Talk:12-hour clock/Archive 1

This page is an archive of past discussions. Do not edit the contents of this page. If you wish to start a new discussion or revive an old one, please do so on the current talk page.

6تا6667ع7778000007غلخهتنخ9

a.m. A.M. am AM & What about p?

There is an article a.m., but no article p.m. - I was just wondering whether it was necessary to have an a.m. article, when it seems that this one will suffice. If it is deemed necessary, perhaps there should be a p.m. article as well (though I think the first solution might suffice).

Additionally, although I have not yet found a mention in the Wikipedia:Manual_of_Style, I think it might be a worthy effort to begin discussion here concerning whether to use A.M., AM, a.m., or am (Which is what seems to be in use here and is preferred by myself). Should anyone feel that they can add to this please do. User:Moogle 15 Dec 04

There already was a "Post Meridiem" article as well as a "Ante Meridiem" article in addition to the A.M. article. I made the shorter Ante Meridiem a redirect to the A.M. article and made Post Meridiem redirect to a new P.M. article for consistency. I'm not so sure we need either, but I don't have strong feelings about it. As far as A.M., AM, a.m., and am I prefer am with A.M. as a second choice. I think a.m. is just wrong, and AM isn't as typographically pretty as am. Zenyu 18:37, Dec 15, 2004 (UTC)

I thought that the A.M. article was silly the first time I saw it, so I'll merge it and P.M. here. Typographically, I dislike "am" because it's a word, and I also dislike periods (although I don't see how you can claim that they're wrong). Thus I use "AM", although I agree that it's not pretty; that's because it really should be in small caps: "AM". -- Toby Bartels 00:12, 4 Jan 2005 (UTC)

In making my other changes, I edited enough of the article that I decided to put the whole thing in my style (complete with small caps), rather than the previous style. But please don't interpret this as a request to institute a Wikipedia-wide style convention; it's enough that each article be internally consistent (to look good). And I'll hardly be upset if a future rewrite of this article establishes some other style here for a while! ^_^ -- Toby Bartels 02:32, 4 Jan 2005 (UTC)

At some point, a Wikipedia-wide style convention was developed, and is recorded in the Manual of Style. The winners were "a.m." and "p.m.". I have updated this article accordingly.--Srleffler 04:39, 8 April 2006 (UTC)

"especially common in the USA?"

Isn't it more widespread than just the U.S.? I'm pretty sure it's used throughout Canada (except possibly Quebec) as well.

Yes it is widely used in Canada, generally excluding quebec and other French dominated regions such as Northern Ontario

Begging your pardons, but this debate, and the resulting explanation soup of "usage" on the front page of this entry, seems to entirely miss the key point. Which is that the statement "12 PM is not noon" leaves the door wide open for all those less fortunate who will read it and say AHA, so "12 PM is midnight" or so "12 AM is noon" (*tears hair out*). I have visited an optometrist, located in an upscale mall and associated with a national chain, who still proudly displays a sign stating that they are open until 12 AM every Saturday. Yet, they don't understand why they have to keep saying no to Saturday afternoon and evening apointments.

Please have a signature for your messages. Thank You! --S iva1979^{Talk to me} 04:47, 26 April 2006 (UTC)

Reason for use

Can any one explain why certain countries still use the 12-hour clock --AzaToth ^talk 11:47, 15 November 2005 (UTC)

Because most Americans think that the 24 hour clock is used only in militry settings, for some damned reason or another? Because the vast majourity of my countrymen are ignorant curs? I'm curious too, because most people here are so ****ing ignorant of the outside world, and every one of my teachers wants to hurt me for writings dates as 2005 12 01 or 01 December 2005, et al; and my own family tells me to "Go live in France" when I use the 24 hour clock. 63.146.46.202 05:37, 2 December 2005 (UTC)

It is cultural. 60/5 is 12. 24 is wrong.

Infinitely short periods in time?

"Noon and midnight are only infinitely short points in time, and therefore it is not practical to use any other convention than that which also applies immediately afterwards, when the clock still displays 12:00."

Using this logic, the minute hand would never be able to leave the 12:00 position as anything immediately after would also be 12:00...

The fact that a point has no length has nothing to do with it - the 3am point also has no length but it is definitely am.

I believe that the reason that it doesn't make sense to allocate either am or pm to 12:00 is that it is a point that joins two continuiums, am and pm (similar to two rulers, one red and one blue that are laid end to end). The join point (12:00) is neither am or pm, just like the join point of the two rulers is neither red nor blue. User:24.5.38.74

The point I was trying to make when using the term "infinitely short points in time" could be explained far more precisely, but unfortunately such an explanation would assume that the reader is familiar with the notion of open and closed ends of intervals in mathematics. The explanation could then go like this:

It is customary that the time interval associated with a given time display is closed at the start and open at the end. For example, the fourth hour of the day, during which a U.S.-style digital clock displays 3:MM:SS am, covers the time interval [03:00, 04:00). Likewise, the first minute , during which a U.S.-style digital clock displays 12:00:SS am, covers the time interval [00:00, 00:01), etc. If we accept that convention, then it only makes sense – in the interest of consistency – to also interpret the time intervals denoted by the "am" and "pm" periods as closed at the start and open at the end. In other words, "am" covers the interval [00:00, 12:00) and "pm" covers the interval [12:00, 24:00). Doing it the other way round would lead to the oddity that a clock would have to flip between am and pm an infinitely short time after it flips from 11 to 12.

Such an explanation might appeal more to readers with a good background in mathematics, but I don't know whether we can assume that here. (On the other hand, people come to Wikipedia to broaden their mind, so perhaps it is ok to explain first what open and closed interval ends are?) Markus Kuhn 09:47, 26 April 2006 (UTC)

Neutrality

Does not meantion the advantages of a 12-hour clock. Yes, there are advantages e.g. 60/5=12. (anonymous)

And what exactly is the advantage of that? Sorry, this really is not at all obvious to me. The day still lasts 24 hours, even with the 12-hour notation. The neutral-point-of-view policy does not imply that we must have a section of advantages. Any advantage or disadvantage that we name must be relevant, clearly explained, and well backed by sounds arguments and/or references. In my opinion, the recently added new advantages section falls seriously short of these requirements. Markus Kuhn 12:50, 10 May 2006 (UTC)

It is my personal opinion that the disadvantages outweigh the advantages by far. That does not say there are no advantages. Also aspects perceived as advantages by some people can be relevant to mention. This is common practice in many other articles with contested neutrality. Instead of removing alleged advantages right away, we could try to formulate them more clearly as they are, fact or perceived. −Woodstone 13:14, 10 May 2006 (UTC)

But we do not have to list clearly misleading arguments. In particular the item

Some people dislike the idea that midnight has two different notations, depending on whether it is the beginning or end of the day.

is merely a (deliberate?) misunderstanding by the original author. The 24-hour notation – as widely implemented by digital clocks and computers – ranges from 00:00 to 23:59 (or in mathematical interval notation [00:00,24:00[), therefore the notation "00:00" is the unambiguous notation for midnight and the new day starts with the rollover from 23:59:59.999 to 00:00:00.000. The 24:00 notation (mentioned in ISO 8601) exists primarily to remind people about what the exact meaning of 00:00 is if it is used together with a date. As a secondary use, 24:00 may actually be written down at the end of an interval, to avoid having to reference the next day explicitly. The above is a very misleading attempt to misrepresent a clear advantage as a disadvantage. It is neither fair nor based on how the 24-hour notation is used in practice. It should therefore be removed. Markus Kuhn 16:21, 10 May 2006 (UTC)

I see this as a disadvantage of the 24-hour system. (anonymous)

One can only say what is an advantage or a disadvantage in their own mind. Therefore one can't say what is not an advantage or a disadvantage in someone elses mind. The same point can be both an advantage and a disadvantage to different people. Wikipedia should state both sides of the issue and let the reader deside for him/herself, thank you. Zginder 22:16, 11 May 2006 (UTC)

What of the above exactly does the anonymous contributor see as a disadvantage? The lack of ambiguity? You must at least be able to explain your point of view in a coherent way before it can even be considered to remain in an encyclopedia article. Markus Kuhn 10:34, 12 May 2006 (UTC)

First of all, "the anonymous contributor" is me, "Zginder." I get forget to sign in sometimes and today it said I was signed in when I wasn't. Second, I wrote a perfect piece of prose that can't be disagreed with and you disagree with me. Third, if you do not use the 12-hour clock you have no right to tell me how to think in regard to the advantages of it. Fourth, all of the sentences under Criticism and practical problems are statements of fact while when I make the advantages also statements of fact they are changeed to "Some people" a weasel word making them statements of opinion. Fith, I have not changed the Criticism and practical problem, but all anybody else semes to do is to delete or change the Advantages. Sixth, the advantages should always go before the disadvantages; however, someone moved the advantages to after Criticism and practical problems. Zginder 1:46 PM, 13 May 2006 (UTC)

Well, the article now lists several advantages of the am/pm system. I can offer even more. You'll sell a lot more am/pm clocks than 24-hour clocks in the U.S. (To make it clear, that means more profit for the seller!) Also, if you have a Web form to be filled out and one of the fields is the time, and you set it up for the 24-hour format, I guarantee someone will put down 5:00 when they should be putting down 17:00 (Advantage: to avoid mistakes and ambiguity). Yet another is the existence of clock towers. A 24-hour clock on a clock tower would be much more difficult to read and not look as good.

Still, I greatly prefer the 24-hour format on computers, especially for file dates. I find the use of am/pm times in a list of files extremely awkward and ugly. [Alan E. Feldman 2006/07/02 16:15 UTC]

What is wrong with these statements?

The 24-hour clock does not say what day midnight belongs to (to take this to completion one cannot say two days instead one must say two days and two midnight. There are 365.2424 days and midnights in a year. etc.) by saying midnight is 12:00 a.m. the 12-hour clock assigns midnight to the day which is beginning.

In using the 12-hour clock one can state which of two events, one in the morning and one in the evening, without a time. (e.g. "The a.m. meeting or the p.m. meeting?" "The a.m.") In these situations the time is known to all. The 24-hour clock doesn't have any way to do that without saying early morning, morning, dawn, forenoon, noon, afternoon, dusk, evening, night, or midnight with are vague, some longer to say, and can change with season.

Zginder 4:20 p.m., 20 July 2006 (UTC)

For the first bullet, the 24-hour clock very precisely states what day midnight belongs to. It is the boundary between two days: the end of one day and the beginning of the next day. In order to avoid confusion about what is meant by midnight on July 20, 2006 one can say 2006-07-20 00:00 for midnight at the beginning of the day and 2006-07-20 24:00 for the one at the end. Every midnight is shared between two days.

For the second bullet, how would saying the morning(afternoon) meeting be any less clear than the a.m.(p.m.) meeting?

−Woodstone 17:51, 20 July 2006 (UTC)

For the second, a.m./p.m. are more encompassing than any other terms in English. All others (Except for noon and midnight) can change times throughout the year. Zginder 11:48 p.m., 20 July 2006 (UTC)

Do you also think people have trouble knowing when to say good morning/afternoon/evening/night? I hear people saying that correctly all the time. I never heard anyone wishing me good a.m. or p.m. −Woodstone 22:21, 20 July 2006 (UTC)

That's not the point. What if you have some thing at 5:00 a.m. and 7:00 p.m. Whether it is light or not depends on the season. That is the point. They are also shorter and people do say "good, what is it now afternoon or evening." Zginder 1:03 p.m., 21 July 2006 (UTC)

I completely do not understand what you're at. Why is it important if it's dark or light? Is it dark during a.m. and light during p.m.? In your example almost any answer would be clear: morning or night would be at 05:00, afternoon or evening would be at 17:00. Are you implying that people often ask you ; "is it a.m. or p.m.?" I never heard that and there is no need for it. And is saying "a.m." really shorter than "morning"? —The preceding unsigned comment was added by Woodstone (talk • contribs).

If it is dark at 7:00 p.m. it is night; however, if it is light it is the evening, therefore if it is an event that happens at 7:00 p.m. every day or every week one can not say evening or night when reffering to all happening of the events. One must state a time. One more thing you wrote that 7:00 p.m. is 17:00 if you think this how can you be editing this article. Also, I think that in the article the 12-hour clock we should use the 12-hour clock. Zginder 8:17 p.m., 21 July 2006 (EDT)

Saying "a.m." isn't shorter, but writing it is. -- 65.68.75.155 04:59, 28 July 2006 (UTC)

Sumerians and half open intervals

Moved from the article:

Some people claim that changing the notation if the clock destroys cultural identity and abominates the language, because the day begins with midnight and ends right before midnight. In other words, it is customary that the time interval associated with a given time display is closed at the start and open at the end. If we accept that convention, then it only makes sense – in the interest of consistency – to also interpret the time intervals denoted by the period of a day as closed at the start and open at the end. In other words, "day" covers the interval [12:00 AM, 12:00 AM).

The Sumerians had a sexagesimal or base-60 numbering system. So that explains why there are 60 seconds in a minute and 60 minutes in an hour.. Now look at degrees, notice how 40°12'13" is 40 degrees, 12 minutes, 13 seconds... There are 360 degrees in a full orbit, or 60*60/6. Let's try dividing 60 by 5, we get 12... Why 5? well 60 can be factored these primes 2,3,5 any multiple of them makes a nice simple divisor, so 2,3,5,6,10,12,15,18,30 are your basic options, of those the only ones that make sense here are 3, 5, 6. Why they chose 5 not 3 or 6 I can't tell you, maybe they just thought 15 or 18 hours were still excessive, and 10 too few? Maybe 12 was better because it could be divided twice in half? Note, 24 is not in the list 2,3,5,6,10,12,15,18,30. This is a problim with clocks with analog dials.

I really do not see why the story about Sumerian number systems represents an advantage. If you want to include that story it might perhaps better fit in a "background" or "history" section. Anyway it should be rewritten, as it is rather conversationally formulated, not encyclopedic. And it should be Substantiated with sources, as it looks like original research, which is not accepted in wikipedia.

Similarly, I do not see how the story about half-open intervals presents an advantage. It sounds like a very specific (even personal), not general interpretation of what a clock time is meant to mean.

−Woodstone 07:49, 10 May 2006 (UTC)

I agree, but right now it just says "Some people claim that changing the notation of the clock to 24-hours destroys cultural values and identity and abominates the language," which is bewildering. I think we should remove that too. 132.236.113.119 07:03, 9 July 2006 (UTC)

It was removed several times and reverted by others. Anyway it starts by saying "Some people claim ...", so it's not untrue and doesn't really harm (the strength of expression exposes its silliness). −Woodstone 10:36, 9 July 2006 (UTC)

Consistency between 12-hour clock and 24-hour clock

Can we ensure that there is consistency between 12-hour clock and 24-hour clock? The issues are almost identical but the articles are inconsistent.

Perhaps we should have a single article. bobblewik 07:27, 15 May 2006 (UTC)

Use by country

We now have a "Use by country" section in both the 12-hour clock and 24-hour clock pages, which will lead to redundancies and inconsistencies. How about moving this entire section into a new separate article Date and time notation by country, were contributors interested in giving a detailed account on the related conventions have all the space they need? Markus Kuhn 13:10, 15 May 2006 (UTC)

I concer. —The preceding unsigned comment was added by 68.239.157.115 (talk • contribs).

In the "History and usage" section, I suggest United Kingdom be moved from "It is the dominant form of time written and spoken in" to "It is commonly used alongside the 24-hour clock in". In speach the 12 hour clock is almost exclusively used, and shop opening times are, for example "open 9 to 5, Monday to Saturday". But in timetables (bus, train, plane), airport and railway station departure boards, any technical document and many official documents the 24 hour clock is used to remove ambiguity.

Translations of a.m. and p.m.

They both come from Latin and are "ante meridiem" and "post meridiem" in Latin. We all agree on that, right? Now "ante" means "before" and "post" means "after," right? Now for meridiem, which is a Latin idiom literally meaning "the middle day" (see wikitionary under meridies). Meridiem is transliterated into English as meridian. (I don't want to see someone change this back to meridem which not an English word and if you think it is find it in a dictionary.) Now Woodstone thinks that "midday" is the best translation; however, this word has a connotation of the middle part of the day and not a point in time. The middle of the day is a little better but can still have that connotation. Forenoon and Afternoon both have a connotation of the day light time before and after noon. So, I suggest using "before noon" and "after noon"(two words) to express the meanings of these phases even tho "noon" can also mean the middle part of the day, and "after noon" can be confused with "afternoon." Zginder 2:10 PM, 10 June 2006 (UTC)

Maybe we could say (ante meridiem(a.m., Latin for before the middle point of the day.) Zginder 2:15 PM, 10 June 2006 (UTC)

Midday is both a literal and a correct translation. Also in Latin the word "meridies" is both used to stand for the moment of noon and for the afternoon (and also "south"). There is also a Latin word meridianus, which is the adjective form, leading to the English "meridian". Translations as "the middle of the day" are just as correct, since Latin does not have articles. In my view giving a whole list of hardly different translations is only confusing. I have no objection against adding "before/after noon" as (the most idiomatic) translation. −Woodstone 16:22, 10 June 2006 (UTC)

I don't think we should have that many translations either but it keeps being changed and we need to decide what should be there. Zginder 11:39 PM, 10 June 2006 (UTC)

OK, if we accept before/after noon as the correct translation, then it is still a problem. What exactly does one mean by noon? Back when am and pm originated, people used Local Solar Time. In that scenario, noon is the middle of the day (between sunrise and sunset) and is exactly when the sun crosses the meridian which is then the same as "12:00 noon". But with our current time system you'd have to be at one of a finite set of longitudes which vary with date for 12:00 noon to be the instant when the sun crosses the meridian because of the existence of time zones, the equation of time, and such. So with our current timekeeping system, noon is not defined precisely enough to apply the literal translation argument. Our noon is not the same as their noon. So what's the point of all this fuss over a literal translation when we are mixing different "noons" from different eras? No one uses arguments like this for anything else, including AD and BC! Why do some insist on making an exception for am and pm? Alan E. Feldman 2006-12-15 18:11 UTC.

Weird sentence

It is more efficient to split a series up in to two parts. For example we do not have different names for negative numbers; we use the − sign.

I don't understand this sentence, and the use of "we" is problematic. What do negative numbers have to do with time notation? Could someone who knows what the argument is supposed to be rephrase it, and write in a way that doesn't show that the author uses the 12-hour clock? Thanks, PruneauT 12:38, 16 June 2006 (UTC)

Since no one seems to answer, i've deleted the sentence. Pruneau^T 00:29, 20 June 2006 (UTC)

Zginder put it back in. I think the argument is goofy. It is not "more efficient" to split any arbitrary linear sequence into two parts. "More efficient" by what measure? Presentation size? No. Internal numeric representation (say in binary bits or BCD (binary coded decimal))? No. I've removed it but we'll see if it comes back. tbird20d. (sorry, editing as anonymous).

More efficient in human brain power, easier to examine and logic. Zginder 12:46 PM (PDT) 2006 June 21.

Australia - Double negative

"It is unlikely that Australia will never migrate to 24-hour notation." What did the author intend to say? I assume it was "...will ever...", but I've nothing to back it up.

I copied it from 24-hour clock and I was confused by it. Copied question over there. Zginder 13:29, 27 July 2006 (UTC)

Manual of Style

Regular editors of this article may be interested in some proposed changes to the use of am/pm/noon etc. in the Manual of Style (WP:DATE), relying in part on this article. See Wikipedia talk:Manual of Style (dates and numbers)#Imporvement about guidelines in Time. — Joe Kress 16:28, 8 August 2006 (UTC)

Weird statement

"the day starts at 12 midnight instead of 6 a.m. when the sun comes up." The sun doesn't (always) come up at 6 a.m. Why is this here? - anomynous passer-by

Removed content

I removed two paragraphs from the article.

From the section Advantages over the 24-hour clock: because this sentence doesn't show an advantage of 12-hour clock over 24-hour clock. Maybe it fits better in another part of the article but it's a claim that would need some citing.

