Talk:12-hour clock

From Wikipedia, the free encyclopedia
Jump to: navigation, search
Article Collaboration and Improvement Drive This article was on the Article Collaboration and Improvement Drive for the week of October 31, 2007.
WikiProject Time (Rated C-class, High-importance)
WikiProject icon This article is within the scope of WikiProject Time, a collaborative effort to improve the coverage of Time on Wikipedia. If you would like to participate, please visit the project page, where you can join the discussion and see a list of open tasks.
C-Class article C  This article has been rated as C-Class on the project's quality scale.
 High  This article has been rated as High-importance on the project's importance scale.
edit·history·watch·refresh Stock post message.svg To-do list for 12-hour clock:

Here are some tasks awaiting attention:
  • Expand : add a second paragraph to the introduction
  • Verify : add citations
  • Wikify : fix references so they include complete citations
Priority 3

Confusion at noon and midnight - table - midnight, end of day, for clocks[edit]

Should each "digital clock display" row in the table in the "Confusion at noon and midnight" section show an entry for both midnight start of day and end of day? Mitch Ames (talk) 01:11, 12 February 2011 (UTC)

Woodstone, you've removed the entries for "midnight, end of day" for "Digital 24-hour clock displays" and "12-hour digital clocks". I disagree with this. At midnight at the end of the day, none of my digital clocks go blank - they all display something. Just because midnight at the end of Monday is the same as midnight at the beginning of Tuesday doesn't make the former an invalid time. Just because my clocks don't know what day it is doesn't make "midnight end of day" just not happen. I suggest that it is appropriate to put the relevant values in these cells. Mitch Ames (talk) 10:07, 4 February 2011 (UTC)

As explained above, a physical digital 12-hour clock never shows midnight at end of day. It passes right through to begin of the next day. So the lines are rightfully empty. −Woodstone (talk) 14:01, 4 February 2011 (UTC)
As with the "clock system" table, our disagreement here appears also to be the purpose of the table. Again, the purpose of the table is not stated in the article, but perhaps it should be. Again, it might help if you answer these questions:
  1. What is the purpose of the table? Ie what exactly is it intended to show?
  2. As a reader of the article, I assume that those rows of the table tell me what 12- and 24-hour digital clocks show for the time indicated by each column. Is this assumption correct and/or reasonable? (If not, the article needs to address Q1 above.)
  3. Is "midnight end of day" a meaningful time? (eg I went out at 7pm on Saturday for dinner and a movie, and returned home at midnight the same day.) Presumably it is, because there's a column in the table for it.
  4. If yes, what would I see on most 12- and 24-hour digital clocks at that time?
  5. Why isn't the answer to Q4 in the table? (See Q1, Q2.)
I suggest that the table should show "12:00am (start of next day)" or "00:00 (start of next day)", as appropriate. Mitch Ames (talk) 03:28, 5 February 2011 (UTC)
My take on the questions:
  1. The table should show how times within the day are stated (in a document) or shown (on a clock). For convenience comparison is made to the internationally defined standard 24-hour clock system.
  2. The table should not only reflect clocks, also document usage.
  3. Many time periods have a natural extent, like a day, week, month or year. There is a case for using the end of such a period as demarcation (not the beginning of the next period).
  4. Clocks do not show periods, but are real time. They will have to choose. Normally clocks show the time truncated, changing at the beginning of their shortest display interval. It is not logical to show 24:00 (and exceedingly rare, certainly never on clocks that also show a date). In my view we can omit this rarity from the overview table.
  5. I still think it is better to leave the end of day blank for tyhe 12-hour clock, since it duplicates the first entry.
Woodstone 13:27, 8 February 2011 (UTC)
So, we appear to be agreed that:
  1. The table should show how times ... are ... shown on a clock
  2. The table should reflect clocks (as well as other things)
  3. There is a case for using the end of [a day] as demarcation – ie "midnight end of day" is a meaningful time.
Then we get stuck:
4. I ask: "What would I see on a clock at midnight end of day?" You neglected to answer the question explicitly - so I'll answer it for you: 00:00 (usually) or 24:00 (rarely). Note that all clocks show something - a display of some description is not a "rarity".
About the preceding paragraph, Isn't midnight end of day just a clarification? After all, how would any clock display "midnight end of day" for longer than a pico-second? And if no clock can display this for any length of time, why should a table entry reflect something that cannot be displayed for any actual length of time? Ohgddfp (talk) 04:20, 8 March 2014 (UTC)
5. I ask: "Given 1, 2, 3 (which we agree on), why isn't the answer to 4 in the table?" Your answer: "it duplicates the first entry". Assuming that you mean the first column, other rows also have the same value in both midnight columns, so I don't think that is sufficient grounds. Given 1, 2, 3, surely one must logically include the answer to 4 in the table.
Does anyone else have an opinion on this? Can someone point out any flaw in the logical progression that given 1, 2, 3, the answer to Q4 should be in the table? Mitch Ames (talk) 13:16, 9 February 2011 (UTC)
No clock can display "midnight end of day" for any length of time. After a pico-second, it would not be the end of that day. If the table is about what the clock can display, having an entry of a display that lasts for no period of time seems no different than having no entry at all. Ohgddfp (talk) 04:20, 8 March 2014 (UTC)
Clarification on Q4. A physical digital clock can show either the start or the end of the day, not both. So logically one of the lines should be empty. Clocks showing 24:00 are so rare that we don't need to mention them in the overview table. A passing remark in the section is enough. −Woodstone (talk) 17:18, 9 February 2011 (UTC)
You keep saying that the digital clock can show one or the other but not both - which is true, but not the point. In simple terms: I go out 7pm on Saturday for dinner and a movie, and return home at midnight the same day ("midnight end of day"). As I arrive home, I look at my 24-hour clock. What does it show? Answer: 00:00 (midnight start of next day). I still don't see why this answer can't be put into the "display at midnight end of day" column of the "what does the digital clock show?" row of the table.
Because the clock doesn't display "midnight end of day", it displays "00:00 midnight, start of next day", as you said.
We agree (Q1) that the table's purpose is to tell the reader what is displayed, we agree that I can refer to "midnight end of day" as a time, and presumably we agree that (most) clocks will show 00:00 (indicating the start of the next day) at that time - so what's the problem with putting that in the table?
If the table's purpose were to tell me what the time was when a clock was displaying 00:00, then you might have a point that there should only be one entry for that value (although one might argue that the "time" could be either midnight start of day next day or midnight end of day previous day) - but that is not what the table is apparently for.
I repeat my request for others' opinions, and/or explicit answers to my original Q4 (if anyone disagrees with my claim that the answer is 00:00 usually, 24:00 on some clocks), and or someone to point out the flaw in my argument. Mitch Ames (talk) 11:57, 10 February 2011 (UTC)
The point is that the 12-hour clock simply cannot denote or display the end of a day (but skips to the start of the next day instead). The p.m. is reserved for noon and a.m. is by definition before noon (not at the end of a day). So entering that 12 a.m. at end of day is misleading and contradictory. The same is true for almost all 24-hour digital clocks. The only use for 24:00 is in writing, typically only as the end of an interval. −Woodstone (talk) 16:47, 10 February 2011 (UTC)
Yes, and my point is that the table should show what a physical clock will display when I look at it at midnight end of day. Clearly we are going around in circles here. Does someone else have an opinion on the matter? And/or do you want to add some text to the article explaining what the purpose of the table is. Mitch Ames (talk) 11:41, 11 February 2011 (UTC)
There is no such thing as "midnight end of day", except as a convenient way to say "midnight start of next day". In common usage, these phrases can be helpful, but a clock cannot display both. So a clock should choose the only one that is mathematically correct, which is 00:00 midnight begins a new day. And no, 00:00 midnight ends the day cannot exist, because the 00:00 would display only for a pico-second, or a whole minute after midnight would be attributed to the day that just ended. And neither can 24:00 end of day exist because 24:xx would need to display, which is mathematically unworkable. Ohgddfp (talk) 04:20, 8 March 2014 (UTC)
As an aside, I notice that the "Written 24-hour clock, ISO 8601" row has entries for both midnight start and end of day. I suggest that this row should just be "ISO 8601" (which is well-defined "style"), to avoid confusion with an actual physical clock - which we both agree cannot show both 00:00 and 24:00. 8601:2004 does not use the term "clock" at all" except to mention in its definition of standard time of day that "Standard time of day is called “clock time” in IEC 60050-111." Mitch Ames (talk) 11:52, 11 February 2011 (UTC)

