Imperial versus metric

[micha]

On 1/5/2014 at 12:00 AM, SargeRho said:

The system that makes the most sense is DD/MM/YYYY, since days change the most often, then months, then years.

Oh well, since this thread has already been resurrected, I may as well throw in my two ha'pennies

So when telling time, you give seconds-minutes-hours too?

 

Again, as others have mentioned, what people say the date as compared to how it should be written down are two completely separate things. All written date / time should go from largest to smallest, as we do with all the rest of our numbers too.

Interesting semi-related factoid : the reason we write numbers effectively from right-to-left ("little-endian" in computer parlance) is because we imported the arab numbering system - which is a right-to-left written language. But then we (mostly, depending on your language * ) say numbers left-to-write since that's how we read them.  

* In german, for example, it's a real mixed bag.  23 is said three-and-twenty (dreiundzwanzig). Which is right-to-left.  123 is one-hundred-three-and-twenty!! 

Reflecting on it, fron an "intuition" perspective it makes a lot of sense to start with the largest first - that gives you a scale to work with.  If you always start with the smallest value you have to wait until you have the whole number before you get the scale. one-and-twenty-and-three hundred-and-four thousand-and five-ten-thousands-and,..... 

Which brings us back to the yyyy-mm-dd date system  

 

 

[kerbiloid]

6 hours ago, mikegarrison said:

everywhere in the world, airplane altitudes are measured in feet and speed is measured in nautical miles per hour.

In Russia - never, and it's not the last country in aviation.

[JoeSchmuckatelli]

On 12/29/2013 at 9:01 PM, Camacha said:

First of all, I am not out to start a flamewar and this topic should not end up being one. I understand different people do things differently. It will always be like that and we should accept that.

I am very much used to the metric system. Most people will be aware that this is very structured, making conversions easier. For fun, I started doing problems in imperial units. Even though I have some idea of what each one should roughly indicate (inches from construction shows, feet from feet, yards from watching football and miles from car shows) I find it utterly impossible to work with the system in a somewhat intuitive and efficient manner. I understand that it is very much a matter of what you are used to, but I can truly not imagine anyone would want to work with something like that voluntarily. I found myself seriously wondering whether this does not regularly lead to conversion problems and all sorts of mistakes and construction problems. I can imagine that processing units in computer systems can be a bit of a hassle too. However, I am simply not used to the system, so maybe all this is not a problem at all.

I thought it might be interesting to share what your experiences with the different systems are I am very curious about the implications of day to day use to the imperial system. Any good stories about the origin or implementation of the systems are welcome too.

Approaching this from 8 years after this was posted - I'm still compelled to answer.

I grew up used to the American system (I could not possibly tell you a man's weight in Stone - but some Brit friends can).  I can also talk confidently in meters and kilometers - thanks to my time in the military.  So much so that it's almost natural - native language... you could say I am fluent in meters and klicks.  But when it comes to millimeters and centimeters I am somewhat still 'translating' - I 'know' how big 5.56 is, just as I know how wide 120mm is - because they are rounds I have used.  But if you tell me something is 7cm... I have to translate that.

The key for me - in knowing meters and klicks is by having stared at, measured, counted and walked in those terms.  For years and years I lived, talked and operated in those - and I talk in those to other military folks.

I don't drive in klicks, however.  I drive in miles.  I bike in miles - but I can walk, even run in klicks.  I can shoot in meters, even range in meters.  But I cannot, for the life of me, drive in meters or kilometers.  At least not in a civilian car, on a road.  I certainly can in a tank across the desert.  I've done that.

Don't get me started on anything other than liters.  I know about how big a liter is.  I regularly carry one on my back.  I cannot tell you whether a glass is a deciliter or some number of milliliters... it just has no meaning.  Cup, pint, quart, gallon - I know them all.  I can guestimate the conversion from pounds to kilos (about half) - but if you asked me how much a man weighs - I'll say he's about a buck-eighty... never ninety.

Fuggitabout it with degrees of centigrade: that's just too pat.  Water boils at 212 degrees.  117 is bloody hot outside.  50?  You need a jacket.  Ask my friend in France what he wears when it's 50, and he starts talking about Global Warming.

