy

DayB - 16,400 feet - no wind

The ability of the instrument to be precise changes, but it is the accuracy of the result (the estimate) that also changes.

So "precise estimate" should - gosh...I think...actually be "accurate estimate".

An aside: the ratio of killed seemed to average about 200:1. Matchlocks versus Maxim guns, repeating rifles, and 10 pound field artillery......hmmmmm.

The book has fantastic photographs, including the first ever photographs from inside Lhasa.

Gerry
My paintings and sculpture

 

[Gooser]

Does the wind affect the instrument, as blowing across the top of a pot of boiling water? --I wonder.

Is this just a pot of boiling water and a thermometer? Or is there more to this "boiling water barometer"?

A water version of a run-of-the-mill mercury barometer would have to be about 3.5 stories tall, so it seems that would be impractical for a group travelling on foot in the mountains. Though they may have had some sort of rubber hose with a glass tube on the upper end--pure speculation. If that were, however the case, wind blowing and shaking the hose could cause fluctuations in the water-level reading in the glass tube at the top. These fluctuations could, in fact cause a degradation of precision due to the observer having to use a range, rather than a value for the input into their algorithm to determine altitude from pressure. Of course, this is all complete speculation, but would you guys agree that this would cause the estimates to be less precise ?

Some of this conjecture is based on what I read here


v/r

Gooser

Why do today
that which may not need to be done tomorrow --me

 

[CajunCenturion]

Yes, that makes sense to me.


--------------
Good Luck
To get the most from your Tek-Tips experience, please read FAQ181-2886
As a circle of light increases so does the circumference of darkness around it. - Albert Einstein

 

[CajunCenturion]

==> but would you guys agree that this would cause the estimates to be less precise ?
No, I would not agree. It would likely cause the estimates to be less accurate. Estimates are values, and values are measured in terms of accuracy.

--------------
Good Luck
To get the most from your Tek-Tips experience, please read FAQ181-2886
As a circle of light increases so does the circumference of darkness around it. - Albert Einstein

 

[fumei]

Of course, this is all complete speculation, but would you guys agree that this would cause the estimates to be less precise ?

According to CC definitions, I would not agree.

The estimates are the result. They are less accurate. The measurements are less precise. Again, if the calculations used under different conditions (the actual measurements) are identical, then the calculated result will have the same precision - no matter what the measurements. Its accuracy will change.

Gerry
My paintings and sculpture

 

[Gooser]

Not necessarily...

Assume they normally come out with a reading of altitude within 30'.

If the fluctuations caused them to come out with a range, where they could only come out with a value somewhere between 15300 and 15750, then they could drop the precision of their estimate to 500' increments, call the estimated altitude 15500' and call it a day. This decreases the precision of their estimate, and more than likely, the accuracy as well.



v/r

Gooser

Why do today
that which may not need to be done tomorrow --me

 

[CajunCenturion]

The boiling water baromoter is essentially just that - a thermometer used to measure the temperature at which water boils.

It's proper name is: hypsometer

--------------
Good Luck
To get the most from your Tek-Tips experience, please read FAQ181-2886
As a circle of light increases so does the circumference of darkness around it. - Albert Einstein

 

[fumei]

You are tossing something we know nothing about into the mix. You are talking about a person making an explicit DECISION to alter how they achieve their calculated result.

Gerry
My paintings and sculpture

 

[CajunCenturion]

==> This decreases the precision of their estimate, and more than likely, the accuracy as well.
No.
It decreases the precision of their process.
It decreases the accuracy of their estimates.

--------------
Good Luck
To get the most from your Tek-Tips experience, please read FAQ181-2886
As a circle of light increases so does the circumference of darkness around it. - Albert Einstein

 

[SilentAiche]

If you are shooting a gun...

[blue] You probably dislike guns. I mean, Really dislike guns...[/blue]

Tim

[blue]_______________________________________________________
"Although many figures are strange, prime numbers are truly odd."[/blue]

 

[Lightning]

[stirring pot]

Estimate
Pi = 3.141

Precise Estimate
Pi = 3.14159265358979

In common usage (for example, calculating the volume of a cylinder) both are quite accurate, yet both are only estimations.