Some people claim that changing the notation of the clock destroys cultural identity and abominates the language.

From the section Pronunciation: because the paragraph is about the pronounciation of 24-hour times. I'm including this in here: 24-hour clock.

Military circles use the 24-hour clock exclusively and would typically pronounce times ending in :00 minutes as the hour followed by "hundred" with an optional "hours". For instance, 16:00 would be pronounced "sixteen hundred" or "sixteen hundred hours".

--Bisco 00:06, 1 October 2006 (UTC)

Typography - half hours

From the Typography section:

hence 5:30 is half past 5 or merely half five

Is this correct? I thought 5:30 would be half six.--Bisco 00:11, 1 October 2006 (UTC)

I happened to notice this older remark when inserting mine below. Yes, it is correct. In English "half five" is short for "half past five". There is no expression "half to five" that could be shortened to it (unlike in German). −Woodstone 14:14, 15 October 2006 (UTC)

Chiming 24 times?

Does anyone know of a clock chiming up to 24 times? It is mentioned in the article, but I doubt if any exist. −Woodstone 14:10, 15 October 2006 (UTC)

It was certainly common centuries ago - see striking clock for a quote about a famous example from 14th century. But I very much doubt whether you'd find a similar clock that still does it today, and even in Italy they changed most of their clocks to conventional striking ages ago. Cormullion 15:16, 15 October 2006 (UTC)

XXIIII

I am just wondering whether anyone knows anything about this, I am not sure, that being said, I believe it should be XXIV to conform with roman numeral system, however I have an old grandpa clock that has four inscribed as IIII not IV so does anyone know if its a common thing to have that in clock? 4, 14, 24 on clocks might be commonly stated on clocks as IIII, XIIII, XXIIII 86.108.117.175 23:18, 27 October 2006 (UTC)

IIII and related Roman numerals are discussed at Roman numerals#IIII or IV?. — Joe Kress 05:31, 30 October 2006 (UTC)

Conformity and consistency are a bit of a modern concern - I don't think it was much of an issue in a world where every country and sometimes every state or city had its own conventions and units. As for 24, here are some examples of numbering schemes for the varying numerals (4, 9, 14, 19, 24):

Florence, Venice Rialto, Venice St Marks, Padua, Clusone: IIII ... VIIII ... XIIII ... XVIIII ... XXIIII
Cremona: IV IX XIV XIX XXIV
Mantua: IIII IX XIIII XIX XXIIII
Shepherd clock Greenwich: IIII IX XIV XIX 0 (;-))

so with this set of data, the 'I's have it, just... Cormullion 14:22, 31 October 2006 (UTC)

Advantages of the 12-hour clock

This section needs work on a number of counts. Adding some citation is certainly ok, but logic stands on its own merit, citation does not make logic any more or less correct.

The language needs to be cleaned up / clarified
The content is not very consistent with the section title. The section seems to address:
- disadvantages of 24 hour numbering
- advantages of *using* a 12-hour nomenclature (not just advantages of its existence per-se)
The arguments about "some people", while easily believable without citation, are weak arguments unless the thoughtfulness of those people can be backed up by more explanation, or the significance of that population could be backed up with some citation, etc. "Some people" may object to a heliocentric solar system, but if the *purpose* of the section is to present the strength of their position, and not just state that it *exists*, then the reader needs to be given more support for why that position has merit. Presenting the arguments well will add more value than attributing them to any particular source.

Incidentally, I removed the word "rare" from the description. Among countless other users internationally, the standard use by United States military (as mentioned elsewhere in the article) is alone enough to argue that 24 hour time is not rare. DKEdwards 18:59, 26 May 2007 (UTC) (Some earlier discussion exists in the archive: Talk:12-hour_clock/Archive_1#Neutrality.)

This section is full of weasel words and sounds like something an American wrote to "defend" the 12 hour clock. Not very encyclopedic.

"""It's always possible to make the distinction between a.m. and p.m. in writing, whereas in isolation a time of "7:00" with no suffix as with the 24-hour clock might be in the morning or evening (however, this is paradoxically caused by the prevalence of 12-hour notation)"""

This argument doesn't really make sense for pro-12 argument, since 7:00 is perfectly unique in a 24h clock, while its ambiguous in a 12-hour notation. In fact I would say that this is by far the biggest disadvantage of the 12h clock. -- 83.135.222.207 05:27, 5 July 2007 (UTC)

Please excuse my bluntness, but you have it exactly backward. Let's say I come across a time given as 10:00. How am I to know whether this was written assuming the 12-hour system or the 24-hour system? It's only unambiguous if there is some way to tell for certain that it was written assuming the 24-hour system. Many 12-hour clocks don't have a.m. or p.m. indicators, so this is not a contrived example. Now, if you always use a.m. or p.m., then there is no ambiguity because the mere presence of a.m. or p.m. tells you unambiguously that the 12-hour system is assumed. I believe that was the point.

--Alan E. Feldman 2007-07-05 5:48 UTC

The above comment really shows a distorted view. You see a time 10:00 (no suffix), without knowing if the writer means a 24-hour or 12-hour clock. In that situation there no certainty any way. However you should assume 10:00 in the morning is meant. If the writer thinks 24, this is always right, if the writer thinks 12, you have 50% chance of getting it right. What is more unambinguous? −Woodstone 18:15, 5 July 2007 (UTC)

I'll tell you what is less ambiguous: 10:00 a.m. or 10:00 p.m. -- the case you completely neglected in your response! I may not have phrased my argument quite right. The point is that with the 12-hour system it is possible to be unambiguous (in a multi-system world) by specifying whether the time is a.m. or p.m. This is not possible in the 24-hour system unless you add in parentheses: "This clock goes by the 24-hour system". (OTOH, times with leading zeros in the hours column would be a tipoff that the 24-hour system was most likely in use, but this is not 100%.) While this is an advantage of the 12-hour system, it is not the fault of the 24-hour system per se. If it were a choice between 24-hour system and 12-hour system from the start, and the world chose the 24-hour system and everyone used that, then this would be a moot point. But when both systems are in use, only in the 12-hour system can you succinctly make it clear which system in use. Now this is normally a very minor point because the context usually makes it clear which system is in use. If you're in the United States, you know the clocks go by the 12-hour system. In a 24-hour country, you'd know it was a 24-hour clock. I now bring you one case in which this minor advantage actually is important.

Let me preface this by saying that I hate the 12-hour system on computers. What could be uglier and more difficult to read than a list of times and dates for files or emails using the a.m./p.m. system? However, for this one situation I am about to describe, logic favors the a.m./p.m. system. Imagine a company whose Information Technology (IT) dept. is split between a city in a 12-hour country and another city in a 24-hour country. Further, this company uses tracking software to track IT changes and the form requires the user to enter a start date/time and a finish date/time for his or her change. If one had chosen to implement the 24-hour system, you know what would happen. Since most IT changes start at 17:00 (or thereabouts), you know that many in the 12-hour country (esp. the U.S.!) would type in 500 for those changes and that would be a big problem! So in this case the logical choice is the 12-hour system. Not a perfect choice, but overall a better choice for the situation (well, if my supposition is correct).

[Oh, and yes, times from 13:00 through 24:00 would be a 100% guaranteed tip-off that the 24-hour system was in use.]

--Alan E. Feldman 2007-07-05 23:47 UTC

If you are working with locations in both 24 and 12 hour oriented countries, it is likely that there would be time zone differences. So to be unambiguous, in that case there is not much other choice than to go for the ISO-8601 system with date/time/zone indication (and implicitly a 24 hour system). Seeing a date like 2007-07-06 makes interpretation of a following 10:00 automatically interpreted as not meaning 22:00. Also it avoids ambiguity whether it means July 6 or June 7. −Woodstone 07:36, 6 July 2007 (UTC)

Well, fine, but many users in the 12-hour country are still likely to put down 5:00 when they should be putting down 17:00. The time zone thing is irrelevant and I skipped it for brevity. I mentioned the yyyy-mm-dd format as my preference -- not as what might be used in the issue-tracking program. Even so, it still again doesn't affect my point. Users will still be likely to put down 5:00 when they should put down 17:00. Time zones and date formats will not help these users avoid this mistake!!!

--Alan E. Feldman 2007-07-07 00:24 UTC.

Alan, you most likely do not live in a 24-hour oriented country. People do not make that kind of mistake in writing down times. They may say five o'clock, but write only 17:00. Definitely in any professional environment, where logging is required, your fear of mistakes is ungrounded. Furthermore, especially for logging, the ISO notation is eminently suitable, because it is so easy to see the sequence and intervals. −Woodstone 10:56, 7 July 2007 (UTC)

I live in the U.S. and I guarantee that many people here would make this mistake. But, I agree with you that people who live in "24-hour countries" would be unlikely to make this mistake. Remember: my example had one city in a 12-hour country and the other in a 24-hour country. In any case, it was a minor "advantage" because it applies only in a narrowly defined scenario. I myself strongly prefer the 24-hour system for the system clocks in computers. For many other uses I can go with either system. But it is certainly true that the presence of a.m. or p.m. in the time pretty much guarantees it to be in the 12-hour system and that there is no commonly used way to do the same in the 24-hour system as succinctly. --Alan E. Feldman 2007-07-07 20:31 UTC

24-hour countries do use 12-hour clocks, which I confirmed on a trip to Europe. In a small Swiss town I visited a clock shop where virtually all clocks were sold in the 12-hour form, despite the fact that the same shop indicated its hours of operation on its door in the 24-hour form.

Please sign your name in the Wikipedia form by writing four tildes ~~~~. Wikipedia will automatically replace them with your username plus the date and time. You can get the username of your choice by clicking "Sign in / create account" at the upper right of any Wikipedia page. You now appear as an anonymous user with the IP address 71.127.xxx.yyy in the page history. — Joe Kress 03:49, 9 July 2007 (UTC)

Fix Citation needed

I regularly look at articles that are in need of citation and find links were I can give it the appropriate citation , thereby making wikipedia more encyclopedic , this article has proved problematic it only has one {{fact}} tag and when trying to check it out I can get to ISO 8601 and look at it, however ANSI standards seem to be for purchase and I cannot access the relevant info to allow me to remove the citation needed tag, if anyone out there can help out or maybe even look at rewording the article it seems like a simple job to remove this tag and yet without relevant links info etc I cannot do it. So for all those sharp minds out there I hope one or more of you can get on the case thanks --Matt 08:08, 17 January 2007 (UTC)

The whole article needs citations so I tagged it. Lots of "Many people" and "common practice" and sentences like "Children learn only the 12-hour analog dial at school." As for access to non-free ANSI standards, can't help you there. Gront 16:16, 24 March 2007 (UTC)

The standard can be seen on-line at omg.org. I do not know how legal that is. −Woodstone 17:04, 25 March 2007 (UTC)

11:59 am = midnight???

Traffic regulations in San Francisco use 11:59 am and 11:59 pm, for midnight and noon respectively.

This statement appears to be incorrect (at best, midnight and noon are reversed), and since I don't know where to check the facts I have just deleted it. Ray Spalding 22:11, 1 Jan 2005 (UTC)

San Francisco isn't the only place that does this. According to NIST, train and plane schedules in the US do a similar thing. (Only correctly, of course.) -- Toby Bartels 00:12, 4 Jan 2005 (UTC)

A perhaps more accurate description than "11:59 am = midnight in San Francisco and elsewhere" would be to say that these places never schedule any event exactly at midnight or noon, because they lack a convenient notation for these two points in time. Markus Kuhn 15:52, 1 January 2007 (UTC)

I looked up the SF traffic code, the only reference to 11:59 is midnight, which is covered in the article and not unique to san francisco. No mention of any 11:59 of either kind for noon (well, if the quote were correct, this 11:59 pm could conceivably mean noon, except that'd be a funny time for new meter regulations to take effect.) --Random832(t c) 13:42, 31 January 2007 (UTC)

Reason for 12-hour increments

Why did the Sumerians or Egyptians use 12 hour increments? Did this concept of "12" derive from the number of notes in the chromatic musical scale? Or perhaps it has some religious significance relating to astrology? -- 216.102.9.150.

The article that I linked above is unclear; but I don't think that the chromatic scale is anywhere near that old. -- Toby Bartels 22:34, 5 Jan 2005 (UTC)

The Sumerians had a base-60 numbering system. So that explains why there are 60 seconds in an hour and 60 minutes in an hour.. Now look at degrees, notice how 40°12'13" is 40 degrees, 12 minutes, 13 seconds... There are 360 degrees in a full orbit, or 60₆₀. But having 180 hours in the day (30₆₀), well that's excessive. Let's try dividing by 15, we get 12... Why 15? well 60 can be factored these primes 2,3,5 any multiple of them makes a nice simple divisor, so 2,3,5,6,10,12,15,18,30 are your basic options, of those the only ones that make sense here are 18, 15, 12, 10. Why they chose 15 not 12 or 18 I can't tell you, maybe they just thought 15 or 18 hours were still excessive, and 10 too few? Maybe 12 was better because it could be divided twice in half? --Zenyu 21:55, Jan 5, 2005 (UTC)

The used base 60 system not a 180. 12 is 60/5 not 15. 68.239.208.239 19:58, 7 May 2006 (UTC)

I'm new to wikipedia, so I don't know if this edit is appropriate, so forgive me if it's not. I would like to know why the 12 hour increments instead of 10. If the reason is that the Sumerians had a base 60 numbering system, thats great, but raises the question why a base 60 and not 10? Surely with 10 fingers its a more obvious choice? —The preceding unsigned comment was added by 195.248.96.13 (talk • contribs).

12 hours for daytime and 12 hours for nightime come from the Eqyptians who divided the night and day in that manner. The reason for a base 60 numbering system in Sumeria/Babylonia is unknown, but it was actually a base 6×10 numbering system, so there was a kind of base 10 system within it (see Babylonian numerals). Sumeria/Babylonia never used a 24-hour day or a 12+12 hour day. Instead their day had 12 double hours (each called a beru), similar to the 12 double hours in China. But the basic time unit in Babylonia was not the double hour — it was the time-degree (called an us), the time it takes for the celestial sphere to rotate one degree (1/360th of a rotation) around a motionless Earth (4 minutes), so a beru had 30 us. — Joe Kress 04:02, 13 December 2006 (UTC)

Meaning of 12:00 am/pm

Noon and Midnight are neither AM nor PM. The designations are Latin for the sun being either before the Meridian or After. At noon and midnint the sun is, by at the Meridian. Thus, 12:00 AM and 12:00 PM are both incorrect, but there seems to be an emerging convention for the convenience of computer users.

Noon is 12:00 PM, and Midnight is 12:00 AM. This is subject to debate? Theanthrope 17:00 6 Jun 2003 (UTC)

That's my thought exactly. Although it is rather odd that AM and PM change at 12 instead of 1, I've never ever run into a situation where the two were confused. Still, I tend to abbreviate them as 12m instead of 12a, and 12n instead of 12p, just for accuracy. --rj 5:46 12 Aug 2003 (UTC)

A digital clock will display "12:00 AM" for midnight, and "12:00 PM" for noon. This is probably because it makes the engineer's job easier, as "12:00 AM" (midnight) is one minute before "12:01 AM", and "12:00 PM" (noon) is one minute before "12:01 PM". Any other convention would require more work for the engineer making the clock. --Juuitchan

Apologies for dredging up an old issue here, but I wonder if this is a European vs North American difference? When growing up (in Britain) I was taught that noon and midnight were neither AM or PM, and that to write 12AM or 12PM was ambiguous nonsense. Only when I moved across the Atlantic did I realise they were commonplace. In fact, my first encounter with them was in a small town in upstate NY, where a roadside sign read "No Parking 12am to 12pm" - I couldn't interpret it, and had to go and park elsewhere! Does anyone know different? If not, I'll put a remark to this effect in the article. Cambyses 16:48, 27 May 2004 (UTC)

It would be nice to have in the article a photo of such a "No Parking 12am to 12pm" sign! Markus Kuhn 17:44, 8 April 2006 (UTC)

But the US standards body NIST agrees with the allegedly European convention: noon is neither AM nor PM, while midnight is both. I suspect that the real disconnect is official standard vs common practice, rather than Europe vs North America -- but the common practice is more common in North America. I will explain the NIST standard in the article. -- Toby Bartels 00:12, 4 Jan 2005 (UTC)

Noon is not 12:00 PM. "Noon" is an exact reference point in time - the meridiem itself. 12:00:00 PM is the one-second interval immediately following the meridiem; 12:00 PM is the one-minute interval immediately following the meridiem. Digital clocks indicate the current time interval; they (correctly) read 12:00 PM for the one-minute interval immediately following noon. At noon, no matter how finely measured, a (12 hour) digital clock should be busy changing its display from "11:59 AM" to "12:00 PM." The confusion over "noon", "midnight", "12:00 AM," and "12:00 PM" vanishes when the convention of differentiation between a point in time and a time interval is adopted. -- Scott Wagner 19:17, 4 Jan 2005 (UTC)

What do you mean "noon is ... the meridiem itself"? You must have meant that noon is when the sun crosses the meridiem. But that is incorrect because of the equation of time (true sun vs. mean sun), time zones, and daylight saving time. See my entry below for more detail. --Alan E. Feldman 2005-12-04 18:32 UTC.

I don't think that it's as simple as saying that "12:00" means an interval, not an instant. The same notation is used for both. Nevertheless, you're quite right that when a digital clock says "12:00", then it means an interval. So I guess that there is no conflict between the time display and the NIST standard that noon (an instant) is not PM. However, when you set your alarm for noon, then I claim that there is a conflict! -- Toby Bartels 22:34, 5 Jan 2005 (UTC)

The first two paragraphs of section "Ambiguity at noon and midnight" are incorrect. I quote them below and comment:

[begin quote] According to the actual meaning of the terms ante meridiem ( a.m.) and post meridiem ( p.m.), as well as standards bodies such as the National Institute of Standards and Technology in the United States, noon (which falls precisely at the meridiem or celestial meridian) is neither a.m. nor p.m., because noon is neither before nor after itself. [end quote]

What does "noon which falls precisely at the meridiem or celestial meridian" mean? Noon doesn't fall anywhere. OK, this was really meant to say that the sun is on the celestial meridian at noon, but that too is incorrect! First of all there is local sun time vs. standard time. The sun *does* cross the celestial meridian at 12:00 local solar time, but we don't go by local solar time -- we go by standard time. The actual position of the sun in the sky determines local solar time. But standard time differs from local solar time in two ways: 1.) Local solar time goes by the position of the true sun while standard time goes by the mean sun (which gives us mean solar time) as determined by atomic clocks. 2.) Local solar time varies with longitude while standard time varies with time zone.

The difference bewteen local solar time and mean solar time is caused by the non-uniform "motion" of the sun acorss the sky. This non-uniform "motion" is caused by the tilt of the earth's axis with respect to its orbit and by the fact that the speed of the earth in its orbit around the sun is not constant but varies according to Kepler's Second Law. The difference between local solar time and mean solar time is given by the equation of time. The result is that the sun is as much as about 15 min. slow or fast compared to mean solar time, depending on the time of year.

So, while the sun *does* cross the celestial meridian at 12:00 local solar time, it almost never does so at 12:00 standard time.

Here's an example: Assume it's late October and you live at 75 deg. longitude. The equation of time then gives the sun as being approx. 16 minutes fast. So, using the strict interpretation of the original definitions of AM and PM and the article's definition of "noon" we would get the following times:

  11:42 am
  11:43 am
  11:44 noon (= 12:00 local solar time which is when the sun crosses the celestial meridan)
  11:45 pm
  11:46 pm
  ...
  11:59 pm
  12:00 pm
  12:01 pm
  12:02 pm

And the crossover point -- "noon" according to local solar time -- would vary throughout the year! The crossover point would also change by 4 min. for each degree of longitude the observer is from his time zone's central meridian. This shows the fallacy of slavish adherence to the old, obsolete, archaic definitions of AM and PM. Current usage is clearly not according to the original definitions! This is because the original definitions refer to local solar time while in the modern era we instead go by standard time. (And I haven't even touched on daylight time!)