I have a sneaky suspicion that the metric system is 'too made up'.  - speed of sound is close to 300m/s.  Speed of light is close to 300,000 km/s  --- waaaay too convenient.  Someone is cheating.  186,000 miles per second makes sense to me.  Counting to 5 for every mile between the flash and bang of lightning tells me how far away the storm is... but I can also count three seconds from the flash of a tank firing and the boom and know its almost a klick away.

All of this is to say - these units of measurement are like a language.  You can learn it, but only translate it to what you know... until you live it.

Only then does any made up system of measurement become real.  

Only by living it - will we ever be fluent in it.

Edited June 18, 2021 by JoeSchmuckatelli

[kerbiloid]

On the lighting flash start counting seconds "21... 22..." and divide by 3 the thunder time., It's in kilometers.

Water boils at 100, freezes at 0, it's very obvious.
Normal body temp is 36.6. Standard office conditioner 22. Cool water is 18..20. Fridge is 4.
-15 is a normal winter temperature in Moscow, and it's too cold for me.

Standard glasses are 180 and 200 grams or milliliters.
Standard vodka bottle 0.5 l. Champagne 0.75. Pepsi glass bottle 0.25 l (and it's strange because since the Soviet times it was 0.33).

Light is about 300 thousand km/s, much more handy than 186 in miles.

Btw, which miles? 1852 or 1609?
If a sailor drives a car, which miles should he use? His miles or these miles? The 1.15 ratio looks weird.

***

The British system could make sense if people were using 12-base numeric system (more handy, of course), but they usually use 10.

But even then it's full of non-standard units, so the metric one is more handy.

Metric is for fiction, Imperial is for fantasy.

[JoeSchmuckatelli]

Its all made up stuff anyway.  As long as enough people agree on the rules, the game is fair.

Think about using watches, clocks and calendars on a planet that rotates once every 27 hours and takes 15.3 months to circle its star. 

A second is a heartbeat, right?  Totally subjective.

 

Eddie: what time is it?

 

Edited June 18, 2021 by JoeSchmuckatelli

[kerbiloid]

21 minutes ago, JoeSchmuckatelli said:

A second is a heartbeat, right?

1/60 th of 1/60 th of 1/24 th, as the Sumerian gods set it.

They would better make it 1/72 th of 1/72 th of 1/24 th. Still more useful.

Also the Ancient Hebrew time units differ.
https://en.wikipedia.org/wiki/Biblical_and_Talmudic_units_of_measurement#Day

Quote

• 1 part (heleq) = 76 moments (rega'im) (each moment, rega, is 0.04386 of a second; 22.8 rega'im is 1 second)
• 1 hour (sha'ah) = 1080 parts (halaqim) (each heleq is 3⅓ seconds)
• 1 day = 24 hours (sha'ah)

Still wondering, how did they measure 3600/(1080 * 76) = 0.04 second intervals.
It's comparable to the CPU clock generator. Makes to start thinking about the stone-and-clay computers.

[YNM]

12 hours ago, RCgothic said:

if all knowledge of the Imperial system were burnt in a fire nothing of value would be lost.

Looks over at us referring to U.S. standards and their official version is in imperial, and there are typos in the metric version and we carried over the typos leading to illogical results

11 hours ago, StrandedonEarth said:

An acre is the area of a rectangle whose length is one furlong and whose width is one chain.

Would've made sense when measuring chains are all you have ! (railway mileage used to be done in chain-miles in the UK, they now technically still do that but many design documents use metric)

Also still more sensible than using ft for railway mileage like in the US, even more so in design documents... like seriously it's waay too small for the scale of stuff you're working...

6 hours ago, micha said:

So when telling time, you give seconds-minutes-hours too?

Again, as others have mentioned, what people say the date as compared to how it should be written down are two completely separate things. All written date / time should go from largest to smallest, as we do with all the rest of our numbers too.

I will say, I've always used DD/MM/YYYY nearly all my life (all 19 years ever since I properly understand stuff) since that's how we would explain it in words, but after learning languages where they naturally do things YY年MM月DD日 that alternative makes sense too.