In some scientific usages, however, neither may be accurate enough for the purpose required.

So, accuracy is not a technical term, but a subjective one.

??
[/stirring]

 

[Scott24x7]

And how about the constant e?

e to 1 digit is 2
e to 2 digits is 2.5
e to 3 digits is 2.66

All are accurate, none are estimates, and none really reflect the "significant" value of e which, if we look at it to 10 places becomes: 2.71828182845

So, to the point of accuracy being subjective, I think we only need to look toward constants, like e and pi.

Pi is so weird in fact, that the further out you get in calculating the value the forumal changes. (I'll leave that the digit heads to explain of someone wants to go down that path...)




Best Regards,
Scott

"Everything should be made as simple as possible, and no simpler."

 

[CajunCenturion]

[Responding to stirring]
Personally, I wouldn't use 'estimate' when referring to pi, e, or any transcendental number for that matter. I would use the term 'approximation', but that's a different discussion.

If you are using pi as part of a calculation, such as calculating a volume, then you should refer to the precision of pi, because pi is part of the process. You'd talk about the accuracy of the volume, because volume is the end-result.

Accuracy and precision are both technical terms and there is nothing subjective about when you should use precision, and when you should use accuracy.

There may be, however, considerable subjectivity in what constitutes a sufficiently accurate result, which in turn, determines the required degree of precision in the necessary approximations and calculations.

Lightning did mention 'common usage'. All of the previous discussion not withstanding, common usage will be -- right or wrong -- whatever the common person decides it to be.

--------------
Good Luck
To get the most from your Tek-Tips experience, please read FAQ181-2886
As a circle of light increases so does the circumference of darkness around it. - Albert Einstein

 

[Scott24x7]

CC,
Doesn't that make your whole argument against subjectivity irrelivant then??? I understand your point about precision and accuracy having very specificially defineable utilization, but as you pointed out, they are only relevant to the degree that the common person needs them...
Do I really care that a gallon of milk actualy has .99997 gallons, or 1.0000034 gallons? Not really. I'm not likely to build, as a commoner, a rocket that uses milk as fule, and if I could, I'm sure I'd be able to specifically measure my burn rates, and trajectories for getting it to its target. (But hey, this is just fun, right?)


Best Regards,
Scott

"Everything should be made as simple as possible, and no simpler."

 

[CajunCenturion]

I'm not arguing against subjectivity.

My position is, as you've stated, that precision and accuracy have specifically definable utilizations. It's the misuse of those terms that is behind my position. Precision is used in reference to the process, and accuracy is used in reference to the result.

Subjectivity comes into play when you talk about acceptability. If my volume is calculated to two decimal places, then my volume is accurate to that degree. There is nothing subjective about that being a degree of accuracy. It's not a degree of precision, it's a degree of accuracy. What is subjective is whether or not that is an acceptable degree of accuracy.

--------------
Good Luck
To get the most from your Tek-Tips experience, please read FAQ181-2886
As a circle of light increases so does the circumference of darkness around it. - Albert Einstein

 

[CajunCenturion]

Oh, and more than anything else, yes, I hope it's fun.

--------------
Good Luck
To get the most from your Tek-Tips experience, please read FAQ181-2886
As a circle of light increases so does the circumference of darkness around it. - Albert Einstein

 

[stackdump]

A loud scream was heard.

"What's the matter", said Matt.
"Look where we've ended up", Jenny said in a high pitched voice, "we seem to have crossed into another thread!".
"My god, it's full of geeks arguing about accuracy and precision, let's get out of here", said Matt in a panic.
"What's the difference?", asked Jenny.
"No idea, but I think our threads that way", he said pointing towards a trap door in the floor of the thread.
"This thread gives me the creeps", said Jenny, "it's as though that Maniac character is nearby".

They opened the door, climbed down and there was gentle clunk as the door closed behind them. The silence was broken by a faint shriek, Jenny could be heard screaming, "Oh no!, speed versus velocity!". "Our situation has not improved", exclaimed Matt as they scuttled away in the direction of the Desktop Hardware forum.