[beg quote] Despite this definitive logic... [end quote]

Well, I have shown about that this is not "definitive logic". The original definitons for AM and PM came into being in the 1600's or so (reference forthcoming) when people went by local solar time. It made perfect sense for that. But obviously we have come a long way when it comes to timekeeping. We have smoothed out the non-uniform apparent motion of the sun across the sky using atomic clocks. We have set up time zones to simplify (civil) time's dependence on longitude and daylight saving time to save daylight. The result is that the sun almost never crosses the celestial meridian at noon (civil time) and therefore it is high time to replace the archaic definitions with the modern de facto convention, namely, that midnight (00:00) is 12:00 AM and noon (12:00) is 12:00 PM in the 12-hour AM/PM system. This convention is so simple, so logical, so natural, so sensible, and so mathematical that it practically forces itself upon us. And in fact it aleady has. Note that digital clocks in practically all devices use this convention which has made it the de facto convention. (This is assuming that the clock uses am and pm to distinguish between noon and midnight. I have seen 12:00 M for midnight and 12:00 N for noon on Cable TV on-screen TV schedules. Not that there's anything wrong with that!)

It is the best and the de facto convention because it allows us to treat the am/pm value as just another column in the date-time construct (which, of course, simplifies programming). By considering pm to be greater in value than am (you could do this by using am=0 and pm=1) you simply use the normal rollover rules. Yes, am/pm times are typically written with am/pm out of order, but we already have the components of the date in different orders that vary by country yet they still follow normal rollover rules. In natural order we get

   yyyy-mmm-dd am/pm hh:mm:ss.cc

It's very simple if you also assume 12 in the 'hh' column to be equivalent to zero. When hh=11 rolls over to 12 (effectively 0), the am/pm value is incremented. When pm (1) rolls over to am (0), the day is incremented. Etc.

So since the original definitions of am and pm apply to local solar time, and since we live by civil time to which the de facto convention outlined above applies, the "ambiguity section" in the article needs to be rewritten. And while the 24-hour format avoids this problem altogether, if one is goint to use am/pm, this is the way to do it.

I acknowledge that NIST and other similar bodies insist that noon and midnight are neither am nor pm, but I believe they should rethink this issue and their position only continues to cloud the issue. They don't even mention the de facto convention which is needed for people who look up this FAQ question to determine how to use and interpret 12:00 AM and 12:00 PM (which is needed for using digital alarm clocks, VCR's, personal computers, etc.) They could at least do that!

They could say that while technically, according to the original definitions of ante meridiem and post meridiem, noon and midnight don't fall into either class, the modern world runs on the new de facto convention (12:00AM = midnight, 12:00PM = noon). They could even explain the difference between local solar time and civil time and such. And they could still mention the use of "tricks" like 12:01, etc., to avoid possible confusion.

If all this *still* doesn't convince you, consider AD and BC. Imagine the same slavish adherence to the original definitions of these terms. The calendar as we know it would completely fall apart. Why? Because we don't know for sure just what year Christ was born. Scholars almost universally agree that Christ was born during the BC period, probably between 7 B.C. and 2 B.C. So to be strictly correct, we'd have to re-number all our years! But it's even worse. Since we don't know the exact year, we wouldn't know HOW to re-number the years. And we probably wouldn't re-number the years *even* if we *did* know: Imagine the confusion that would result!!! Therefore, we have done the sensible thing: We use the de facto definitions of AD and BC which are such as to simply keep the current numbering of years unchanged. --Alan E. Feldman 2005-12-04 20:03 UTC. (Minor correction at 2005-12-04 22:17 UTC)

The long story above just shows that am and pm have no hard relationship to the apparent position of the sun. This is not only true for noon and midnight, but depending on time zone and season the difference between solar time and clock time can easily deviate for more than an hour. During daylight saving time 12:30 pm is almost surely before solar noon (before the sun passes the meridian). So the discussion on the precise meaning of 12:00 am/pm has nothing to do with the position of the sun. It looks to me like the theoretical meaning can only be wether the time is before or after noon. So the discussion in the article stating that 12:00 is neither am nor pm is wholly correct. Of course this does not deny that the most logical extension of the meaning is for 12:00 pm to be shortly before 12:01 pm, as is the usual designation on digital clocks. But even this is only marginally logical since 11:59 pm is then not shortly before 12:00 pm. −Woodstone 19:24, 4 December 2005 (UTC)

The article is not wholly correct:

[begin quote] According to the actual meaning of the terms ante meridiem ( a.m.) and post meridiem ( p.m.), as well as standards bodies such as the National Institute of Standards and Technology in the United States, noon (which falls precisely at the meridiem or celestial meridian) is neither a.m. nor p.m., because noon is neither before nor after itself. [end quote]

Contrary to the above quote from the article, noon (civil time) does not fall precisely at the meridiem. Noon is a time, not something that can be on the meridiem. Even with the obvious correction, "noon is when the sun 'falls' on the meridiem", it is still wrong as I demonstrated above. Therefore, the article is not wholly correct. If anyone needs to know how noon and midnight are classified w.r.t. am and pm, they are almost certainly asking about civil time for which there is a clear, sensible, logical, and mathematical de facto convention. The original definitions, which apply only to local solar time (which almost no one uses today) are irrelevant in our modern world. Given the am/pm system as it is, the de facto convention is correct for civil timekeeping in the modern world. Denying noon and midnight am/pm assignments does no good and in fact causes further confusion; it is unfortunate that NIST and other time authorities refuse to recognize this. And finally, if you interpret 12 as zero, 11:59 pm being followed by 12:00 am is totally logical. It's certainly no less logical than January, February, ....

--Alan E. Feldman 2005-12-04 22:17 UTC.

Oops. Meridiem means middle of the day. So noon falling on that makes sense. But the article equates that with celstial meridian, so it's still wrong! And now we get into what "middle of day" means. OK, if middle of day means 12:00 civil time, then yes, strict application of the old definitions gives noon and midnight no am or pm designation. But the de facto convention still exists and is the most useful answer to the question. If you discover a file on your computer and it is labeled 12:00 am, you need to know the de facto convention to know if the file's time stamp is noon or midnight. So we may as well redefine am and pm to be what they already are in the modern world. am covers the interval [00:00,12:00) and pm covers the interval [12:00 24:00) using parenthesis to indicate open ended (exclusive of the end point) and brackets to indicate closed (inclusive of the end point). And the times are given in 24-hour format. --Alan E. Feldman 2005-12-05 06:22 UTC.

Well, the new write-up is better, but it still worships the old definitions too much. The article says that "despite this strict logic" resulting in no am/pm assignments for noon and midnight, it is "common practice" to have noon as 12pm and midnight as 12am. It is more than common practice: it is the de facto convention, confusing or otherwise.

Do we stick with the original definitions (or usage) of will and shall? NO. These words have evolved. Why can't am and pm evolve? What is so terrible about a new definition? If people who wrote articles about am and pm clearly described the new de facto convention and why it came about and how logical and mathematical and sensible it is, there would be LESS confusion. It would be a reference. In English, spelling is confusing. But we don't re-spell everything phonetically to avoid confusion, do we? People have to learn the words. And if a writer is unsure of a spelling, he or she can look it up in a dictionary! The same should be for am and pm. The American Heritage Dictionary, to its great credit, says the following:

Usage Note: By definition, 12 A.M. denotes midnight, and 12 P.M. denotes noon, but there is sufficient confusion over these uses to make it advisable to use 12 noon and 12 midnight where clarity is required. (see <http://dictionary.reference.com/search?q=AM>)

That's the right way to do it!

Then there is the part talking about the "technical inaccuracy" of the de facto convention. I ask this: How do we really know that the people in the 1600's didn't happen to use the de facto convention? Does it even matter? They used local solar time. We're pretty sure about that. But in today's modern world we use UTC and standard time. Why is standard time okay but not the new de facto am/pm convention? They may or may not have assigned 12pm to noon and 12am to midnight. Why does it matter? In any instance where the time is given as 12:00 AM or 12:00 PM you need to know what it means. Back then, if someone asked you the time, they expected to get the local solar time. Do that today and you're likely to miss the plane! So "What time is it?" has evolved from local solar time to standard time (which is based on UTC, of course). And am and pm have also evolved.

In today's world, 12 AM is midnight at the beginning of the day and 12 PM is noon, period. Why do so many worry about adapting AM and PM to our modern world? Do we do this with ANYTHING else? Words evolve. Timekeeping evolves. The second used to be 1/86400th of Jan 1 in 1900 or 1940 or something like that. Now it is a certain number of oscillations of a certain atomic process. The meter has gone through SEVERAL re-definitions. If we are going to stick with c. 1600 definitions, shouldn't we go back to local solar time? I mean that's what was used when am and pm came into being. And why is no one similarly fussy about using "strict logic" to interpret AD and BC? No one says that our current usage of AD and BC is "not strictly correct" or "technically inaccurate". Moreover, application of "meridiem" is "technically inaccurate" because it assumes local solar time, which is not in general use today.

If one wants to use, or suggest to use, tricks like 12:01 and 11:59, etc. in written schedules, notices, and similar, be my guest. There's nothing wrong with that. But I think the article should be a little more accepting of the new de facto convention and give better reasons for it: namely, that in the case where one is restricted to using am and pm to distinguish noon from midnight, it is the logical choice because then the am/pm status simply becomes another column in the time construct that obeys the usual rollover rules (okay, it mentions the rollover bit, but in a different way). And it should say that it is necessary to know and use this convention with many electronic devices, including computers. (In fact, it is for that very purpose that people look up am and pm in the first place!!!) And ultimately, there is nothing wrong about 12pm being noon and 12am being midnight. That's just the way it is. You need to learn how to spell, even though spelling is confusing. Similarly, you need to learn that 12pm is noon and 12am is midnight at the beginning of the day.

If you differ with me on this, can you PLEASE answer the questions I have posed? thanks. --Alan E. Feldman 2005-12-24 06:06 UTC.

When most people hear PM they think day light becuase the PM (12:00h or 12:00PM) starts in the daylight. When most people hear AM they think dark becuase the AM (00:00h or 12:00AM) starts at night.Perhaps I am oversimplifying this old and dead issue -greataff

Not so they haer the numbers like they hear the numbers in the 24-hour clock.

The new write-up for "Amguity at noon and midnight" is much better than the previous versions I remarked on above. I'd like to add that I have been seeing increasing use of 12 PM for noon (and also one occurrence of 12 AM for midnight) over the last year or so. And I haven't seen any mistakes! It looks like we are finally beginning to break away from the original archaic defintions of am and pm -- as we should be. Alan E. Feldman 16:27, 2 July 2006 (UTC)

I am flabbergasted at this discussion in which the consensus seems to be that 12:00 a.m. is midnight and 12:00 p.m. is noon. This is exactly backwards from, for instance, the U.S. GPO Style Manual, which states:

 9.54. References to meridian in statements of time are abbreviated as follows:

 10 a.m. (not 10:00 a.m.)
 2:30 p.m.
 12 a.m. (formerly 12 m.) (noon)
 12 p.m. (midnight)

Fitzaubrey 18:47, 15 September 2006 (UTC)

Your link gives a blank page. I'll check it again on Monday (or the next time I remember, if later). Anyway, what you quote makes no sense. There's a problem with writing "10:00 a.m."?

And what sense does it make to have

11:58 am
11:59 am
12:00 am
12:01 pm

be the standard? A digital clock shows 12:00 for 60 seconds. For all but the first instant of that period it is clear that its am/pm is the opposite of the minute before it. So why should the first instant of that period be different? The sequence

11:58 am
11:59 am
12:00 pm
12:01 pm

makes far more sense. This way, the am/pm field is simply just another column (though out of place) that follows the usual rollover rules. Why would anything else be any better? Every digital clock that uses am/pm assumes that 12:00 am is midnight and 12:00 pm is noon. It is the de facto convention, and rightly so. I have seen increasing use of 12:00 pm in signs and notices and such to mean noon and never the other way around. The American Heritage dictionary says 12:00 am is midnight and 12:00 pm is noon. Alan E. Feldman 2006-Nov-24 6:30 UTC.

Somebody changed the opening sentence for the "Confusion at noon and midnight" section to "The abbreviations for the terms ante meridiem (before noon) and post meridiem (after noon) should not be applied to noon and midnight." Why? This is obviously controversial. The previous statement (last seen on Aug 1) was much better, and certainly far more neutral. Why not say that use of am and pm at noon and midnight is highly controversial or that some claim that am and pm shouldn't be used for noon and midnight because their literal translations and so on. Then the "However, ..." part would follow nicely. In fact, the way it reads now is self-contradictory. It says that am and pm should not be used for noon and midnight, and then immediately gives very good reason why they should be used for noon and midnight. Why do some people find it so important to apply a literal translation of Latin terms to time when said terms originated hundreds of years ago when people used Local Solar Time and analog clocks as opposed to the elaborate system of Standard Time with time zones, daylight saving time, digital clocks and the like that we use today? Where are these people when it comes to the terms A.D. and B.C.? Using the same logic with these terms would cause the current Gregorian calendar to completely fall apart since no one knows for certain when Christ was born. Can someone please explain why the literal translations should be used for ante meridiem and post meridiem but not for Anno Domini and Before Christ? And what horrors would ensue if we would simply all accept the de facto, and quite sensible, new definitions of am and pm? It's not like we're redefining 2 plus 3 to be minus 97! Alan E. Feldman 2006-12-15 16:55 UTC.

I fixed the noon and midnight section. There is no reason that am and pm shouldn't be used for 12:00 noon and midnight. In fact at midnight it is actually an advantage in that it obviates the ambiguity of which midnight is being referred to: at the beginning of the day or at the end. Alan E. Feldman 2006-12-22.

(I'm sorry this entry is so long, but I'm trying very hard to be clear.)

Marcus Kuhn removed my addition "(Note that, according to the de facto convention, 12:00 a.m. does unambiguously refer to the beginning of the day.)" because he says that it is ambiguous. He missed the part that said "according to the de facto convention". The de facto convention is what the article refers to as "common practice". But the article is slightly incomplete because ***according to the de facto convention***, 12:00 am is the midnight at the beginning of the day. It is equivalent to, in 24-hour format, 00:00, not 24:00. In fact, if you create a file on your computer at midnight, you will see that I am right. So the article should really say instead:

"However, it has become common practice in countries that use the system (such as the United States) to define 12:00 am as the midnight at the beginning of the day (equivalent to 00:00 in the 24-hour notation) and 12:00 pm as noon (equivalent to 12:00 in the 24-hour notation)."

So, 12:00 am is ambiguous _only_ without the de facto convention. With the de facto convention, it is 100 percent unambiguous, which is what I was saying.

OK, so because the article incorrectly omitted the fact that 12:00 am is the midnight at the beginning of the day, I understand why Marcus deleted my statement. So I think the article should be fixed so that somehow it clearly explains that 12:00 am is at the beginning of the day. (Why is it at the beginning of the day? Because then the am/pm "column" follows the normal rollover rules of ordinary numbers. Explicitly, when the minutes column goes from 59 to 00, the hour is incremented by one. When the hours column goes from 11 to 12, the am/pm column is "incremented". When the am/pm column goes from pm to am, the day is incremented by one. It is 100% mathematically logical if you simply think of 12 as 0, AM as 0, and PM as 1. That is why 12:00 am is at the beginning of the day. It is also the easiest way to implement time in a digital or computer clock and is consistent with the "applies immediately after" part of the article..)

Despite the continuing efforts of the "literalists" (those who decree that you can never under any circumstances use am or pm at 12:00), despite the fact that they insist on engraving the original archaic, obsolete definitions of ante meridiem and post meridiem in stone for all eterninty, despite the fact that said terms originated in an era when Local Solar Time was in use which is completely different from the Standard Time system we use today, despite the fact that they have no problem with Anno Domini being applied to our current calendar system even though it suffers from ambiguity for ALL years; a new definition is taking hold, whether they like it or not. First digital clocks, and then computers, combined with logic, common sense, mathematical sense, etc., have practically forced this de facto convention upon us. Not only that, I see it being used more and more in train schedules and the posting of business hours. And I've never seen 12:00 am or 12:00 pm used incorrectly in such things in recent years (I think I saw it used wrong once many years ago, but I've also seen a huge number of words spelled wrong and an even larger number of apostrophe errors). Since this de facto convention is literally taking over, I think it should be clearly and correctly explained in this article. That is what would help people most when they encounter a situation in which they need to know what 12:00 am or pm means, not how obselete archaic terms don't apply.

If sources like Wikipedia, NIST, etc., would simply state correctly what the de facto convention is, THAT would be the most helpful. The literalists are doing us no favors by brining up the problem of noon not being before or after itself without acknowledging and explaining the de facto convention. The terms ante meridiem and post meridiem originated some 400 years ago and applied to a *different* timekeeping system. It's time to move on. The second has been redefined at least once. The meter has undergone at least four redefinitions. Would the world end if we acknowledge and carefully document the de facto convention for am/pm?

Don't get me wrong. If someone wants to point out that am and pm don't literally tell us how to assign themselves to 12:00 noon and midnight, fine. If someone wants to explain tricks like 11:59pm, 12:01 am, and such (and use such tricks), fine. But don't do it instead of clearly and correctly explaining the de facto convetion for am and pm. I'm sorry to have written at such length, but I find that's what it takes for me to be clear on this point. Alan E. Feldman 2007/01/02 UTC.

Oops! I got carried away with my "despite..."'s and screwed up. I meant that the literalists "insist on engraving the original archaic, obsolete definitions of ante meridiem and post meridiem in stone for all eterninty despite the fact that said terms originated in an era when Local Solar Time was in use which is completely different from the Standard Time system we use today" and "despite the fact that they have no problem with Anno Domini being applied to our current calendar system even though it suffers from ambiguity for ALL years"; not that the de facto convention is taking over because of these "despite..."'s. Now, to clarify the de facto convention: 12:00 am is 00:00 (the beginning of the day) in the 24-hour format and 12:00 pm is 12:00 in the 24-hour format (noon). There is no ambiguity in the definition; there is only ambiguity because some people are unaware of this convention. The sooner we all acknowledge it, the sooner the confusion will end. Alan E. Feldman 2007/01/14

It is unfortunate the the Chicago Manual of Style and the govt. publication cited in the article don't go with the de facto convention. They need to get with the times. Every computer using the 12-hour am/pm format uses the de facto convention as far as I know. And since computers are so dominant today, why do anything differently? If you see a file on your computer dated January 14, 2007 12:00 am, that means it was the beginning of January 14. If you see a file on your computer dated at 12:00 pm, you know that means noon. Since it would be highly impractical to reprogram all 12-hour computers and rewrite the time stamps of all 12:00 files, why on earth would it not be best to always use the de facto convention (given that one is already using the 12-hour am/pm clock, of course)?

BTW, the link to the gov't site (ref. 3) goes to a blank page and the link to the Chicago Manual of Style goes to a page that says nothing about 12:00. It does say not to use a.m. or p.m. with "o'clock" but to use morning, afternoon, evening, or night instead (8:00 a.m., but eight o'clock in the morning). Well, that won't work in the land of the midnight sun! (OK, the midnight-sun remark is a nitpick.) Alan E. Feldman 2007/01/14

Sorry, Alan E. Feldman, but you are too repetitious, so I didn't read all you wrote here. Nevertheless, perhaps we can agree that the authorities do not agree. Therefore there is no clear convention. Therefore, you are asking Wikipedia to take sides in a dispute instead of merely reporting the dispute. Our Wikipedia style book recommends the "literalist" approach because it is the only approach which causes no confusion and requires no intensive training to use. I agree. All your attempted logic about digital clocks overlooks the obvious: a At noon you see the clock change from am to pm. Noon is the instant of that change, and so is neither am nor pm. Just because the clockmaker was too lazy to make it say noon and midnight doesn't mean I have to pretend that noon is after noon. That will always be confusing to each generation of children growing up, so why not stick with the Latin meanings, which are not ambiguous if used properly: "midnight at the end of June 3", etc. Korky Day 06:57, 16 March 2007 (UTC)

While there is no clear "official convention", there is a clear de facto convention. Every 12-hour am/pm digital clock uses the de facto convention I have described. The world runs on this convention. If you see a file on your computer whose timestamp is given as 12:00 AM or PM, if you get an email whose timestamp is 12:00 AM or PM, if you see these times in computer output, be it a printout or electronic display, you need to know the de facto convention. I am not asking Wikipedia to take sides on this. I am simply reporting what the de facto convention is. If the authorities said that clocks turn counter-clockwise and by simple observation you note that they all turn clockwise, would you be taking sides by reporting that clocks actually turn clockwise? If you think the de facto convention is "incorrect", so be it. But then you should also complain that the de facto convention for AD and BC is also "incorrect" by very similar reasoning (search for AD and BC in my previous posts).