4 hours ago, kerbiloid said:

The British system could make sense if people were using 12-base numeric system (more handy, of course), but they usually use 10.

There was actually a proposal to use base 12 numbers.

Edited June 18, 2021 by YNM

[kerbiloid]

There is no problem with finger-counting of  base 12

Spoiler

And see what true professionals were able to do with base 10.

Spoiler

Finger positions used for counting up to 9999 from Luca Pacioli's 1494 Summa de arithmetica, based on the earlier Arabic system.

I believe that using a double-base system would be the best (6 an12 as double-6), but it's hard to do this on my own, when everything around is base 10.

[SargeRho]

12 hours ago, micha said:

Oh well, since this thread has already been resurrected, I may as well throw in my two ha'pennies

So when telling time, you give seconds-minutes-hours too?

No, it may as well be YYYY/MM/DD, whether it's left-to-right or right-to-left doesn't really matter to me.

What really scrambles my neurons, is YYYY/DD/MM, or MM/DD/YYYY

[kerbiloid]

From MS SQL Server dox.

Spoiler

Date and Time styles

For a date or time data type expression, style can have one of the values shown in the following table. Other values are processed as 0. Beginning with SQL Server 2012 (11.x), the only styles supported, when converting from date and time types to datetimeoffset, are 0 or 1. All other conversion styles return error 9809.

 Note

SQL Server supports the date format, in Arabic style, with the Kuwaiti algorithm.

DATE AND TIME STYLES

Without century (yy) (1)	With century (yyyy)	Standard	Input/Output (3)
-	0 or 100 (1,2)	Default for datetime and smalldatetime	mon dd yyyy hh:miAM (or PM)
1	101	U.S.	1 = mm/dd/yy 101 = mm/dd/yyyy
2	102	ANSI	2 = yy.mm.dd 102 = yyyy.mm.dd
3	103	British/French	3 = dd/mm/yy 103 = dd/mm/yyyy
4	104	German	4 = dd.mm.yy 104 = dd.mm.yyyy
5	105	Italian	5 = dd-mm-yy 105 = dd-mm-yyyy
6	106 (1)	-	6 = dd mon yy 106 = dd mon yyyy
7	107 (1)	-	7 = Mon dd, yy 107 = Mon dd, yyyy
8 or 24	108	-	hh:mi:ss
-	9 or 109 (1,2)	Default + milliseconds	mon dd yyyy hh:mi:ss:mmmAM (or PM)
10	110	USA	10 = mm-dd-yy 110 = mm-dd-yyyy
11	111	JAPAN	11 = yy/mm/dd 111 = yyyy/mm/dd
12	112	ISO	12 = yymmdd 112 = yyyymmdd
-	13 or 113 (1,2)	Europe default + milliseconds	dd mon yyyy hh:mi:ss:mmm (24h)
14	114	-	hh:mi:ss:mmm (24h)
-	20 or 120 (2)	ODBC canonical	yyyy-mm-dd hh:mi:ss (24h)
-	21 or 25 or 121 (2)	ODBC canonical (with milliseconds) default for time, date, datetime2, and datetimeoffset	yyyy-mm-dd hh:mi:ss.mmm (24h)
22	-	U.S.	mm/dd/yy hh:mi:ss AM (or PM)
-	23	ISO8601	yyyy-mm-dd
-	126 (4)	ISO8601	yyyy-mm-ddThh:mi:ss.mmm (no spaces) Note: For a milliseconds (mmm) value of 0, the millisecond decimal fraction value will not display. For example, the value '2012-11-07T18:26:20.000 displays as '2012-11-07T18:26:20'.
-	127(6, 7)	ISO8601 with time zone Z.	yyyy-MM-ddThh:mm:ss.fffZ (no spaces) Note: For a milliseconds (mmm) value of 0, the millisecond decimal value will not display. For example, the value '2012-11-07T18:26:20.000 will display as '2012-11-07T18:26:20'.
-	130 (1,2)	Hijri (5)	dd mon yyyy hh:mi:ss:mmmAM In this style, mon represents a multi-token Hijri unicode representation of the full month name. This value does not render correctly on a default US installation of SSMS.
-	131 (2)	Hijri (5)	dd/mm/yyyy hh:mi:ss:mmmAM

¹ These style values return nondeterministic results. Includes all (yy) (without century) styles and a subset of (yyyy) (with century) styles.

