July 20, 2012 at 11:32 am
Friends
Well, we HAVE come a long way from T-SQL and TIme Zones, eh?
I've been keeping my head down as I didn't want to step into a flame, but there are a couple of things I thought I would throw out there:
"12 AM/PM"
I always felt slightly uneasy that I have not got all the information I need to be certain I understood what was meant when I hear/read those.
GSquared's excellent illustration slightly leads me to wonder if I was right all along ... but I have an extension
First, I'm pretty sure the whole point of AM/PM use is to assist translation from 12-hour to 24 hour representation. So far so good.
Whilst I agree "12 AM" should strictly be interpreted as "the 12 O'Clock before the Meridian", and aside from describing pole-to-pole circles around the Globe and "imaginary" lines connecting acupuncture points, "Meridian" apparently has a synonym "Zenith" (highest - as in position of the Sun in this context), so "12 AM on <date>" would seem to be <date>T00:00:00.
"12 PM on <date>" would, by similar interpretation as above, be the second after <date>T23:23:59.
I'm expressing it that way because the notion of 24:00:00 is abhorrent to me, and clearly that "second later" is not on <date> at all - more correctly it would be <date+1>T00:00:00 - the oxymoron of which GSquared spoke.
My question here however would be "in AM/PM representation, what happened to the 12 O'Clock that is neither before, nor after, but is the Meridian?".
"midnight on the <date>"
At the risk of entering Chicken and Egg territory (and assuming that "night follows day"), just don't get me started on "Midnight on <date>".
Seems to me such an animal just can't exist as it would again be <date+1>T00:00:00 and nothing to do with <date> at all.
Even the notion of "boundary between dates" is lost when expressions like <date>T00:00:00 are clearly accepted and <date>T24:00:00 is not.
OK - "confusion bit" over ...
As for the Million, Billion, Trillion issue - I believe the "British" notion that these were each six orders of magnitude different was abandoned some time ago in favour of the "US" three orders of magnitude.
All this aside, the colinearity (is that even a word?) with the SI magnitude representation is clear to me - am I alone?
1 thousand Dollars = 1 KiloDollar
1 thousand KiloDollars = 1 MegaDollar (formerly "Million")
1 thousand MegaDollars = 1 GigaDollar (formerly "Billion")
Even the (strictly non-SI) "centi" already has a place:
1 penny (or cent) = 1 centiDollar
100 centiDollars = 1 Dollar
meaning (in England) of course that ..
1 thousand GigaPoundsSterling = 1 pack of smokes and a Beer
.. but that's next year of course - if the UK Government has it's way.
Have a great weekend all - at the risk of starting a whole new thread fork, it's Beer O'Clock here in London, England so I'm off out to play.
Russ
PS - I still think datetime should be an Object in T-SQL - just like XML.
Sorry ... couldn't resist!
July 20, 2012 at 1:06 pm
Russell Lavelle-langham (7/20/2012)
at the risk of starting a whole new thread fork, it's Beer O'Clock here in London, England so I'm off out to play.
That is one definite problem with eliminating time zones. Without them, it won't always be 5:00 somewhere. 🙂
Jay Bienvenu | http://bienv.com | http://twitter.com/jbnv
July 23, 2012 at 6:24 am
jbnv (7/20/2012)
Russell Lavelle-langham (7/20/2012)
at the risk of starting a whole new thread fork, it's Beer O'Clock here in London, England so I'm off out to play.That is one definite problem with eliminating time zones. Without them, it won't always be 5:00 somewhere. 🙂
Yeah, but the sun will be past the yardarm somewhere, and that's more fun to say anyway. (http://wiki.answers.com/Q/What_is_meaning_The_Sun_is_past_the_yardarm)
- Gus "GSquared", RSVP, OODA, MAP, NMVP, FAQ, SAT, SQL, DNA, RNA, UOI, IOU, AM, PM, AD, BC, BCE, USA, UN, CF, ROFL, LOL, ETC
Property of The Thread
"Nobody knows the age of the human race, but everyone agrees it's old enough to know better." - Anon
July 23, 2012 at 6:25 am
jbnv (7/20/2012)
Russell Lavelle-langham (7/20/2012)
at the risk of starting a whole new thread fork, it's Beer O'Clock here in London, England so I'm off out to play.That is one definite problem with eliminating time zones. Without them, it won't always be 5:00 somewhere. 🙂
Yeah, but the sun will be past the yardarm somewhere, and that's more fun to say anyway. (http://wiki.answers.com/Q/What_is_meaning_The_Sun_is_past_the_yardarm)
- Gus "GSquared", RSVP, OODA, MAP, NMVP, FAQ, SAT, SQL, DNA, RNA, UOI, IOU, AM, PM, AD, BC, BCE, USA, UN, CF, ROFL, LOL, ETC
Property of The Thread
"Nobody knows the age of the human race, but everyone agrees it's old enough to know better." - Anon
July 23, 2012 at 6:32 am
Russell:
On the point of "midnight on <date>", it's another ambiguous term that really has no meaning. If one considers the moment of midnight to be both days (most common historical meaning), then there are technically two midnights every calendar date, one at the beginning of the day, one at the end.
The artifice of assuming the day starts at exactly 00:00 and ends one-quantum-unit-of-time before the next 00:00:
a) Assumes infinite accuracy in time measurement, which is absurd
b) Assumes that time is digital, not analog, for which there is no evidence and plenty of counter-evidence
c) Assumes all frames of reference are identical, which violates all of modern physics
We force time to match out watches because we have such limited tools for measuring it. Our flawed tools shape the language, that's all.
- Gus "GSquared", RSVP, OODA, MAP, NMVP, FAQ, SAT, SQL, DNA, RNA, UOI, IOU, AM, PM, AD, BC, BCE, USA, UN, CF, ROFL, LOL, ETC
Property of The Thread
"Nobody knows the age of the human race, but everyone agrees it's old enough to know better." - Anon
July 23, 2012 at 10:01 am
Ponderance...
The definition of noon basically comes down to when the sun is highest overhead. Due to the distance between the sun and the earth, it takes 8+ minutes for us to actually observe that the sun is indeed at it's highest point. And the earth does not go around the sun in a circle, the path is more elliptical in nature so there is a variance on how long that the light gets here; with a variance of about 16 seconds throughout the year. Not to mention that the orbit of the moon affects our orbit, so the earths path may overall be elliptical, it is not a straight curve (for lack of a better term); rather it is sinusoidal path overlayed onto that ellipse.
That would mean that the length of an observable day ( from noon to noon) only average out to 24 hours, and some would be longer and some would be shorter, with a trigonometric offset based upon the moons position to earth.
Suffice it to say, I think we should define noon as when it actually occurs, and not when we can physically observe it.
OK... time for me to get some coffee
Viewing 6 posts - 106 through 110 (of 110 total)
You must be logged in to reply to this topic. Login to reply