The reason not to stick with the Latin meanings is because they are obsolete and doing so only exacerbates the confusion. People need to know the de facto convention because in the digital age because that is what is used. I have seen many uses of 12:00 a.m. and p.m. in business-hours signs and all were correct according to the de facto convention. I've also seen numerous uses in computer output at work and on the Web -- all correct. I use JIRA at work which uses the de facto convention and I need to know what that convention is. So there should be a reference. Another reason is that we don't do this with anything else. Why not stick with Local Solar Time, which is what was used when the terms a.m. and p.m. came into being? Why not stick with the original definition of the meter? Look at the definition for noon at www.m-w.com and you'll see that it used to mean midnight! There was a change in the defintion of noon! So why not a change in the defintions of a.m. and p.m., which has already happened anyway? AD and BC are just as "incorrect" as assigning 12:00 a.m. to midnight at the beginning of the day and 12:00 p.m. to noon, but no one complains about that! Your argument about changing from a.m. to p.m. at noon assumes the original, archaic defintions. They are obsolete and it's high time the authorities recognized this. So you would have digital clocks show "12:00 noon" for the entire 60-second interval from 12:00:00 to 12:01:00? Are we to retrofit all existing 12-hour clocks to be in accord with this obsolete convention? Are you going to pay for this retrofitting? Children would not be confused if they were simply taught the de facto convention. You cannot say "midnight at the end of June 3" on a digital clock. The existence of digital clocks means that we must assign a.m. and p.m. values to noon and midnight, and the de facto convention has done that in the most sensible, logical, mathematical and economical way. And finally, I ask you (or should it be "thee"?), why should the original archaic defintions of a.m. and p.m. be etched in stone for all eternity despite their immense impracticality in the digital world whereas all other words and terms are allowed to evolve?

--Alan E. Feldman 2007-03-18 15:31 UTC.

Oh, just one more thing to avoid a huge misunderstanding: I have no objection to the use of things like "midnight at the end of June 3" where practical. I also have no objection to usage of 12:01 AM and 11:59 PM in signs and schedules and such. I only object to ignoring the reality of the de facto convention for digital 12-hour clocks, claiming that it is somehow "wrong", not including and explaining the de facto convention in references, and insisting that the original archaic defintions of a.m. and p.m. be etched in stone for all eternity, despite their immense impracticality. I object because these things only exacerbate the confusion.

--Alan E. Feldman 2007-03-18 16:08 UTC

To Korky Day: According to your logic: At noon you see the hour change from 11 to 12. Noon is the instant of that change, and so is neither 11 nor 12. (Okay, a slight cheat, but very slight.) The insistence on sticking with the original, obsolete definitions (which are highly impractical on digital clocks) as opposed to the nearly ubiquitous de facto convention is the cause of all this confusion. You are apparently upset that the de facto convention is not self-evident or self-defining. So what? Very few things are. Teaching kids only the original Latin meanings is what will confuse them because they will encounter 12:00 AM and 12:00 PM in numerous computer displays and output (and other digital clocks) and they won't know what these times mean unless you teach them the de facto convention.

-- Alan E. Feldman 2007-03-25 15:27 UTC.

I, Korky Day, now take 20 short paragraphs to anwer Alan E. Feldman and others. Scott Wagner said it so well and concisely near the top (19:17, 4 Jan 2005 = 2005 Jan 4 19:17), and no one has beaten his logic.

Alan Feldman seems to think that Scott and I are illogical because we won't adopt his favoured convention of 12 am and 12 pm. We are not illogical. I agree to state in the article that there is a strong trend among most computer and digital clock interpreters to assign midnight and noon to the morning and afternoon, respectively. However, other conventions and interpretations are different and the article should be even-handed. If we want to fully inform the readers, we will tell them all the important conventions and interpretations, not favour Alan's.

Who knows what the trends will be a decade from now? It is not our role to campaign for our preference in an encyclopedia text. Because Alan seems to be so immersed in computers and digital clocks he might think that the whole world shares his view. This is an on-line encyclopedia, but that doesn't mean the readers necessarily want everything slanted towards the current cyber fads. Maybe the newer digital clocks will say noon and midnight (or N and M, as Alan says he saw on telly). That would be better, I think, but I'm not saying that in the article because it's just my preference.

Alan's sun vs. standard time argument fails because standard time doesn't pretend an internally ILLOGICAL thing: that noon is before or after noon. It merely defines noon differently. It says to consider the whole time zone, east to west, as the same time as if it were geographically at the centre of that time zone (approximately), where sun and standard time are the same. The former "pretending" (convention) is confusing and illogical, the latter "pretending" is easy and logical.

Also, Alan's digital clock apparently omits the required space between the numbers and the am/pm, so he writes 12:00AM instead of 12:00 AM, because he's a slave to his (current!) gadgets and can't think for himself (I'm just trying to rib him in a gentle way!). Also, he writes the :00, even though it's superfluous, just because that's how he interprets his digital clock's demands. If he could read a clock with hands, perhaps he would realize that. (More teasing.)

Similarly, Alan's BC-AD argument breaks apart. When BC and AD were coined, the Latin-speaking monk who did it (if I remember right) was trying to base it on Christ's birth. It's now long past the practical time to argue about when to start counting the years in that system. A mistake was probably made, we know it, we accept it, it doesn't cause any real day-to-day problems, so we don't want to change the numbering now.

That's nothing like today's am-pm debate. The latter hasn't been going on for centuries. There is no convention that's been accepted by enough people and organisations to make it a done deal. And rejecting 12 am as noon wouldn't cause any problems, not like re-numbering all the years in all the books. Rather, rejecting 12 am and 12 pm would end the confusion. It would mean that people would just have to interpret their old digital clocks differently (or remember their illogical convention). I always have interpreted them correctly: when the 11 changes to 12 and the am changes to pm, that's the instant of noon. Noon isn't the interval after noon (pm) any more than it's the interval before noon (am).

Instants and intervals are important mathematical concepts to teach in schools! Alan, you talk about them in "Alan E. Feldman 2005-12-05 06:22" above.

I presume, Alan, that you understand that a point in space has no length, width, or height. A horizontal line (pointing ahead of and behind you) has length, but no width or height. A horizontal plane has length and width, but no height.

Similarly, why not accept that noon and midnight are instants, not intervals, as Scott Wagner wrote? The same with any time: 5:23 pm is an instant, not the interval between 5:23 and 5:24, even though your sacred digital clock (according to your interpretation) insists that it is that entire interval.

I quote Alan some more:

"So you would have digital clocks show "12:00 noon" for the entire 60-second interval from 12:00:00 to 12:01:00? Are we to retrofit all existing 12-hour clocks to be in accord with this obsolete convention? Are you going to pay for this retrofitting? Children would not be confused if they were simply taught the de facto convention. You cannot say "midnight at the end of June 3" on a digital clock. --Alan E. Feldman 2007-03-18 15:31."

First answer: yes, show "noon" for 60 seconds if the clock doesn't show seconds. If you show it instantly, it will be invisible. People will have to learn the convention that it means that it was noon the instant they first saw "noon", not the whole time they see "noon". (If they don't understand that, no harm is really done, unlike if they don't understand your favoured convention.) That's much more logical and easy for children than your favoured convention.

Retrofit? No. Your wonderful modern gadgets don't last long enough to worry about. I'm wearing my father's watch with hands which still work and are unambiguous 43 years after he died.

Yes, you can have dates on a digital clock. Better throw out yours and get one which says whether the midnight is the end or the beginning of the date. If they don't make them yet, they will. (Friendly sarcasm. I'm a green who hates waste.)

Then Alan writes, "According to your logic: At noon you see the hour change from 11 to 12. Noon is the instant of that change, and so is neither 11 nor 12. (Okay, a slight cheat, but very slight.)" (Alan E. Feldman 2007-03-25 15:27.)

Wrong, Alan. It's like I said above: Noon is the instant the 12 first appears.

Alan, you argue passionately that we should adapt to our machines and their alleged logic (based on the imperfect human computer programmers) rather than making our machines adapt to us.

Maybe so, but Wikipedia is not a activist organisation. And, anyway, I'm an activist on the opposite side. What you who cannot accept the concept of "instant" should do, if you really want to "win", is to invent new terms. Maybe N+ to mean noon plus the interval from noon to midnight. Maybe M+ to mean midnight plus the interval from midnight to noon. So at noon your new, advanced (still holy?) digital clock would roll over to 12 N+.

If people can re-name BC and AD (they now call them BCE and CE), then why not? Alan said, "The terms ante meridiem and post meridiem originated some 400 years ago and applied to a *different* timekeeping system. It's time to move on." (2007/01/02).

So go ahead and "move on", Alan--and be sure to leave the old, no-longer-meaningful terms AM and PM behind, too. And when you get support for new terms in the "real world", you can add them to Wikipedia. (If you adopt my coinage, give me credit for inventing N+ and M+ on 2007 April 2 12h09.) -- The 20 paragraphs above are by Korky Day 02:41, 3 April 2007 (UTC)

Be ye informed that there is a reason to write(or type) 12:00Am. This is that 12Am means the whole hour, while 12:0Am means ten minutes and 12:00Am means one minute. Zginder 22:06, 3 April 2007 (UTC)

You invented that just now, did you, Zginder? I think I prefer the old way, "midnight to 1 am" means the whole hour. Or 0:00/1h Korky Day 23:21, 5 April 2007 (UTC)

I was talking about accuracy and significant figures. e.g. 12:00 a.m. means from 11:59:30 p.m. to 12:01:30 a.m. which is two minutes; whether it is one of two minutes can be discussed another time. Zginder 12:47, 9 May 2007 (UTC)

As digital clocks are intended for everyday use by ordinary people, it would be impolite, if not downright rude, not to consult them on how these devices should operate. Therefore, I urge all programmers of digital clocks to ask some ordinary people two sets of three questions:

"What time is 10 pm?" (If the answer is not ten o'clock in the evening / at night, move on.) "And what time is 11 pm?" "And what time is 12 pm?

"What time is 10 am?" "And what time is 11 am?" "And what time is 12 am?"

The second set could be asked before the first, if desired. Armed with the answers, you should now be in position to programme digital clocks the way ordinary people expect them to be. MansLaughter 17:56, 11 May 2007 (UTC)

Response to Manslaughter: This is called leading the witness. One could easily lead the witness the other way. If you would think through how a digital clock works, especially a mechanical one, you would understand why it's done the way it is. It makes sense the way it is. I don't see what sense it makes to have 12 AM immediately followed by 12:xx:xx.xx... PM. Shouldn't all the times beginning with 12 (in the same hour) be the same AM or PM? Besides, it's a little late to change it now. And I can't remember the last time anyone asked me how anything should be built. --Alan E. Feldman 2007-05-14 02:15 UTC.

Response to Korky: Well, if you can't be bothered to read all that I wrote, it is not fair to criticize it, no?

I was trying to make these points:

1.) In the 12-hour world, unless you're in Amish country or something like that, you will find that digital clocks are ubiquitous and dominant. They all work the same way w.r.t. 12:00. So I think it's important to know what that way is and it should be documented to set a standard, just like we have standards for length, area, weight, etc.

2.) It makes sense the way AM-PM digital clocks work.

3.) It is okay to re-define AM and PM to accommodate digital clocks. We've redefined numerous other means of measure. Why not this?

I claim my points remain valid, your 20 paragraphs notwithstanding. I'll respond to them later when I have more time. And I do understand intervals and instants. I didn't think it was necessary to go into such detail, but I will. --Alan E. Feldman 2007-05-14 02:16 UTC

I like the article the way it is now. Though I doubt that many people will come across many Gov't papers mentioning 12:00 very often. And the 1941 Chemistry and Physics handbook? How many people have a copy of that, much less use it to determine AM and PM for noon and midnight? --Alan E. Feldman 2007-05-14 02:16 UTC

I, Alan E. Feldman, hereby respond to Korky's "20 short paragraphs":

I, Korky Day, now take 20 short paragraphs to anwer Alan E. Feldman and others. Scott Wagner said it so well and concisely near the top (19:17, 4 Jan 2005 = 2005 Jan 4 19:17), and no one has beaten his logic.

His logic was beaten by Toby Bartels only two responses down (and by my own writings, of course).

Alan Feldman seems to think that Scott and I are illogical because we won't adopt his favoured convention of 12 am and 12 pm. We are not illogical. I agree to state in the article that there is a strong trend among most computer and digital clock interpreters to assign midnight and noon to the morning and afternoon, respectively. However, other conventions and interpretations are different and the article should be even-handed. If we want to fully inform the readers, we will tell them all the important conventions and interpretations, not favour Alan's.

It's more than a strong trend. It is "always". It is not most -- it is almost certainly all. Even-handed? The terms "12 noon", "noon", "12 midnight", and "midnight" are encompassed by the de facto convention. The only change the de facto convention makes from the original convention is that 12 a.m. and 12 p.m., instead of being forbidden, are used to designate 12 midnight at the beginning of the day and 12 noon, respectively. There's nothing un-even-handed about it. And the de facto convention is very important. I like the last May 31 edition of the article, and it does not favor my convention.

Who knows what the trends will be a decade from now? It is not our role to campaign for our preference in an encyclopedia text. Because Alan seems to be so immersed in computers and digital clocks he might think that the whole world shares his view. This is an on-line encyclopedia, but that doesn't mean the readers necessarily want everything slanted towards the current cyber fads. Maybe the newer digital clocks will say noon and midnight (or N and M, as Alan says he saw on telly). That would be better, I think, but I'm not saying that in the article because it's just my preference.

I didn't know Wikipedia was in the prognostication business. A decade from now? What does it matter? How does that apply to this article and not all others? The entire a.m./p.m. world uses the de facto convention. It is certainly not a "fad". It's always been this way as long as digital clocks have been around. Fads don't last for decades. When newer digital clocks behave differently, Wikipedia can be updated to reflect that.

Re campaigning: I do not campaign in the article. On this discussion page I am responding to people who post material like Korky's. My "campaigning" is limited to trying to explain to people like him the following essential points:

1.) The a.m./p.m. world goes by the de facto convention. AFAICT, all a.m./p.m. digital clocks use this convention. This includes those in computers which means that software and the Web also use this convention. Most recently, even printed matter such as timetables and business-hours signs now use the de facto convention. It is therefore pervasive and very important. It is in fact the de facto convention.

2.) This involves a slight addition to the old convention, namely, the use of 12 a.m. for midnight at the beginning of the day and 12 p.m. for noon. All else remains the same. The terms "noon" and "midnight" are not forbidden.

3.) The de facto convention is logical, elegant, sensible, mathematical, economical, "green", and practical.

4.) The result is a slight change: at noon and midnight we do not worry about the literal meanings of ante meridiem and post meridiem. I argue that this change is not only acceptable, but also desirable for the following reasons:

A.) No other terms in measurement are immutable for all time. The unit of time called the second has been redefined at least twice. The meter has been redefined four times. And noon itself has been redefined at least once. The terms BC and AD were effectively redefined when it was determined that Christ was almost certainly not born later than 4 BC [ref: http://www.tondering.dk/claus/cal/node3.html#SECTION003142000000000000000]. Why should the definitions of this one set of terms (a.m. and p.m.) be "etched in stone" for all eternity when all others are not only allowed to evolve, but HAVE, in fact, evolved?

B.) Quite generally, English words derived from Latin do not share their exact original Latin meanings. For example, quantum is from the Latin for "how much", but quantum mechanics is hardly synonymous with "how much" mechanics. Also, milli- is from the Latin for 1000, but a millimeter is 1/1000th of a meter, not 1000 meters. Other examples abound. (One could also argue that since the terms include "meridiem", which is almost certainly based on the sun crossing the meridian, they are applicable only to local solar time, not the standard time we use today. Unfortunately, I don't know enough about Latin to determine the validity of this one particular point. But "meridiem" is certainly a lot more similar to "meridian" than it is to "noon". Therefore, it may be that the definitions that Korky favors are not really "literal" after all!)

C.) It is highly impractical to construct digital clocks with any other convention. In fact, if one were to construct a digital clock without any regard to what happens at 12:00, one would automatically end up with the de facto convention.

D.) It is already the de facto convention and most people know it and accept it. For those who haven't learned it yet there should be a reference.

I "campaign" ONLY in the discussion pages and only in response to those who insist 12 a.m. and 12 p.m. are "incorrect and should not be used". Though I see no problem in explaining in the article why this convention is used by digital clocks, and the article, in fact, does this quite well and very succinctly!

Alan's sun vs. standard time argument fails because standard time doesn't pretend an internally ILLOGICAL thing: that noon is before or after noon. It merely defines noon differently. It says to consider the whole time zone, east to west, as the same time as if it were geographically at the centre of that time zone (approximately), where sun and standard time are the same. The former "pretending" (convention) is confusing and illogical, the latter "pretending" is easy and logical.

It is "illogical" only if one insists on permanent, etched-in-stone-for-all-time, definitions of a.m. and p.m. If one simply allows for a small change in the definitions, it is very logical. Given point (4) above, why in this one case does Korky insist on making an exception to both adapting terms and conventions as needed and terms derived from Latin not having their exact literal Latin meanings?

Also, Alan's digital clock apparently omits the required space between the numbers and the am/pm, so he writes 12:00AM instead of 12:00 AM, because he's a slave to his (current!) gadgets and can't think for himself (I'm just trying to rib him in a gentle way!). Also, he writes the :00, even though it's superfluous, just because that's how he interprets his digital clock's demands. If he could read a clock with hands, perhaps he would realize that. (More teasing.)

This is meaningless nitpicking.

Similarly, Alan's BC-AD argument breaks apart. When BC and AD were coined, the Latin-speaking monk who did it (if I remember right) was trying to base it on Christ's birth. It's now long past the practical time to argue about when to start counting the years in that system. A mistake was probably made, we know it, we accept it, it doesn't cause any real day-to-day problems, so we don't want to change the numbering now.

My point was that if we allow for a non-literal interpretation of BC and AD, why not do the same for a.m. and p.m? The only "real problem" caused by the de facto convention is that people like Korky needlessly resist it. Additionally, this particular paragraph of his discredits his own literal-interpretation argument.

That's nothing like today's am-pm debate. The latter hasn't been going on for centuries. There is no convention that's been accepted by enough people and organisations to make it a done deal. And rejecting 12 am as noon wouldn't cause any problems, not like re-numbering all the years in all the books. Rather, rejecting 12 am and 12 pm would end the confusion. It would mean that people would just have to interpret their old digital clocks differently (or remember their illogical convention). I always have interpreted them correctly: when the 11 changes to 12 and the am changes to pm, that's the instant of noon. Noon isn't the interval after noon (pm) any more than it's the interval before noon (am).

Of course it's not. People sensibly accept the re-definition of AD and BC by abandoning "strict literal interpretation" of the original definitions but for some reason a few people like Korky refuse to do the same with a.m. and p.m.