² The default values (0 or 100, 9 or 109, 13 or 113, 20 or 120, 23, and 21 or 25 or 121) always return the century (yyyy).

³ Input when you convert to datetime; output when you convert to character data.

⁴ Designed for XML use. For conversion from datetime or smalldatetime to character data, see the previous table for the output format.

⁵ Hijri is a calendar system with several variations. SQL Server uses the Kuwaiti algorithm.

 Important

By default, SQL Server interprets two-digit years based on a cutoff year of 2049. That means that SQL Server interprets the two-digit year 49 as 2049 and the two-digit year 50 as 1950. Many client applications, including those based on Automation objects, use a cutoff year of 2030. SQL Server provides the two digit year cutoff configuration option to change the cutoff year used by SQL Server. This allows for the consistent treatment of dates. We recommend specifying four-digit years.

⁶ Only supported when casting from character data to datetime or smalldatetime. When casting character data representing only date or only time components to the datetime or smalldatetime data types, the unspecified time component is set to 00:00:00.000, and the unspecified date component is set to 1900-01-01.

⁷ Use the optional time zone indicator Z to make it easier to map XML datetime values that have time zone information to SQL Server datetime values that have no time zone. Z indicates time zone UTC-0. The HH:MM offset, in the + or - direction, indicates other time zones. For example: 2006-12-12T23:45:12-08:00.

When converting smalldatetime to character data, the styles that include seconds or milliseconds show zeros in these positions. When converting from datetime or smalldatetime values, use an appropriate char or varchar data type length to truncate unwanted date parts.

When converting character data to datetimeoffset, using a style that includes a time, a time zone offset is appended to the result.

 

[wumpus]

On the whole idea of "base 10 vs base 12 or base 60" (the base 60 system that time and circles still use dates back to Babylon)...

Sometime around 2000 the New York  Stock Exchange went from stocks having their traditional fractions to going decimal.  I suspect IBM was behind it, because they're the only computer company that really includes a "base 10 mode" (Intel has one, but it is so limited it appears to be included as "checkbox engineering").  The fractions NYSE used were all naturally base-2, and were as easy for a computer to do the calculations as any old timer could.  Simply define the "decimal point" three bits to the right and your 16 bit word (expect to special case Berkshire Hathaway, or just use 64 bits for everything) now deals with integer values of 1-8192 and fractions 1/8 to 7/8.

Lets just say anyone who thought decimalization made things easier for computers doesn't get the "10 types of people" joke.  Probably including far too many modern programmers.

16 hours ago, NFUN said:

Every single Mech E major I knew would convert the imperial units they were given in problems to metric to do the work and would convert them back at the end, even though they were mostly American and they were taught in Imperial in most classes since the beginning of school

 

Just say no.

My first job was at a startup working under a French engineer who would often argue with the American fabricator/technician about "sillymeters" (I was also told that "soccer" was pronounced "foitball" and football was pronounced "football" (emphasis more on the long oo than the t)).  The Frenchman once mentioned that he didn't have much of a problem working with Imperial units as he would convert everything to metric and convert the final answer back.  The tech admitted that he did it that way, and so did everybody else.

[mikegarrison]

3 hours ago, wumpus said:

My first job was at a startup working under a French engineer who would often argue with the American fabricator/technician about "sillymeters" (I was also told that "soccer" was pronounced "foitball" and football was pronounced "football" (emphasis more on the long oo than the t)).  The Frenchman once mentioned that he didn't have much of a problem working with Imperial units as he would convert everything to metric and convert the final answer back.  The tech admitted that he did it that way, and so did everybody else.

These are mainly old-timey engineering stories. Nowadays everybody uses software that will automatically calculate any units you want, so it doesn't really matter.

[RCgothic]

For file names I definitely prefer YYYY/MM/DD. Sorts alphanumerically in an order that makes sense.