Rejecting 12 a.m. causes problems because while most people know what it means and accept it, organizations like NIST refuse to even acknowledge it so those few who do get confused upon seeing it should have a clarifying reference. At least Wikipedia is -- at least as of this writing -- picking up the slack. I think the fact that every a.m./p.m. digital clock, including those embedded in computers (resulting in the same convention in software, the Web, and printing) uses this convention makes it a done deal. It is far more practical, economical, green, and easier to provide a reference to an easy rule than to replace all the world's a.m./p.m. clocks and all the electronic devices (including computers) that use them!

Instants and intervals are important mathematical concepts to teach in schools! Alan, you talk about them in "Alan E. Feldman 2005-12-05 06:22" above.

I presume, Alan, that you understand that a point in space has no length, width, or height. A horizontal line (pointing ahead of and behind you) has length, but no width or height. A horizontal plane has length and width, but no height.

Similarly, why not accept that noon and midnight are instants, not intervals, as Scott Wagner wrote? The same with any time: 5:23 pm is an instant, not the interval between 5:23 and 5:24, even though your sacred digital clock (according to your interpretation) insists that it is that entire interval.

Well, if noon and midnight are mere points with zero length, width, and height, they don't even exist, right? If you have something of zero size, you have nothing. If you have zero grams of something, you have nothing. So why all this fuss over nothing?! See how ridiculous you can get with logic that looks valid superficially?

[Seriously, now:] Whether a given time represents an instant or an "interval" depends on the context. Most of the time a written time means an instant. However, if it is a time printed in a computer log file, then it represents an interval starting at that time and ending just before the next possible time (as would be consistent with the relevant precision). When setting a clock, a time is an instant. Usually, an input time represents an instant and an output time represents an interval, except when the output consists of times in a schedule or timetable. You could say that an output time that records the time of an event represents an interval due to limitations of its precision and that output times that predict future events represent instants. But that, too, could vary with the context. ... Context matters!

I quote Alan some more:

"So you would have digital clocks show "12:00 noon" for the entire 60-second interval from 12:00:00 to 12:01:00? Are we to retrofit all existing 12-hour clocks to be in accord with this obsolete convention? Are you going to pay for this retrofitting? Children would not be confused if they were simply taught the de facto convention. You cannot say "midnight at the end of June 3" on a digital clock. --Alan E. Feldman 2007-03-18 15:31."

First answer: yes, show "noon" for 60 seconds if the clock doesn't show seconds. If you show it instantly, it will be invisible. People will have to learn the convention that it means that it was noon the instant they first saw "noon", not the whole time they see "noon". (If they don't understand that, no harm is really done, unlike if they don't understand your favoured convention.) That's much more logical and easy for children than your favoured convention.

I don't think a digital clock should show "noon" for 60 seconds. ... Digital clocks show the truncated time. Therefore, when it shows 12:00 it is indicating that the time is somewhere in the interval [12:00, 12:01). It would certainly be wrong for all points in this interval except the first to be shown as "noon". And since noon is a fleeting instant, as Korky says, why bother with it? The same applies to midnight, of course.

Trust me, children can handle the de facto convention. If they can't learn that, they can't learn anything else either. Look at it this way: An a.m. time is the amount of time that's passed since midnight; a p.m. time is the amount of time that's passed since noon; with 12 in the hours column being interpreted as zero. Very logical. Very simple. Very sensible, elegant, practical, economical, green, and mathematical. And Korky says "People will have to learn..." Well, all of a sudden it's okay for people to have to learn something! So why is he suddenly not concerned about that?

Retrofit? No. Your wonderful modern gadgets don't last long enough to worry about. I'm wearing my father's watch with hands which still work and are unambiguous 43 years after he died.

Irrelevant. Besides, his father's watch likely doesn't indicate a.m. or p.m., so his use of "unambiguous" is ambiguous!

Yes, you can have dates on a digital clock. Better throw out yours and get one which says whether the midnight is the end or the beginning of the date. If they don't make them yet, they will. (Friendly sarcasm. I'm a green who hates waste.)

So if you hate waste, why waste the additional time and materials it would take to build such clocks and the additional energy it would take to run them?

Then Alan writes, "According to your logic: At noon you see the hour change from 11 to 12. Noon is the instant of that change, and so is neither 11 nor 12. (Okay, a slight cheat, but very slight.)" (Alan E. Feldman 2007-03-25 15:27.)

Wrong, Alan. It's like I said above: Noon is the instant the 12 first appears.

Korky just contradicted himself. In these same "20 short paragraphs" he said, "I always have interpreted them correctly: when the 11 changes to 12 and the am changes to pm, that's the instant of noon."

Alan, you argue passionately that we should adapt to our machines and their alleged logic (based on the imperfect human computer programmers) rather than making our machines adapt to us.

Most people already have, and for good reason. It's really not that hard.

Maybe so, but Wikipedia is not a activist organisation. And, anyway, I'm an activist on the opposite side. What you who cannot accept the concept of "instant" should do, if you really want to "win", is to invent new terms. Maybe N+ to mean noon plus the interval from noon to midnight. Maybe M+ to mean midnight plus the interval from midnight to noon. So at noon your new, advanced (still holy?) digital clock would roll over to 12 N+.

I am not asking Wikipedia to be "activist". I am trying to ensure that the very important de facto convention is present and properly explained in this article.

If people can re-name BC and AD (they now call them BCE and CE), then why not? Alan said, "The terms ante meridiem and post meridiem originated some 400 years ago and applied to a *different* timekeeping system. It's time to move on." (2007/01/02).

So go ahead and "move on", Alan--and be sure to leave the old, no-longer-meaningful terms AM and PM behind, too. And when you get support for new terms in the "real world", you can add them to Wikipedia. (If you adopt my coinage, give me credit for inventing N+ and M+ on 2007 April 2 12h09.) -- The 20 paragraphs above are by Korky Day 02:41, 3 April 2007 (UTC)

Most people have already "moved on" by having learned and accepted the de facto convention. A woman in the finance department at my company wrote in a recent email that our timesheets were due "12:00 pm Monday, July 2nd". Schedules and timetables I come across aren't bothering with things like 12:01 a.m. any more. Businesses are correctly using the de facto convention in their business-hours signs. As far as I can tell, all a.m./p.m. digital clocks use this convention. This includes computers which means software and the Web use the same convention. It is only a very few people like Korky who are causing (or rather, exacerbating) confusion. I, for one, I am trying to clear up the "confusion".

--Alan E. Feldman 2007-07-02 18:45 UTC

Phrase needs verification

In chapter "Confusion at noon and midnight" it says

[begin quote] Noon and midnight are only infinitesimal points in time, and therefore it is not practical to use any other convention than that which also applies immediately afterwards, when the clock still displays 12:00. This convention is standardized for computer usage in American National Standard ANSI INCITS 310 (which extends the international standard ISO 8601 time notation with a 12-hour a.m./p.m. variant for the U.S.-market). [end quote]

The version of ANSI INCITS 310 which is currently available at [ansi.org] (go to the shop, search for INCITS 310) does not deal with the 12 hour format at all. It gives some explanation on how to use the 24 hour format in data exchange in the U.S., but no clarification, description or explanation on the 12 hour format. Where does the information given above come from? Was an explanation of the 12 hour format contained in a different version? Is the reference wrong? Or should the last sentence of this phrase rather be deleted from the article? Daisy G 12:37, 21 December 2006 (UTC)

I recall having found sometimes in 1997 in the standards room of the Purdue University Engineering Library an ANSI standard that looked very similar to a national adoption of ISO 8601, but had also a 12-h notation option as an extention. I do not recall the exact document reference, whether it was "ANSI INCITS 310" or some predecesor of it, or whether what I found then is still a currently active ANSI standard as of 2006. Someone with easy access to a US library with a good ANSI collection might be able to answer this one. (ANSI standards are not easily available here in Britain.) It would be nice to have the correct reference and a description of what that specification did with midnight and noon. Sadly, the web sites of most standards bodies are very bad at providing access to information about historic documents. Markus Kuhn 15:46, 1 January 2007 (UTC)

The standard I meant was probably ANSI X3.43-1986 "Representations of local time of day for information interchange" (also FIPS PUB 58-1) or ANSI X3.51-1975 "Representations of Universal Time, local time differentials, and United States time zone references for information interchange" (also FIPS PUB 59), but it seems that these FIPS documents were withdrawn in 1997 [1]. So the only US standard now left for representing time in information interchange is essentially identical to ISO 8601. (Given the various ambiguities inherent to the am/pm notation, that's certainly a wise choice, IMHO.) Markus Kuhn 08:22, 23 January 2007 (UTC)

While there is apparently no mention of the de facto convention in the ANSI standard (at lesat not easily findable on the Web), I have yet to see a single computer not use it. --Alan E. Feldman 2007-03-18 15:56 UTC

Begin and end of day (24:00 issue)

The article contains this line:

Some people dislike the idea that midnight has two different notations in the 24-hour clock, depending on whether it is the beginning (00:00) or end (24:00) of the day.

Under which user 155.208.231.234 added:

This seems an odd argument to someone growing up with the 24 hour system, as we know that even though it is a TWENTYFOUR hour system, the time will never be displayed as 24. From 23:59:59, the time shifts to 00:00:00. Midnight will never be referred to as "at 24 o'clock" in writing nor in speech. In speech it is always "midnight", in wirting it can be referred to as "midnight" or in the format hh:MM (00:02 for two minutes past midnight).

This comment ignores the fact that written notices often contain 24:00 to explicitly indicate the end of the day. For example in ranges used for opening hours as from 07:00 to 24:00. See also the example of a time table using the 24:00 as well as the 00:00 notation in 24 hour clock. −Woodstone 15:16, 22 December 2006 (UTC)

While I tend to agree that midnight should always be 0000 I have seen 2400 used in many countries, on many clocks and machines and very commonly on timetables (especially in Italy where I am at the moment). 24.00 (and sometimes even 24.05, etc.) is found at the end of timetables because if you need a bus in the late evening you naturally look at the end of the timetable and not the start (where you might normally expect to see 0005, etc.). As for spoken language I don't really know because most people seem to pronounce the 24h times as 'undici ora' (eleven o'clock) rather than 'ventitre ora' (twenty-three o'clock).--Xania

^talk 00:55, 31 January 2009 (UTC)

12:01

Is there any information on what the various sources saying that noon should be something other than 12:00 pm, have to say about 12:01, 12:30, etc? --Random832(t c) 13:27, 31 January 2007 (UTC)

I don't think there is any explicit mention of times like 12:01 and 12:30 in any reference, but both the original, archaic, obsolete defintions of a.m. and p.m. and the newer de facto convention agree that all times between 00:00 and 12:00 are a.m. and all times between 12:00 and 24:00 are p.m., which covers your question. There is generally confusion only about the times 00:00 and 12:00. --Alan E. Feldman 2007-03-18 15:51 UTC.

Digital decoders' time notation

Are there any providers of cable or satellite television that still use the 12-hour notation in their set-top boxes? I ust want to find out which pay television companies have 12-hour time and which ones have 24-hour time. Scott Gall 04:40, 8 February 2007 (UTC) PS: My one uses 24-hour.

Leading Zero in 12 Hour Clock

What do people think about changing the section that refers to using a leading zero with the 12 Hour Clock? In my (completely amateurish) opinion the use of a leading zero in the twelve hour clock is just wrong. To my eyes writing 09:02 pm is simply incorrect, not as the article seems to suggest merely a matter of style. Before the arrival of the 24 hour clock would there have been instances of using a leading zero? I suspect not; using a leading zero in the 12 hour system would presumably only have occured after the 24 hour system came into being, at which point the leading zero may have been erroneously copied to the other format. In addition I don't think this section uses adequate citation either. "Some style guides suggest," which style guides? I also feel the statement of "many digital clocks" is not rigorous enough either. I have never seen a digital clock that uses a leading-zero when displaying 12 hour format - I'm not making this up to add to my argument, it's just I've never seen this. Although I'm sure if you looked hard enough you could find some, I'd venture that this probably doesn't count as "many" and those that do use a leading zero are just poorly manufactured (I've got a clock laying in a drawer that seems to reset itself every five days - but that doesn't add evidence to the case that time starts over at the end of every working week - it's just cheap tat that shouldn't have passed quality control). No offence meant to the original contributor of this piece, I agree with their basic premise and it's probably a better start than I could have made. But I think in attempting to preserve neutrality they are giving credence to a state of affairs which is more likely to have come about by error or exceptionally rather than deliberately. Be interested to hear what other people say - I daresay I am now about to be shot down by a whole bunch of people who regularly write 01:30 pm etc... DistractionActivity 14:24, 21 February 2007 (UTC)

Seen as I posted this in Feb and no-one has replied, I'm guessing the above isn't controversial. I'm deleting the leading-zero section. I think it confuses more than it helps. DistractionActivity 16:34, 20 July 2007 (UTC)

Monosyllabic Words

Regarding the section "Advantages of the 12-hour clock", "eleven" is also not a monosyllabic word. - Lake Ontario 12:33, 21 March 2007 (UTC)

Twelve Hour Clock

Combining the number twelve with ante meridiem or post meridiem is ambiguous and incorrect. The descriptor would be either noon or midnight sans the number. Garnered from the Physics Laboratory Time and Frequency Division of the National Institute of Standards and Technology Betwixt Words 19:46, 7 June 2007 (UTC)

Yes, that's what NIST says. It is ambiguous and "incorrect" only if you insist on sticking with the original convention for a.m. and p.m. But the world goes by a new convention: the de facto convention. While it's not "official", it is extremely pervasive and is far more important.

NIST is in serious disagreement with every a.m./p.m. digital clock on the planet (not to mention many published transit timetables, event times, business-hours signs, and the like both in print and on the Web). These clocks and usages have established a new de facto convention for the 12-hour clock. It is a sensible, logical, mathematical, and practical convention and is in fact the de facto convention. To construct digital clocks otherwise would be a huge waste of time and resources and would cause other major problems. (Note that 12 noon and 12 midnight [the latter being ambiguous!!!] is actually okay in the de facto convention. The de facto convention only legitimatizes 12 a.m. and p.m. for use as midnight at the beginning of the day and noon, respectively. And it is the best convention for the 12-hour clock. And it is, in fact, de facto.)

"... Time is a totally artificial, man-made construct, a lie we have agreed to tell so we can know what time to show up for dinner. It's like a mile or a pound -- it's not important what they are as long as we all agree that they're the same thing. ..." (attribution given below in "The Entire Article" section)

You say "incorrect" as if it were some deep "God-given" truth. Well, you can't change the value of pi but you can change a convention. In fact, every other convention in measurement has been modified as needed, especially those for time. Why should this one alone be etched in stone for all eternity? The "authorities" once told us the Sun went around the Earth. Of course that wasn't a convention, but you get the point. Authorities can be wrong and it is plainly obvious to anyone who has used a digital clock that NIST is behind the times. So 12 a.m. and 12 p.m. are not "incorrect" but adhere instead to a different convention, a far more useful and important convention: the de facto convention of a.m./p.m. digital clocks (including system clocks in computers). You may be upset that the "literal translation" wouldn't hold at these two times, but so what? Quantum is Latin for "how much" but no one would seriously consider quantum mechanics synonymous with "how much" mechanics. The same goes for most other words derived from Latin, Greek, etc. Note that the SI prefix "milli-" (used in millimeter, e.g.) comes from the Latin word for 1000. But a millimeter is not 1000 meters but instead 1/1000th of a meter. Are you going to protest this one, too? Are you going to say "millimeter is incorrect"? According to "literal interpretation logic" you would have to! The same argument applies to AD and BC. See my comments in "The Entire Article" section and elsewhere on this discussion page. --Alan E. Feldman 2007-06-27 12:27 UTC

Re the first entry in this section: Please note that even the term "midnight" is ambiguous. Usually it means the midnight at the end of the day but at 12:10 a.m. one could say it's ten past midnight with this midnight meaning, of course, the midnight at the beginning of the day. So much for avoiding ambiguity!!!

--Alan E. Feldman 2007-07-04 01:36 UTC

If I hear "12 am" or "12 pm" I have literally no idea what time is indicated. If there is a "de facto convention" on this, it has entirely passed me by. For what it's worth I grew up in the UK.

199.89.64.179 18:00, 18 July 2007 (UTC)

24-Hour Timekeeping System

I wish there was a 24:00, as I absolutely could use a 25 hour day. Our day begins at zero hundred hours with midnight and ends at 23:59.

 1)00:00-00:59  2)01:00-01:59  3)02:00-02:59  4)03:00-03:59  5)04:00-04:59  
 6)05:00-05:59  7)06:00-06:59  8)07:00-07:59  9)08:00-08:59 10)09:00-09:59 
11)10:00-10:59 12)11:00-11:59 13)12:00-12:59 14)13:00-13:59 15)14:00-14:59 
16)15:00-15:59 17)16:00-16:59 18)17:00-17:59 19)18:00-18:59 20)19:00-19:59
21)20:00-20:59 22)21:00-21:59 23)22:00-22:59 24)23:00-23:59:59

This system would leave no room for alarm errors. Betwixt Words 21:18, 9 June 2007 (UTC)

You really believe that?! Each day ends a minute before midnight?

But this does highlight something that bothers me about this debate. In everyday language people refer to midnight of a particular day as the end of that day. Am I mistaken about this? Why is this never addressed? Dyaimz 22:07, 20 June 2007 (UTC)

Greetings...The day ends at 11:59:59 pm or 23:59:59 and begins at midnight or 00:00. Betwixt Words 01:04, 24 June 2007 (UTC)

Then what day is it during the 60 seconds between 23:59 and 00:00? --Alan E. Feldman 2007-06-30 19:02 UTC

Greetings...End of time now reads 23:59:59, for ease of understanding; as all minutes and seconds within the boundries of any particular hour are actually considered included.

The contributions in this article were probably reworked by several hundred people, not one person. I myself added only two lines, a reference and the words noon and midnight. The rest of the article I did not bother, so the inference about illogic of the rest of the article is not applicable in my situation. Thank you though Betwixt Words 15:35, 5 July 2007 (UTC)

If you say the day ends at 23:59:59 then I ask, "What day is it between that time and midnight one second later?" Midnight can be either the beginning or end of the day. It all depends on the context. If you say something happened at 12:00 a.m., or 00:00, it is at the beginning of the day. If you say something happened at 24:00, it is at the end of the day. If you say, "Meet me at midnight tonight" it means midnight at the end of the current day. This is one of those cases where it is better to completely rephrase what you want to say. Perhaps just say that the date changes at midnight.

--Alan E. Feldman 2007-07-07 02:17 UTC.

The entire article

The Time and Frequency Division, part of NIST's Physics Laboratory, maintains the standard for frequency and time interval for the United States, provides official time to the United States, and carries out a broad program of research and service activities in time and frequency metrology.^[1]Please see NIST's home page. If this governmental agency maintains the time standard for our country and state that the use of p.m. and a.m. are incorrect when referring to midnight and noon, I feel it a mute point to offer anything else other than suggestions. This entire aritcle needs reworking. Betwixt Words 16:52, 17 June 2007 (UTC)

The article already states that. The article is not saying that the standard(s) is(are) ambiguous but that the vulgar (common) every day use is ambiguous.Zginder 19:32, 17 June 2007 (UTC)

Greetings...Number one reference on the article page states 'The answer is that the terms 12 a.m. and 12 p.m. are wrong and should not be used. Therefore, either 12 a.m. or 12 p.m. could work as a designation for midnight, but both would be ambiguous as to the date intended.'

Could work, but would be ambiguous.

The reference I used in the introduction of 'The Entire Article' is from the same web site, only from the home page. Betwixt Words 01:17, 24 June 2007 (UTC)

Re the first entry in "The entire article" section: You completely ignore the pervasiveness of digital clocks. Are you saying that every a.m./p.m. digital clock is wrong at 12:00? Are all computers using the system similarly wrong? Tell me: What reference shall someone use if they have an email or computer file timestamped with "12:00 a.m." or "12:00 p.m." to tell if it means midnight at the beginning of the day, midnight at the end of the day, or noon? There are two conventions: the original archaic obsolete one, and the de facto one. Both should be in the article. NIST is stuck on the archaic one -- why I don't know. But when people go to the NIST FAQ to find out which of 3 possible times their 12:00 a.m. or. p.m. email arrived, they are ill served. You can write that the de facto convention is not "strictly correct according to the original definitions of ..." if you wish, but excluding the de facto convention would be a disservice to Wikipedia reader.

This reminds me of what Jon Carrol wrote in

http://www.sfgate.com/cgi-bin/article.cgi?f=/chronicle/archive/2005/04/15/DDG27BCFPA1.DTL

"... Time is a totally artificial, man-made construct, a lie we have agreed to tell so we can know what time to show up for dinner. It's like a mile or a pound -- it's not important what they are as long as we all agree that they're the same thing. ..." [I think he means "our system of keeping or telling time ...."]

So it's not like the archaic version of the a.m./p.m. system is some "eternal truth". It's a convention and conventions sometimes change as needed. NIST is behind the times on this one. Both conventions should be in the article. If NIST suddenly declared that the English system of measurement is "wrong and should not be used", but people kept using it anyway, are we to exclude the English system from Wikipedia articles? I think not. (Personally I think the original system has outlived its usefulness. Note that in the de facto system you can still use terms like noon and midnight, so I don't see why it's a problem as it covers all bases.)

   >----o----<

I like the revision as of 20:49, 31 May 2007. I don't see anything wrong with it except for, perhaps, a few obscure references in the second table and perhaps an awkward phrase or two.

In the current version I think the opening paragraph is bad. I quote it here for ease in reference:

"""The 12 hour clock is a timekeeping convention in which the 24 hours of the day are divided into two periods called ante meridiem (a.m., from Latin, literally "before the middle of the day", idiomatically "approaching midday") and post meridiem (p.m., "past midday"). Each period consists of 12 hours numbered 12 (acting as zero), 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11. Time periods run 12:00 through 11:59 either ante meridiem or post meridiem.[1] When referencing twelve o'clock the descriptor is either noon or midnight, without the ambiguity of the meridiems."""

Why do we need to spell out all the numerals from 1 to 11? Then it says that 12:00 through 11:59 can be either a.m. and p.m. Then it contradicts that by saying 12:00 is excluded. Aside from the argument over whether phrases such as "12:00 a.m." and "12:00 p.m." are "legitimate" or not, this paragraph is a bad description of the "original" 12-hour clock system. And if the day is "divided into two periods called ante meridiem and post meridiem, to which periods do noon and midnight belong? Again a self-contradiction. And there is no such thing as a "meridiem" so the "ambiguity of the meridiems" is not a good phrase. (Certainly "a.m." and "p.m." are not "meridiems"! They are before and after "meridiems", no?) And midnight is _still_ ambiguous, save for the common usage of it being the end of the day. What was so terrible about the May 31 version?

Trying to explain the 12-hour system in an article is a little like trying to define the word 'the'. You'll never really learn what it means by reading its definition in a dictionary. You learn it by learning the language.

I also prefer the first table from that version. What was so terrible about it? One could add an asterisk to the a.m. and p.m. phrases in the 12:00 boxes and say "* according to the de facto convention which is what is used on digital clocks". And why do we need every hour in the table? Why not include all the minutes? 12:00 midnight (a.m.*), 12:01 a.m., 12:02 a.m., ..., 3:45 p.m., 3:46 p.m., ....

Another problem with the current table is that is implies that midnight is at the beginning of the day when in the original system it can be the beginning or end of the day and it is the opposite of the very common usage of midnight as being at the end of the day.

--Alan E. Feldman 2007-06-24 15:07 UTC

Hey Alan, something we agree on! "it is the opposite of the very common usage of midnight as being at the end of the day." I have been trying to get an answer to whether the use of the word midnight, in speech or writing, to mean the end of the day is disputed or not. Certainly in my whole life as a native English speaker I've not known it used otherwise than the end of the day. Dyaimz 22:17, 27 June 2007 (UTC)

I find it ironic that according to the "literal interpretations" of ante meridiem and post meridiem, midnight is actually ambiguous with respect to whether it is at the beginning or end of the day. But somehow this doesn't bother the "literalists" (those who insist 12 a.m. and 12 p.m. are "ambiguous and incorrect"). Consistency would demand that they declare "midnight" to be "ambiguous and incorrect"!

Yes, common usage in most situations usually means at the end of the day, but that is NOT deducible from the literal interpretations of a.m. and p.m. Also, one could imagine a situation in which someone on a subway platform just missed a train that left at midnight and only minutes later tells a fellow straphanger that the last train left at midnight in which case it would not be the midnight at the end of the current day! One could in fact say when asked the time, "It's a few minutes after midnight". In the de facto convention, 12 a.m. is not ambiguous at all.

On the other hand, one could argue that the term 'midnight' is ambiguous until determined by the context in which it is used, but only if such context is decisive.

--Alan E. Feldman 2007-06-27 13:17 UTC

Greetings...I do believe the twelve hour clock antiquated and love the simplicity of the 24 hour clock.

Regarding the fellow on the platform, time runs in a circular motion. When one takes the sleep period out of the equation, one understands. Midnight is midnight and noon is noon.

Day begins at midnight or 00:00(:00); twelve hours after the fact is noon and ends at 23:59(:59). If there was a 24:00 there would be a 25-hour clock.

When I discovered this Wikipedia article I was on the am/pm 12 hour day bandwagon, but I now love the logic of a 24 hour clock due to the logic of NIST. My computer? ..set on a 24 hour day, same as NIST's.

I believe that time should be universally set. I also believe that our standard of measurement should be reset to reflect a universal standard.

Aren't opinions grand...makes life interesting. Betwixt Words 15:05, 5 July 2007 (UTC)

I actually prefer the 24-hour system, at least on computers. I hate seeing a column of a.m./p.m. times and prefer the date as yyyy-mm-dd. I also set all computers I use, but don't share with others, this way. HOWEVER, if, for some reason, one wants to or has to use the 12-hour system, I see no reason to exclude 12:00 from a.m. and p.m. Most people in the 12-hour world get their time from a.m./p.m. digital clocks and they use the de facto convention I have described. Most people have learned the de facto convention, accept it, and use it without much trouble. I cannot understand why some are so resistant to such a minor, but eminently useful change in the definition, especially since it really is the 'de facto' convention for a.m. and p.m. I have given ample reasons on this discussion page for this. The NIST argument is fine for the pre-digital clock age. But it leaves people hanging when they have an email, file, or other thing timestamped with 12:00 a.m. or p.m. and may be confused as to which of the 3 possible interpretations it is. Well, it's always the one given by the de facto convention when computers are involved. Moreover, I've seen more and more correct use of 12 a.m. and 12 p.m. in all sorts of contexts. And it isn't going to change anytime soon. So the best solution, it seems to me, is to go with the de facto convention. This would keep confusion to a minimum. I don't understand the horror of special-casing 12:00. We do this for other things, as in the definition of prime numbers. If we simply say a prime number is a number divisible only by one and itself, a host of useful theorems are lost. If one simply adds the special case excluding the number 1 as a prime number, all the beautiful and useful theorems are regained. (I had another example in mind but it escapes me at the moment.) Furthermore, the NIST argument is not entirely correct: midnight is ambiguous unless it is pinned down by context. If you say midnight, July 12, is it midnight at the beginning or at the end? There is no way to tell. If at 00:10 you say, "It's 10 past midnight", clearly you are referring to the midnight at the beginning of the current day. If you say your store is open from 8 a.m. through midnight, context makes it clear that it means the midnight at the end of the day. So the whole point of the argument of not assigning an a.m. or p.m. value to midnight due to ambiguity falls apart because without sufficient context, the term 'midnight' itself is still ambiguous.

The time 24:00 is commonly used in 24-hour countries to mean midnight at the end of the day. (It is equivalent to 00:00 the next day.) In fact, there is an image of a train schedule in another Wikipedia article showing exactly this. So, whether the day starts or ends at midnight depends on the context established by your statement. I don't see where you get that the day ends at 23:59. Digital clocks show the truncated time, so it is only 23:59 at the moment it first appears on the clock.

You can avoid the special-casing of 12:00 by describing the 12-hour system thusly: The day starts at midnight and noon is halfway through the day. Times before noon are given as the number hours and minutes (and seconds if such precision is desired) that have passed since midnight (with 12 representing zero) and are given an a.m. suffix. From noon forward, times are given as the number of hours and minutes that have passed since noon and given a p.m. suffix. Very simple with no special cases! I see no reason to fret about literal interpretation of ante meridiem and post meridiem because most words derived from Latin do not share the original Latin definition. Furthermore, as I have also argued above, if you consider p.m. to be times after the sun crosses the (celestial) meridian, then that applies only to local solar time (or apparent solar time), and since we go by standard time, they don't strictly apply anyway, so why all the fuss?

--Alan E. Feldman 2007-07-06 00:20 UTC

Pronunciation

In some countries (I know Norway is one example) a phrase such as "half five" is not short for half past five as stated in the article but rather means 4:30 (i.e. half way to five) Does anyone know of other countries where this is true? Seems like it should be mentioned in the article but since so many other people have worked on it I'll let someone else make any changes.

Kirkmona 16:13, 20 June 2007 (UTC)

This is an interesting point. Norway is definitely not the only country where it has this meaning. But as this is the English version of WP should we only be concerned if this usage happened in an English speaking community? Dyaimz 20:22, 22 June 2007 (UTC)

One of the references is wrong!!!

In the article it says the following:

The abbreviation a.m. stands for ante-meridiem (before the Sun has crossed the line) and p.m. for post-meridiem (after the Sun has crossed the line). At 12 noon, the Sun is at its highest point in the sky and directly over the meridian. It is therefore neither "ante-" nor "post-".

http://www.nmm.ac.uk/server/show/conWebDoc.17917

This is clearly wrong. The sun is in general not at its highest point at noon because of the use of Standard Time and the equation of time. While it is an "official reference", it is simply wrong. Consider just the Standard Time portion of my argument: The sun is certainly not at its highest point in the sky at the same instant for an entire time zone. That alone makes this reference wrong. Now, add the equation of time and it gets even worse: the meridians (lines of longitude on the earth) for which the sun is at its highest point in the sky at noon move throughout the year according to the equation of time.

Therefore, this should be immediately excised from the article. --Alan E. Feldman 2007-07-03 16:02 UTC.

Further proof that this reference is wrong:

Another page on the same Web site says I'm right!

From

http://www.nmm.ac.uk/server/show/conWebDoc.351

I quote the following passages:

"If a sundial is used to determine the time it rapidly becomes apparent that it does not indicate the same time as clock time. The difference amounting to some 16 minutes at certain times of year. This difference is also seen as an asymmetry in the times of sunrise and sunset. It is called the Equation of Time. ... The Sun will be on the meridian at noon at both solstices and equinoxes and so the equation of time due to obliquity will be zero at these times. Between the solstices and the equinoxes the Sun will be slow relative to clock time with minima near 5 February and 5 August. Between equinoxes and solstices the Sun will be fast relative to clocks with maxima near 5 May and 5 November."

But that's just due to the tilt of the Earth's axis. Add to that the varying speed of the earth around the sun and the variation gets more complicated:

"The total of these two effects gives the equation of time, which is formally defined as the difference between clock time and apparent solar time."

So the statement that the sun is at its highest point in the sky at noon only applies to "apparent solar time" (or what I prefer to call "local solar time"). Now, add the fact that we actually go by Standard Time and you see that the referenced quote in the article is wrong according to the same Web site!!!

Please, someone remove this error (the part that says that the sun is at its highest point in the sky at noon) from the article at once. I'd do it myself but I don't want to appear as a lone hostile vandal.

The is more proof that you can't believe everything you read, even well-respected references. It also shows that there is a problem with using the literal interpretations of ante meridiem and post meridiem to claim that 12 a.m. and 12 p.m. are "incorrect and should not be used." This is clearly a blow for the argument that noon is neither ante meridiem nor post meridiem!

So while references are certainly very important, they are subject to error like everything else and shouldn't be treated as if they were perfectly infallible.

Alan E. Feldman 2007-07-04 00:56 UTC

I only see a well-sourced description of what AM and PM mean in the article. Whether in practice, the usage of those terms appears to be incorrect in relation to the actual position of the sun, is irrelevant. That's simply a result of us not using sun dials anymore, but rather a 24-hour time system adopted over the whole world.--Atlan (talk) 16:36, 4 July 2007 (UTC)

Where do you see a "well-sourced description"? "In the article" is vague. I'm talking about this particular quote. IT IS WRONG. There is no doubt about it. Therefore, it should be removed from the article ASAP. Well-sourced or otherwise -- it is still wrong.

Let's look at it again, shall we?

The abbreviation a.m. stands for ante-meridiem (before the Sun has crossed the line) and p.m. for post-meridiem (after the Sun has crossed the line). At 12 noon, the Sun is at its highest point in the sky and directly over the meridian. It is therefore neither "ante-" nor "post-".

http://www.nmm.ac.uk/server/show/conWebDoc.17917

This is clearly wrong. The Sun does not cross the "line" at noon, civil time. It crosses the "line" (the celestial meridian, actually) at noon, local solar time (or apparent solar time, if you prefer). But we go by civil time, not local solar time. The sun does not cross the line at noon, civil time except at specific lines of longitude that vary throughout the year according to the equation of time. Therefore the quoted material is wrong. In fact, other pages on the same Web site prove it is wrong!

You claim this is irrelevant because we don't use sundials. Please explain how that makes this quote correct when it is still plainly wrong. Do you mean to say that the wrongness of the quote is irrelevant? And that that makes it okay to post false statements?

Tell me: When someone comes across a file with a timestamp of 12:00 a.m. or 12:00 p.m., when someone has an email whose timestamp is one of these two times, when someone uses software that displays one of these two times, and this someone hasn't learned the de facto convention yet, or for some reason (perhaps lack of sleep) can't remember it -- what reference is this person to use to resolve the issue? Tell me. How will this person determine which of the three possible times (00:00, 12:00, 24:00) this timestamp denotes?

If civil time is relevant and sundials are irrelevant, tell me why this incorrect quote is relevant, even though it is 100% wrong, and every a.m./p.m. digital clock in the world, including those embedded in computers and other electronic devices, and the de facto convention all these clocks use, is somehow irrelevant?

Could you please also be more clear with your comments? Thanks.

Alan E. Feldman 2007-07-05 5:36 UTC.

There's no such thing as a 'meridiem'

Betwixt Words, please give me a reference showing that meridiem by itself is actually a word. You can't, because it isn't. It's like 'memoriam". That is not a word either, but 'in memoriam' is (a compound word).

Alan E. Feldman 2007-07-07 00:41 UTC

The root word is only complete with either unhyphenated prefix and does not stand alone. I wish I could take credit for authoring more than two lines in the article, but I can not; therefore the question belongs to someone else. Thank you though Betwixt Words 20:01, 8 July 2007 (UTC)

Meridiem is the accusative case of the fifth declension Latin noun meridies (midday or noon). The accusative case is usually used when the noun is the object of a verb. But Cicero did use it with the adjectives ante and post, so meridiem is not used alone. However, the Anglicized form is meridian, which the OED indicates is "long rare" in the sense of noon. Nevertheless, it is used in United States law in the forms ante meridian and post meridian.[2][3][4][5] — Joe Kress 03:49, 9 July 2007 (UTC)

Small correction. The ante and post are prepositions (not adjectives) and they always take the accusative case. The normal meaning of meridian in English is the great circle passing through the poles and the zenith at a place on Earth. So ante meridian would mean: before the Sun passes that circle (as originally, when local solar time was used). Admittedly all of this has no real bearing on the article. −Woodstone 20:29, 18 July 2007 (UTC)

Translations of a.m. and p.m. (2)

I am continuing the discussion fund here Talk:12-hour_clock/Archive_1#Translations_of_a.m._and_p.m.

The current translations are ante meridiem (a.m., from Latin, literally "before the middle day", idiomatically "approaching midday") and post meridiem (p.m., "after the middle day"). If I did not participate in the above stated discussion I would have no clue what "approaching midday" means. We need a translation that people will understand.Zginder 02:02, 26 August 2007 (UTC)

My guess is that "approaching/past" were chosen as a mnemonic. The editor looked for words beginning with "a" and "p" to facilitate remembering which is which. Literally "ante" and "post" mean "before" and "after", "meri" is a form of the word "medius", meaning "middle", and "dies" means "day". My favorite translation is therefore "before/after midday". Because latin has no articles, it could equally well be translated as "before/after the middle of the day". −Woodstone 10:36, 26 August 2007 (UTC)

So what is the translation of millimeter? Milli is Latin for 1000. Yet a millimeter is not 1000 meters. Actually, literally it would be 1000meter. Please, can we stop this semantic super-nitpicking about the "translations" of a.m. and p.m.? Quantum mechanics literally would mean "how much" mechanics because quantum is Latin for "how much". But it does not mean "how much mechanics". That would be ridiculous. There are many other examples I won't give here right now. This shows that a word's meaning and its origin are not the same. There is usually some relation, but often not equivalence. So let's stop fretting so much about exactly what these words mean and just do what's sensible for our day and age (not the 1600's), which is, of course, to go with the de facto convention that a.m./p.m. digital clocks already use anyway (including those in computers). One more thing: Even if we accept the "literal translations of a.m. and p.m.", whatever we determine them to be, is it so terrible to include an exception in the rule itself? We do this for prime numbers. The exception is that 1 is not considered to be a prime number, even though it is divisible only by 1 and itself. But if we didn't make 1 an exception, all the beautiful theorems about prime numbers would disappear. So it's sensible to make 1 an exception. The same goes for midnight and noon with a.m. and p.m. All other terms and units of time have changed over the years as needed. Why should a.m. and p.m., especially with the reasons I just gave, be any different?

Alan E. Feldman 2007-09-21 03:02 UTC

Precession of the equinoxes

Two weeks ago, 204.210.99.14 added the hypothesis that the 2x12 hour clock has been derived from the precession of the equinoxes. The hypothesis doesn't make sense because the true period is 25,765 instead of 24,000. In addition, it is unlikely that ancient civilisations would have been without a timekeeping convention before somebody figured out the precession of the equinoxes and its period. I removed the hypothesis, but Zginder restored it. Why? Ceinturion 20:11, 1 September 2007 (UTC)

I restored it because it appears to have a source. Has anyone checked it? Zginder 02:46, 2 September 2007 (UTC)

204.210.99.14 uses that source only for the statement that a number of ancient cultures knew the precession of the equinoxes. Wikipedia has an article about the book: "When Hamlet's Mill first came out, it was severely criticized for a number of things, such as tenuous arguments based on incorrect or outdated linguistic information, lack of modern sources and an over-reliance on coincidence, and the general implausibility of such a farflung of influential civilization existing and not leaving behind solid evidence, and at best given a grudging sort of praise, especially for its use of linguistic arguments and analyses." Unless somebody improves this flimsy section, I'll remove it again next week. Ceinturion 08:35, 2 September 2007 (UTC)

I have removed it. Putting below for better reference:

"While no one is certain of the exact origin of the 24 hour day with its two periods of am and pm it does appear to be a microcosm of the mythological cycle of a Great Year. This was the Greek as well as the Chinese term for one precession of the equinox, a period often rounded to 24,000 years, comprised of two 12,000 year phases. According to Giorgio de Santillana, the former professor of the history of science at MIT, and author of Hamlet's Mill, over thirty ancient cultures believed in this cycle of rising and falling ages marked by the slow movement of the equinox through the twelve contellations of the ancient zodiac." Zginder 01:36, 3 September 2007 (UTC)

12am/pm midnight/noon and advantages

The way I use to decide whether 12am or 12pm indicates noon or midnight is relatively simple. Think not of the moment (instant of time) but rather the whole hour (that starts with that instant and ends one hour later). Is that hour AM (before noon) or PM (after noon)?

You can apply the same logic to however precisely the time is specified. For example, 12:00:00.000am is midnight because that whole millisecond is before noon...