[kerbiloid]

Just now, RCgothic said:

For file names I definitely prefer YYYY/MM/DD

"/" ?

Just now, kerbiloid said:

file names

 

[mikegarrison]

Just now, kerbiloid said:

"/" ?

Yeah, using "/" is not a great idea for file names.

"\" is not good either.

Nor is "." or "?" or "*" or "[" or "]" or ... well, suffice to say there are a lot of characters that shouldn't be used for file names. Letters, numbers, and "_" are the only really safe ones. "." is used in many file names, of course, but some systems will get really confused if you don't use it exactly as they expect you to.

[kerbiloid]

I use yyyy.mm.dd-hh.mm.ss for files.

[RCgothic]

Haha, that's what I meant. Just the order is important, not the separator as long as it's not something incompatible. I generally use "-".

[tater]

If people were serious, they'd use the decimal time they also invented along with metric.

 

[YNM]

1 hour ago, mikegarrison said:

Nowadays everybody uses software that will automatically calculate any units you want, so it doesn't really matter.

Would say that it does matter when we're referring codes and standards. The differences are slight, sure, and conservatism is welcome, but where do you cut the line becomes more glaringly obvious than otherwise.

[wumpus]

1 hour ago, mikegarrison said:

These are mainly old-timey engineering stories. Nowadays everybody uses software that will automatically calculate any units you want, so it doesn't really matter.

Mostly true, but I've used software I rather liked that typically wanted all length measurements in either imperial (mills) or metric (mm), and thought that it made more sense to do the occasional calculation myself when the devices were speced the "wrong" way than to flip the thing back and forth between metric and imperial (even though it was probably storing everything as a double and it wouldn't have been an issue).

Also the story took place in a startup incubator on a college campus, so including school calculations without the crutches made sense.

Finally found it: long, long ago when i was studying high school chemistry, the teacher was teaching the "factor label method".  This was a huge help to me in that it not only solved basic high school chemistry problems, it also was used to solve any conversion issues in any physical science problem in any other class (or professionally, although you'd probably use some sort of similar software).  This video even includes his "bar notation", something I haven't seen during my occasional googles of "factor label method" https://www.youtube.com/watch?v=K33txxFsnrg

[Spacescifi]

 

I prefer metric.

 

I only recently found one of no doubt several sites that converts various rates of gravity acceleration into meters traveled per second.

 

With that knowledge one can easily calculate how long travel will take at any gravity of acceleration listed for any distance measured in kilometers. Since there are 1000 meters in  a kilometer, if you had a 1000g orion drive missile, one pulse burn would allow it to coast  10,000 kilometers in a bit over TEN SECONDS!

Obviously very useful for space scifi.

Because what scifi space show or book ever uses Imperial measures?

 

Kilometers is what everybody uses and everyone KNOWS what you mean.

 

It's just more accessible, and that is what scifi often wants to be to the reader.

 

If you really want to disconnect the reader, throw hawking and relativity equations at them while using imperial measures...during space combat!

Edited June 19, 2021 by Spacescifi

[razark]

Proper date format is DD Mon YYYY or for filenames DD_Mon_YYYY.

Edited June 19, 2021 by razark

[SingABrightSong]

The main advantage of US Customary is that the basic units, in particular units of length are very intuitive for humans. The main advantage of SI units is that they're base-ten and easily convertible. I propose, then, a base-12 system based off the customary foot, such that twelve feet make, let's call it a "wheel" for now, and so forth, and basing the measures for volume off the cubic foot rather than the cubic meter.

[magnemoe]

6 hours ago, tater said:

If people were serious, they'd use the decimal time they also invented along with metric.

 

The French tried but it failed, 10 days weeks was an obvious nerf with one day a week off 
10 hour a day is too long hours for stuff like shifts. 
20 hours and 100 minutes and 100 seconds would work. 
 

[Beccab]

9 hours ago, mikegarrison said:

These are mainly old-timey engineering stories. Nowadays everybody uses software that will automatically calculate any units you want, so it doesn't really matter.

Mars Climate Orbiter would like to have a word with you about how it doesn't matter