Another way to look at it is the perspective of the digital watch/clock makers. Assume that they would actually show some other suffix (or no suffix at all) at noon or midnight. How long would you actually see that time displayed? Well, not for one hour (midday or midnight does not last that long), nor one minute, nor one second ... not any other smaller fraction of a second either because that instant has no duration (or, one could say, it has a zero duration). Therefore, even if clock makers did show, say "noon" and "midnight" in these cases, and they managed to show it for infinitely small (zero) duration, you wouldn't see it. You would just see the suffix that follows...

Now, I have a question - how many regions/locales/cultures/languages (pick whatever) is it common to say (pronounce, not write) time in 12-hour and/or 24-hour clock format? This may be one of the advantages in some places, as time may be mostly spoken in 12-hour way (even if written differently in the same place).

My own first language is rather interesting in this respect - it has two words in use that can both be translated as "hour". I don't believe there is an official standard but one of them is more commonly used with 12-hour clock and another with 24 hour clock. For example, 5 pm could be said as:

5 sati (posle podne - after noon)
17 časova

... although the word čas appears to be used less frequently for expressing times before noon.

What is the situation in other languages?

--Aleksandar Šušnjar 18:05, 6 September 2007 (UTC)

In Thai there are many "hour" words:

with the Thai 6-hour clock, roughly
- ti: hours 01:00 to 06:00
- mongchao: 07:00 to 11:00
- mong: 13:00 to 18:00
- thum: 17:00 to 23:00
with the 24-hour clock:
- nalika (also means "watch")
chuamong: one hour's duration

−Woodstone 20:11, 6 September 2007 (UTC)

Confusion at noon and midnight

Someone added the following asterisk-paragraph:

(*) Digital clocks and computers appear to show the times 12 a.m. and 12 p.m., as in the chart to the right. However, since any particular time is actually an instant, those two read-outs refer to the one-minute periods following the instants of noon and midnight, not to the instants itself. In other words, 11:59 am shows until noon, 12:00 p.m. shows after noon; neither shows the instant of noon.

The problem with this is that it doesn't cover other important cases where the times 12 a.m. and 12 p.m. appear as the times of past or future events or are used to set a clock. A computer log file will often contain time stamps of past events. A time stamp may appear as 12:00:00.00 a.m. or p.m., which means midnight at the beginning of the day or noon, repsecitvely. A computer may give the times of future events. An event may be scheduled for either of these two times. Even further, one may at times wish to set a digital clock (including system clocks in computers) to one of these times. In all of these cases the quoted paragraph needs some help. The paragraph (perhaps inadverntly) gives the actual reason that the de facto convention is what it is, but doesn't complete the thought. The new cases I listed are the "invisible instants". Just because they're invisible on a clock, doesn't mean they can't be used as time stamps, times in a schedule, or times used to set a clock. So someone should rewrite this paragraph to "complete the thought".

Re self-contradiction: You are applying the traditional definitions out of their scope. That is what I meant by the article contradicting itself. Since the terms do not pin down the times for noon and midnight, you cannot use them as you did for 12 p.m. midnight, esp. since your use of them was selective and subjective. That violates NPOV. Yes, the article "continue[s] to explain how the terns [terms] are nevertheless used in an extended way in practice", but that doesn't mean that the traditional definitions can legitimately be used subjectively and selectively as you have done.

Re the airline: I dodged slightly, but only to make another point. I seriously doubt that any airline uses the Gov't Printing Office (GPO) convention. In fact, I doubt that anyone uses it (save perhaps for the GOP istelf).

I'll take on your question directly now. You are assuming you are right about it meaning EOD. So you are really saying you're right because you said you're right. That's a circular proof. The fact that I'd have to wait if you're right doesn't mean that you're right!

I still claim that the reference does not say EOD and that you are applying a subjective interpretation. My point is that since the GOP is illogical about noon, what's to say it is not being similarly illogical about midnight? Nothing! It's not BOD, it's not EOD, it's ambiguous and the article should reflect that.

Re 9.54 being the only mention: Not true. The article says more than just section 9.54. Go to the actual 12-hour article in W and click the GPO link in the table and you'll instead get this (which I already quoted in this discussion!):

 12.9. Units of measurement and time, actual or implied, are

expressed in figures.

 a. Age:

[...]

 b. Clock time (see also Time):
     4:30 p.m.; half past 4
     10 o'clock or 10 p.m. (not 10 o'clock p.m.; 2 p.m. in the 
       afternoon; 10:00 p.m.)
     12 a.m. (noon); 12:15 p.m. (15 minutes past noon)
     12 p.m. (midnight); 12:25 a.m. (25 minutes past midnight)
     4h30m or 4.5h, in 
       scientific work, if so written in copy
     0025, 2359 (astronomical and military time)
     08:31:04 (stopwatch reading)

You say that BOD is not a valid interpretation. Then how to explain this part:

  12 a.m. (noon); 12:15 p.m. (15 minutes past noon)
  12 p.m. (midnight); 12:25 a.m. (25 minutes past midnight)

In the first line, both times refer to the same noon. Are you now going to tell me that in second line -- in the very same line as itself, no less -- that two different midnights are being referred to? I think the most logical interpretation from this line would be that it means midnight BOD. That would also put the times in chronological order! To be truly neutral -- in light of your argument -- one can only say it's ambiguous. The only way to find out for sure is to contact the GPO and point out their folly and ask how they are going to resolve it.

Also, you're trying to make sense out of something that is simply illogical and nonsensical in the first place. You may as well try to apply logic to astrology. By attempting to repair the GPO style, you're violating NPOV.

I suggest that you explain your reasoning in the article, and please include the reason you reject midnight BOD, but not 12 a.m. noon, even though both are invalid on the same grounds, and how that doesn't violate NPOV. AEF 71.127.218.78 15:12, 14 October 2007 (UTC)

Reply to Zginder: I don't mean it's twice; I mean it's ambiguous and the table should reflect that. It is a NPOV violation because (1) the traditional definitions of a.m. and p.m. are being used out of their scope (the scope being all times other than noon and midnight) and (2) they are being used selectively, subjectively (to midnight and not noon). They are being used to reject one invalid point (midnight BOD) but not the other (noon). Since that is subjective, it violates NPOV. AEF 71.127.218.78 15:12, 14 October 2007 (UTC)

While we're on a nitpick-fest, "readout" is not hyphenated and the "one-minute periods _following_ [my emphasis] the instants of noon and midnight" actually include the times 12:01 a.m. and 12:01 p.m., respectively. Therefore, it should be rewritten to reflect this fact.

Alan E. Feldman 2007-09-14 03:44 UTC

Sorry to bother you all again, but I missed an important point: When the clock says 12:00 it means that the truncated time is 12:00; that is, the time is 12:00:xx.xx... where each 'x' can be any of the standard Arabic numerals {0,1,2,3,4,5,6,7,8,9}. Therefore, at the first instant the clock shows 12:00, it is indeed 12:00. This is true of all possible times displayed on the clock: the first instant that the particular time is displayed is the time that it is exactly that time (according to the clock, of course!). When you set a clock to 3:00 p.m., for example, you don't set it to the point between 2:59 p.m. and 3:00 p.m., basically because you can't! You set it to 3:00 p.m. This proves my point. Besides, it takes a finite (non-zero) amount of time to switch from one time to another. So at exactly what point during this transition is it supposed to be exactly the time of the next displayed time? This shows that the last sentence in the quoted paragraph is wrong and should be omitted.

Alan E. Feldman 2007-09-14 03:59 UTC

I really don't see how a time on a digital clock only refers to the 60-second interval after the time in question. The purpose of a clock is to show the time. So if the time is 3:45, shouldn't the clock show 3:45? Doesn't that make sense? And if it shows 3:45 only the after the instant of 3:45, how are you going to explain that to the kiddies?

Alan E. Feldman 2007-09-21 02:48 UTC

Thanks, Alan. I added that. With your encouragement, I keep improving my faulty wording. I'm going to work on it some more right now. Korky Day 13:08, 22 September 2007 (UTC)

I hope my newest revision answers all your objections except one. Please tell me if it doesn't.

The exception is that the flipping might take a finite amount of time to occur, which you say is a problem. I don't think so. If the flipping does take a finite period, someone simply has to define which instant during that process is the instant desired. For example, using an old-fashioned mechanical digital clock, an actual little card with a number on it flips over on a circular wire, like a Rollodex (spelling?). We could define that the instant the card hits the previous card in the pile is the exact time you want. With all-electronic digital clocks, I don't think there is any finite gap of time between the displays of 11:59 and of 12:00. If I am wrong and there is a gap, then you can use the same logic as with a mechanical digital clock: assume that a particular instant in the process is the instant you're looking for. That might be the instant that the 12:00 light comes on, for instance. Korky Day 13:51, 22 September 2007 (UTC)

The new version of the asterisked paragraph in question is much better. It is clear now that it is saying that the digital clock does what it does out of practicality but emphasizes the point that the "de jure" convention (which forbids use of a.m. or p.m. at noon and midnight) is different from what the digital clock shows, which is the "de facto" convention (which says that 12:00 a.m. denotes 00:00 and 12:00 p.m. denotes 12:00). OK. And never mind my points about the time it takes for the clock to change from one displayed time to the next, now that I understand Korky's point better.

Alan E. Feldman 2007-09-24 00:50 UTC.

Re the Gov't Printing office entry of the style table: Woodstone says that p.m. cannot logically be at the beginning of the day. Why, then, can a.m. be noon in the middle of the day? That is just as illogical. The fact is that the term midnight is ambiguous and there is nothing in the reference to say whether it means midnight at the beginning or end of the day. If it's 00:10 one can well say it's 10 minutes past midnight which obviously refers to midnight at the beginning of the day. In "We close at midnight tonight" it obviously means midnight at the end of the day. To choose one over the other is not being neutral. In fact, there is insufficient information to decide the question. This should either be explained in the table or article, or, the entry should simply be removed. ... Its style is totally illogical. Please explain to me the logic that forbids 12 p.m. at the beginning of the day but allows 12 a.m. to be noon. —Preceding unsigned comment added by 72.68.12.151 (talk) 23:45, 8 October 2007 (UTC)

Zginder: Please explain your cryptic statement about reverting the Gov't Printing office entry. Woodstone: Please explain your logic. Thanks! Alan E. Feldman 72.68.12.151 01:45, 9 October 2007 (UTC)

The abbreviation p.m. means "after the middle of the day". How can the beginning of a day be after the middle? Midnight can be defined as the end of the day before it as 12:00 p.m. or as the beginning of day following it as 12:00 a.m. That in itself is confusing enough, since the notation 12:00 p.m. is most often used to stand for noon. −Woodstone 07:41, 9 October 2007 (UTC)

Woodstone: You pretty much just made my case for me. You are applying your logic selectively only to midnight. This reference is obviously violating that logic for noon, so why do you think it should apply at midnight? The fact is that the reference is ambiguous as to what it means by "midnight" because it only calls it "midnight" which is clearly ambiguous. And if you apply the context of the reference it looks a lot more like the beginning of the day than the end. Let me quote it here, please:

     12 a.m. (noon); 12:15 p.m. (15 minutes past noon)
     12 p.m. (midnight); 12:25 a.m. (25 minutes past midnight)

The phrase "25 minutes past midnight" means that the time is 00:25 and therefore refers to the midnight at the beginning of the day. So by using logic based solely on what's actually in the reference, I come to the opposite conclusion. You are imposing your interpretation on the reference with no justification. It is a clear violation of Wikipedia's neutrality policy. The entry should be fixed accordingly, perhaps by adding a fourth column called "midnight". --Alan E. Feldman 71.125.91.108 02:32, 11 October 2007 (UTC)

The argument about noon is quite different. Noon is strictly speaking neither a.m. or p.m. Midnight on the contrary is both a.m. and p.m., depending on which side of it you are looking from. It only becomes clear if you combine it with a date. If you had an airplane ticket for October 11 at 12:00 p.m., would you go to the airport in the evening of October 10? You would have to wait a long time, because the flight is after the noon of October 11. I cannot see any other sensible interpretation of the Printing Office's standard. −Woodstone 13:01, 11 October 2007 (UTC)

Very short version: You can't apply one convention to another!!!

Short version: You are applying your convention to disambigufy the ambiguity of midnight in the Gov't Printing Office convention. Since the whole point of the table is to compare conventions, you cannot do this. Furthermore, you applied your convention only to the entry you didn't like. You should have also applied it to 12 a.m. being noon to be consistent. Since you were not consistent, this constitutes bias and therefore violates Wikipedia's neutrality policy.

Long version: I'll quote you and respond.

You: The argument about noon is quite different.

Me: You missed the point. In both cases you are imposing external ideas (namely, another convention) onto the Gov't Printing Convnetion (GPC).

You: Noon is strictly speaking neither a.m. or p.m. Midnight on the contrary is both a.m. and p.m., depending on which side of it you are looking from.

Me: Already there is a problem. The whole point of the table is to show the different conventions as defined by different entities. Here you are already declaring what a.m. and p.m. are. That's a defnition (or convention)! And it's only one definition. The article shows that there are others.

More specifically, since you declared that noon is neither a.m. nor p.m. you should reject the 12:00 a.m. entry just like you reject the 12 p.m. is midnight at the BOD (beg. of day) entry. Since you accept one, but not the other, you are being inconsistent and therefore biased. This is a clear violation of Wikipedia's neutrality policy.

You: It only becomes clear if you combine it with a date. If you had an airplane ticket for October 11 at 12:00 p.m., would you go to the airport in the evening of October 10? You would have to wait a long time, because the flight is after the noon of October 11.

Me: No, I would try to arrive at the airport about 1000 or 1030 as I'd expect the flight to leave at 1200. 12:00 p.m. is noon. Airlines don't use the Gov't Printing convention. Why do you think they would? Since the airlines use computers, it should not be surprising that they use the digital clock convention. Actually, at least one uses 12:00N for noon. I doubt that ANY use 12:00 p.m. for midnight! Check some airline schedules on-line. I don't think you'll find any flights that take off at midnight anyway and if you do it will most likely be clear either from proper use of the de facto convention, use of 11:59 p.m., use of 12:01 a.m., use of 12:00N (to signify noon), or it will be clear from the context.

And if your plane ticket said midnight, how would you know which midnight? The Gov't Printing convention has the same problem because its midnight is also ambiguous.

You: I cannot see any other sensible interpretation of the Printing Office's standard.

Me: There is no unique sensible interpretation. I presented a perfectly "sensible" interpretation in my last post which places 12 p.m. as midnight at BOD. Can you please tell me how that is wrong? If you can say 12:25 a.m. is 25 minutes past midnight, then that midnight is BOD. That implies that midnight can be at BOD in the GPC. What is wrong with that? Midnight in the GPC is ambiguous. I can hear it now: "But Alan, p.m. means after noon." But Woodstone, that's according to ***another*** convention. The whole point of the table is to show how different entities define things differently, including p.m. So if the Gov't printing office defines a.m. and p.m. differently, and it is quite evident that they do, then you cannot say "a.m. means before noon and p.m. means after noon" because the convention in question says a.m. can be noon itself. So it is certainly not out of the question for p.m. to be something other than after noon, at least for noon or midnight. IOW, their a.m. already violates the after/before stuff, so what reason can you give to say that their p.m. doesn't, especially since I showed a perfectly reasonable way in which it can?

The point of the table is to show different conventions. The Gov't Printing convention is ambiguous when it comes to midnight. It is not right or neutral to apply another convention in an attempt to "disambigufy" it.

Why is it a problem to have 12 p.m. for both BOD and EOD but it's perfectly fine to have midnight for both in some of the other entries???

I'm sorry to have to write at such length, but I haven't been able, so far, to get my points across to you. And you don't address them either. So when you respond, please address my points. Thanks. Alan E. Feldman 71.127.215.237 01:32, 12 October 2007 (UTC)

P.S. One more thing, Woodstone: The article itself says: "The terms ante meridiem (before noon) and post meridiem (after noon) do not apply literally to noon and midnight." But you are applying them to midnight. Therefore you are contradicting the very article you are contributing to!

Everyone agrees what a.m. and p.m. mean for all times other than 00:00 and 12:00. The whole point of the table is to show what different entities use or recommend or declare or whatever for how to treat these two times. If a reference like the Gov't Printing Office Style Manual is ambiguous about what it means by the term "midnight", it is not justified to apply other conventions to clear up the ambiguity, namely because they are other conventions. It is also not justified to use the "before noon" and "after noon" ideas to clear up the ambiguity because doing so contradicts the article itself!

--Alan E. Feldman 71.255.81.235 10:11, 12 October 2007 (UTC)

In several places, the article and my comments use a phrase like "stricly speaking" or "literally" and then continue to explain how the terns are nevertheless used in an extended way in practice, You may call that "contradicting" oneself, but it is just a way to describe reality.

On the point of the airline you dodge the point. The problem does not disappear by ignoring it. If the printing office would promise you to deliver a print at the stated day and time, you would face the same long wait.

The actual standard as referenced says only this:

  9.54. References to meridian in statements of time are 
abbreviated as follows:

10 a.m. (not 10:00 a.m.)
2:30 p.m.
12 a.m. (formerly 12 m.) (noon)
12 p.m. (midnight)

So it does not say that 12 p.m. on October 11 is at the beginning of that day. That is your interpretation. My point is that that is not a valid interpretation. The time October 11 at 12 p.m. is not the same as October 10 at 12 p.m. The question is what should be the meaning of each. The standard is not explicit. So we must look for an interpretation. In my earlier comments I have made clear that the only interpretation that makes sense is that October 11 at 12 p.m. (according to the U.S. Printing Office) is midnight at the end of October 11.

−Woodstone 12:57, 12 October 2007 (UTC)

Woodstone is right about this one, I do not think that the USGPOSM is saying that 12:00 p.m. excites twice! This is the U.S. Government after all, You would not government official confused as to when an event should happen. This is also not a NPOV violation and I was the one that labled this very same article POV a few years ago.Zginder 00:03, 13 October 2007 (UTC)

Reply to Woodstone: Re self-contradiction: You are applying the traditional definitions out of their scope. That is what I meant by the article contradicting itself. Since the terms do not pin down the times for noon and midnight, you cannot use them as you did for 12 p.m. midnight, esp. since your use of them was selective and subjective. That violates NPOV. Yes, the article "continue[s] to explain how the terns [terms] are nevertheless used in an extended way in practice", but that doesn't mean that the traditional definitions can legitimately be used subjectively and selectively as you have done. (I admit there is logic to your argument that a (traditionally defined) p.m. time has to be after 12 noon, but to do that is to apply the traditional definitions selectively and out of their traditional scope. And since it is possible to use other logic to have 12 p.m. midnight mean BOD, the reference is ambiguous as to BOD vs. EOD.)

Re the airline: I dodged slightly, but only to make another point. I seriously doubt that any airline uses the Gov't Printing Office (GPO) convention. In fact, I doubt that anyone uses it (save perhaps the GPO istelf).

I'll take on your question directly now. You are assuming you are right about it meaning EOD. So you are really saying you're right because you said you're right. That's a circular proof. The fact that I'd have to wait if you're right doesn't mean that you're right!

I still claim that the reference does not say EOD and that you are applying a subjective interpretation. My point is that since the GOP is illogical about noon, what's to say it is not being similarly illogical about midnight? Nothing! It's not BOD, it's not EOD, it's ambiguous and the article should reflect that. So EOD is not the only interpretation that makes sense. BOD makes just as much sense in this case.

Re 9.54 being the only mention: Not true. The article says more than just section 9.54. Go to the actual 12-hour article in W and click the GPO link in the table and you'll instead get this (which I already quoted in this discussion!):

 12.9. Units of measurement and time, actual or implied, are

expressed in figures.

 a. Age:

[...]

 b. Clock time (see also Time):
     4:30 p.m.; half past 4
     10 o'clock or 10 p.m. (not 10 o'clock p.m.; 2 p.m. in the 
       afternoon; 10:00 p.m.)
     12 a.m. (noon); 12:15 p.m. (15 minutes past noon)
     12 p.m. (midnight); 12:25 a.m. (25 minutes past midnight)
     4h30m or 4.5h, in 
       scientific work, if so written in copy
     0025, 2359 (astronomical and military time)
     08:31:04 (stopwatch reading)

You say that BOD is not a valid interpretation. Then how would you explain this part:

  12 a.m. (noon); 12:15 p.m. (15 minutes past noon)
  12 p.m. (midnight); 12:25 a.m. (25 minutes past midnight)

In the first line, both times refer to the same noon. Are you now going to tell me that in the second line -- in the very same line as itself, no less -- that two different midnights are being referred to? I think the most logical interpretation from this line would be that it means midnight BOD. Clearly 12:25 a.m. is referring 25 minutes after midnight BOD. That would also put the times in chronological order! To be truly neutral -- in light of your argument -- one can only say it's ambiguous. The only way to find out for sure is to contact the GPO and point out their folly and ask them how they are going to resolve it.

Also, you're trying to make sense out of something that is simply illogical and nonsensical in the first place. You may as well try to apply logic to astrology. By attempting to repair the GPO style, you're violating NPOV and still end up with a nonsensical convention. --AEF 71.127.218.78 15:13, 14 October 2007 (UTC)

Reply to Zginder: I don't mean it's twice; I mean it's ambiguous and the table should reflect that. It is a NPOV-violation because (1) the traditional definitions of a.m. and p.m. are being used out of their scope (the scope being all times other than noon and midnight) and (2) they are being used selectively, subjectively (to midnight and not noon). They are being used to reject one invalid point (midnight BOD) but not the other (noon). Since that is subjective, it violates NPOV. --AEF 71.127.218.78 15:13, 14 October 2007 (UTC)

Nothing wrong with their statement. If they define 12 p.m. as midnight, then 12:25 a.m. is still (just like for any other convention) 25 minutes past midnight. They just do not state the dates going with that. It would be the midnight at the end of one day and 25 minutes past the same midnight regarded as the beginning of the next day.

The essential question to answer is: if the US Print Office uses their convention to express midnight in combination with a date, which one is meant, relative to that date, the begin or the end. As explained many times above there is only choice that makes sense. That their convention overall does make little sense is a separate issue.

You keep saying that the argument is the same for noon, but, I must insists that the reasoning their is quite different.

−Woodstone 22:05, 14 October 2007 (UTC).

You could interpret it as midnight EOD, but you could also interpret it as midnight BOD. But according to you, both 12 a.m. noon and 12 p.m. BOD are invalid. But you reject only the latter. How does that not violate POV? (A question you still have not answered.) Why does it matter if the arguments are not exactly the same? They have the same exact root: the traditional definitions of a.m. and p.m. Please tell me why one applies and not the other. Don't tell me yet again that EOD is the only choice that makes sense -- I got that part; tell me why it's okay not to reject 12 a.m. noon even though it, too, makes no sense.

Please don't repeat your assertion that they're different. Tell me why one applies and the other doesn't. Saying they're different does not answer that question. I don't see how the difference is relevant, esp. as they have the same exact roots. Please explain how and why.

Why does it matter that, according to you, EOD is the only one that makes sense? You say: "That their convention overall does make little sense is a separate issue." What matters is what the reference is saying. If they're saying something that doesn't make sense, either it should be reported as is or omitted entirely. Anything else is biased POV. You're changing their "style": you're putting words in the GPO's mouth: you're unjustifiably clearing up an ambiguity in their style. And that's clearly not a NPOV.

Please stop dodging my questions. Thanks! --AEF 72.68.9.118 00:46, 15 October 2007 (UTC)

I changed 24-hour clock in the table to 24-hour format because most 24-hour clocks never display 24:00. The term "24:00" is for event times and schedules. It is part of the 24-hour format or style, but is not normally part of its clock. Yes, at least one clock shows that (I'm pretty sure that's an expectional case), but it must then show it for a minute, in which case it is NOT ISO-8601, which is also mentioned in the same row! I don't know why it turned red when I changed it -- change the color back if you like -- I don't care. But clock is an inaccurate term for that row. No clock can show both on one display. AEF 71.127.220.58 23:25, 22 October 2007 (UTC)

The name of that system is "the 24-hour clock". It counts through all the 24 hours from 0 to 23. The link turned red because the article "24 hour format" does not exist. −Woodstone 07:37, 23 October 2007 (UTC)

end point convention -- meaning?

"end point convention" (last column of chart) is never defined. I have no idea what it means. Korky Day 18:15, 13 October 2007 (UTC)

Zginder: What is the reason for adding the endpoint column? And why is Antiquated "neither", why is legal "neither", and what's the difference between emdash and neither? Thanks! AEF 71.127.218.78 15:16, 14 October 2007 (UTC)

Table of Time Values

Is there a need for the P.M. numbers to be complete, when the A.M. section is abbreviated with ellipses? I would like to have a uniform and concise table, and one that would be uniform (if possible) with the chart that would replace the one found at 24-hour clock...--Vox Rationis 00:30, 2 February 2007 (UTC)

This table is ridiculous. I'm sorry to be so blunt, but it really is. How many rows do we need to get the idea? Where is the reference?!:-D) What is "undefined (midnight)*" supposed to mean? And an explicit list of the numbers for the hours in the text along side it: 1, 2, 3, ... ? Are we teaching readers how to count? --betaneptune —Preceding unsigned comment added by 71.125.90.46 (talk) 00:46, 9 June 2009 (UTC)

Criticism and practical problems

This section of the main article starts with "Many people who grew up with the 24-hour clock see the 12-hour notation as a less practical and outdated convention". What does this actually mean? You could grow up in many places not seeing a Fahrenheit thermometer or an inch rule, but is it being suggested that some people have never seen an analogue clock face? My point is that most people use 12hour time to read an analogue clock & 24hour time to read a digital clock. The only time when I would do otherwise is if speaking to someone abroad, I would tell them the time as I think they would understand it best. If I'm stopped & asked the time, I look at my analogue watch & tell the person what I see! If the hand points to 3, I say "3 o'clock" & I expect the person asking to know if it's am or pm. Dyaimz 22:41, 19 July 2007 (UTC)

What if it's an 12-hour digital clock? What if it's an am/pm 12-hour digital clock?

What if you see an event listed as 10:00. Is that a.m. or p.m.? Sometimes you will be able to tell by context; often you cannot.

Alan E. Feldman 2007-09-14 3:25 UTC

Well, I grew up without a.m./p.m. clocks which means I'm not used to them which in turn means that: I am confused by them, have to recall the meaning of a.m./p.m. every time, can never remember whether 12 a.m. is midnight or noon, hate it to have to read the time twice because only when I stumble upon those strange abbreviations I know what part of the day I have to put the previous numbers into and yes, I do consider the 12-hour notation impractical and outdated for written time (while intuitive for analogue clocks of course). I just can't help finding a system that starts with 12, then resets one hour later to 1, then resets again 12 hours later to 1-but-a-different-1 and ends at almost-12 not the most logical way of counting from 0 to 24. And this mixing up of more or less international digits with some particular language (Latin/English in this case) isn't that elegant either in my opinion. Anyway, I don't even want to discuss the advantages and disadvantages of having the 12-hour system for analogue and the 24-hour system for written time compared to having the 12-hour system for both. I just wanted to provide some source for the sentence from the article. Although I'm not "many" of course ;) --194.97.120.203 (talk) 14:26, 31 May 2008 (UTC)

Endpoint convention and ambiguity columns

I find the added columns about endpoint convention and ambiguity more confusing than elucidating. The concepts are not wrong, but the term "end point convention" must be considered original research (no relevant hits on Google). It would be better to stick to only a textual explanation of the concept and not mix it in the table. The ambiguity is only real for the term "midnight" itself. −Woodstone 20:18, 15 October 2007 (UTC)

"endpoint convention" has 156 hits and "end point convention" has 99 hits on Google. I did not invent the term. Zginder 22:36, 15 October 2007 (UTC)

But what is the point of it? What does it add to the table? AEF 71.255.87.227 00:24, 16 October 2007 (UTC)

I was talking about relevant hits. Try {"endpoint convention" p.m.}. Only 9 hits (3 with added space), including an echo of this article. Not significant. But even the 156 mentioned above is rather low. −Woodstone —Preceding comment was added at 09:49, 16 October 2007 (UTC)

I think we should remove the endpoint section as we don't have endpoint in the table anymore. AEF 71.127.220.58 23:19, 22 October 2007 (UTC)

I removed the "End point convention" section because it is confusing and almost certainly the brilliant insight of the author (a.k.a. original research, which is verboden). The use of calculus as justification is dubious: the note referring to the necessity of continuity on a closed interval for various theorems is false (even if it is replaced by the slightly more correct "closed, bounded interval"). But, most of all, it is tangential to the section; maybe it is worthy of a sentence therein if someone cares to make a more nuanced edit, but certainly not a whole subsection. Leberquesgue (talk) 23:54, 9 December 2008 (UTC)

Truncated time

The notion that clocks show a truncated time representation is in my view the cleanest way to get around the unclarity at midnight. When the actual time is 12:00:00.001 a.m., a 12-hour digital clock in minutes shows just 12:00 a.m. Of course there is an infinitesimal instant where neither p.m. nor a.m. is (should be) shown, but we would never notice that. The talk about endpoint convention never makes this point very clear. −Woodstone (talk) 10:39, 26 December 2007 (UTC)

Yes, the fact that clocks show the truncated time is the reason 12:00 p.m. is noon and 12:00 a.m. is midnight at the beginning of the day. But your comment about the instants of noon and midnight not being a.m. or p.m. causes problems. Why all the fuss over these "instants of 12:00 a.m. and p.m." in the first place? The entire shmear of telling time is a human-made thing, and being a creation of humans is subject to change by humans. Digital clocks have pretty much done that. The "standards organizations" have already done that with the definitions of the second, the meter, and Universal Time itself upon which civil time is based! And these definition changes were done with good reason. And the dictionary doesn't get in their way: "We're not letting any stupid dictionary tell us how to define units of time, space, and so on!" But for a.m. and p.m. they weasel out and go with whatever dictionary they had at hand says. And civil time used to be local solar time but for convenience reasons was changed to Standard Time based on GMT (or UT), and later, UTC. So everything else about telling time has changed. So why such intense resistance to a.m. and p.m. being redefined as given by digital clocks?

Anyway, the problem is when you work with computers. If you want to input midnight, you need to specify 12:00 a.m. For "noon" you need to input 12:00 p.m. If you're using a computer that is set to use the 12-hour clock, you need to know the digital clock convention. Also, what is the point of including the quote from the National Maritime Museum? The quote doesn't even apply to civil time as explained by the paragraph immediately below it (which refers to another page from the same Web site, no less!). Now it depends on whether you go with "noon" or the crossing of the celestial meridian by the sun as your "noon" reference point for ante and post meridiem. If you go with noon, the quote makes no sense. If you go with the celestial meridian, you still have the problem that civil time is not the same as local solar time, which pretty much ruins the original point. So why do we have this quote in the article, then? betaneptune 71.255.89.104 (talk) 04:04, 21 January 2008 (UTC)

Yet another reason to stop this fuss is that no clock is exact anyway. So even if you do somehow label the instants of noon and midnight without using a.m. or p.m., the clock will be a little off anyway. At exactly noon, any reasonably accurately set clock will show either 11:59 a.m. or 12:00 p.m. Even atomic clocks are not "exact" (except that they are used as part of the definition of UTC, and even then, it is an average of specific clocks, and those clocks almost certainly use the 24-hour format anyway). betaneptune 71.251.41.51 (talk) 15:59, 21 January 2008 (UTC)

Time Measurement and Standards Topics Template

(Reduplicating post of mine from User talk:Woodstone:) Sorry if I stepped on any toes, but I was only removing "See Also" listings from where they duplicate links on the template. Many articles have excessively long See Also lists, that append without adequate explanation, and templates seek to address that.

Specifically regarding 12-hour clock and 24-hour clock articles, references to each are in the body of either article, making the See Alsos redundant, even without the Time MEasurement Template. Also, in the 12-hour clock article, Comparison of the 12-hour and 24-hour clocks is linked twice in the article above, in appropriate places, and should probably not be in the See Also list a third time.

In any case, no page has been left unlinked where it was linked before. I'm very careful about that. Cheers. -- Yamara 16:05, 11 January 2008 (UTC)

Confusing explantion in the "Confusion at noon and midnight" section

This section has the following text:

"The moment when the Sun is in fact at its highest point in the sky on a particular day, is noon by apparent solar time, also known as true time. This occurs not exactly 24 hours after the previous true noon. Clock time (mean solar time) is based on the average[7] interval in the course of the year between one true noon and the next.[8]"

Well, for one thing, the first comma should be omitted!

This paragraph fails to mention the effects of time zones and daylight time, both of which have, on average, even larger effects on the (civil) time at which the sun crosses the celestial meridian. Local civil time, in the vast majority of cases, is an integral number of hours offset from UTC. And even UTC is not really the same as Universal Time (UT1) (which is the old definition of GMT, and UT1 is corrected for polar motion, making it slightly different from "mean solar time" [UT0]) as it can differ from that by as much as 0.9 seconds. So this paragraph is highly incomplete and misleading and therefore needs to be fixed. If we didn't have the quotation from the National Maritime Museum (which is incorrect as shown by a previously included quote reference to another page in the same site! [It is incorrect because it implicitly equates civil time with apparent solar time]) then the explanatory correcting quote quoted above wouldn't be needed. So either the correction needs to be expanded to be reasonably understandable and complete or both the incorrect quote and the correction should be removed from the article. I prefer removal but if the incorrect quote must stay, a much more satisfactory correction is needed. This entire bit has come from an attempt to tie a.m. and p.m. to when the sun crosses the meridian but because of all these things the sun can cross up to approx. 2 hours or more "later" then civil noon, making the whole point about crossing the meridian somewhat archaic. 71.127.218.88 (talk) 21:16, 13 January 2008 (UTC)

The paragraph has since been replaced twice. It's much better now but the whole deal with including the quote from the National Maritime Museum is somewhat pointless (see the Truncated Time section above). betaneptune 71.251.41.51 (talk) 16:18, 21 January 2008 (UTC)

Origins of noon = pm / noon = am conventions

I removed the false attribution:

In the United States, largely because of the preponderance of digital clocks and computers, which change from a.m. to p.m. (and vice versa) when changing the hour from 11 to 12, noon is often called "12:00 p.m." and midnight "12:00 a.m.", as at the beginning of a day.

I heard from a reliable personal source that the origin of the 'noon = pm' convention was the US military, around the time of WWII. Perhaps someone could confirm/clarify the origins of this convention. And is the reverse (GPO) convention confined to that institution? 202.64.168.196 (talk) 09:26, 15 April 2008 (UTC)

That sounds a lot like original research. Wikipedia does not allow original research. See WP:OR. Z gin der 2008-04-15T14:23Z (UTC)

informal speech

I'm in the US, and my family and community (in Buffalo, NY) has always used "a quarter to six", and never "a quarter of six." 74.78.98.109 (talk) 18:00, 8 June 2008 (UTC)

I'm in the US, from Minnesota and a transplant to California and I've never heard "a quarter of six". It's always been "a quarter to six", in both locations. 71.142.176.144 (talk) 18:00, 24 November 2008 (UTC)

Well I'm in the US too (I live in DC and NY), and everyone in these areas says "quarter of" not "quarter to." Why doesn't the article just mention BOTH expressions. —Preceding unsigned comment added by 71.246.193.26 (talk) 03:14, 15 November 2009 (UTC)

Citations needed

The article is tagged as needing citations/references, which is reasonable, and there are also {{fact}}-tags here and there, but their distribution seems less reasonable - actually quite random. Given the general tag at top, should the fact tags flag only controversial or particularly dubious claims, or everything that ideally ought to be sourced somehow?--Noe (talk) 06:35, 12 June 2008 (UTC)

five-to-half

In Sweden it is common to use the terms "five-to-half" and "five-past-half" when referring to :25 and :35 respectively. E.g. 6:35 "five past half seven". Would that be of value to add to the article? I'd add it but there might be other countries that use the same notation. —Preceding unsigned comment added by Poposhka (talk • contribs) 22:08, 4 December 2008 (UTC)

conversion table 12 vs 24 hour

I find the addition of several 01 and 59 minute times around the tricky transitions clarifying. However they make the table too long. How about removing some integral hours in the regular sections. For example suppress 2:00 till 10:00 (and possibly 14:00 till 22:00)? −Woodstone (talk) 21:51, 31 March 2009 (UTC)

Bottom of the "Confusion" Table

It says, "Digital clocks and computers, when set to the 12-hour system, appear to show the times 12 a.m. and 12 p.m., as in this chart." Appear? They DO show 12 a.m. and 12 p.m. Actually, they usually show 12:00 a.m. and 12 p.m., with the a.m. and p.m. usually being implied by a dot which is lit for PM and dark for AM. Please, "appear"? That's like saying the sun appears to rise in the east. --betaneputne —Preceding unsigned comment added by 71.125.90.46 (talk) 00:35, 9 June 2009 (UTC)

I agree with you 100%, you are absolutely correct. They most certainly do and there hasn't been anything shown to the contrary. This content is definitely incorrect. I will go ahead and remove it. Thanks much betaneputne for pointing this out. JackOL31 (talk) 03:00, 15 December 2009 (UTC)

Your edit is still not valid. All 12-hr clocks with am/pm indicators show am/pm as the chart indicates. Therefore, you would need to list all types of clocks. Otherwise the statement implies that some do not, which is incorrect. If you disagree, please discuss. JackOL31 (talk) 12:16, 15 December 2009 (UTC)

In the spirit of compromise, I will agree to Woodstone's edit. The entry certainly seems to be redundant an unnecessary, though. JackOL31 (talk) 12:54, 15 December 2009 (UTC)

Let me explain my confusion. Saying that a clock set to the 12-hour mode follows the 12-hour system seems unnecessary. It's a bit like saying cars put into drive gear will go forward. If you really take a hard look at that table, are you really contributing to the entry with previously unknown information? JackOL31 (talk) 13:32, 15 December 2009 (UTC)

I saw your statement below, The statement for usage of a.m./p.m. at 12 o'clock is only meaningful for digital clocks (not analog ones). Mentioning that specifically is useful information in view of the ambiguity. I don't quite see it that way. As I mentioned below, some analog clocks and watches do keep track of am/pm, depending on the model. Regarding your concern about ambiguity, the table entry states, "12-hour digital clocks with a.m. and p.m." and lists the values 12 a.m. and 12 p.m.. The footnote says digital clocks display 12 a.m. and 12 p.m. as shown in the chart. I don't see the value added.(Sorry for all the replies - I recently realized you were addressing both issues in the below section where I was addressing them separately) JackOL31 (talk) 13:54, 15 December 2009 (UTC)

I was not intending to address this item again, but after seeing the last section I need to reopen this discussion. The following lists an entry in the last table, the footnote for that table and the last section in the article:

12-hour digital clocks with a.m. and p.m. * 12:00 a.m. 12:00 p.m.

Digital clocks and computers, when set to the 12-hour system, show the times 12 a.m. and 12 p.m., as in this chart.

If a 12-hour time format is set, Microsoft Windows and Office applications denote noon by "12 pm" and midnight by "12 am". This convention is also followed in Apple's Mac OS X operating systems.

In summary, the first line above says 12-hour digital clocks display in this manner. The footnote says digital clocks and digital clocks on computers display in the manner described mentioned previously. The last section says digital clocks display in the manner described earlier but repeated again. I do not see the value added by essentially saying the same thing over and over again. Unnecessary repetition is not a good thing. Slight differences in wording do not make them different concepts. Unless a good reason can be provided why this need to be said 3 times, the middle duplication (triplication) needs to be deleted. JackOL31 (talk) 02:09, 19 December 2009 (UTC)

^ National Institute of Standards and Technology's Physics Laboratory, Time and Frequency Division

[1] National Institute of Standards and Technology's Physics Laboratory, Time and Frequency Division

[1]