────────────────────────────────────────────────────────────────────────────────────────────────────All automatically set digital clocks I know of display a truncated time; a clock that displays minutes would begin to display 11:30 at 11:30:00.00... and continue to display 11:30 for a period of 1 minute. The time displayed by a manually set clock depends on the person who set it. If, for example, the clock-setter was listening to CHU (callsign) and started the clock at the moment the minute tone sounded, then as far as humanly possible the clock would display truncated time. I think that's how everyone sets a clock, but I suppose someone could set the clock at the end of the minute.

If a correct digital clock with a truncated display, as described, were to display 24:00, it would only do so for an infinitesimal time, so short that no human could ever observe it, because at 00:00:00 + ε the correct display becomes 00:00 (where ε is a very small positive number such that 0 < ε << 1 s). In as much as a digital clock that displayed 24:00 would be incorrect according to the way nearly all people set clocks, I think we should not include a description of such a clock unless someone can provide a citation to a specific clock model that behaves in this way and was sold in reasonable numbers (following the WP:NPOV policy that the viewpoints of tiny minorities need not be presented in Wikipedia). Jc3s5h (talk) 16:46, 11 February 2011 (UTC)

Comment What is the source for this? μηδείς (talk) 03:26, 12 February 2011 (UTC)

If by "this" you mean the claim that some digital clocks show 24:00 at midnight... I copied it from the last paragraph of 24-hour clock#Midnight 00:00 and 24:00 and added {{Citation needed}} to both. However in the absence of any reliable source I am happy to delete that row from the table. It doesn't affect the main dispute of this section, which is whether each "digital" clock row(s) in the table should show an entry in both the midnight start of day and midnight end of day columns. Mitch Ames (talk) 06:03, 12 February 2011 (UTC)
There used to be a picture of a clock on a kitchen appliance showing 24:00.
A rare example of a digital clock showing midnight as 24:00 instead of the standard 00:00. This variant notation has been used by at least one European supplier of kitchen appliances.
The owner posted this in talk:
I got a bit curious about why some ovens display 24:00 instead of 0:00 and which ones do it. So I emailed the BSH (Bosch-Siemens, who make Neff ovens too). Surprisingly I got a quick reply. Kudos to them for taking an interest! Here is what they had to say:
I had never noticed this and was not aware convention dictates that the clock is expected to show 00:00 as opposed to 24:00.
I believe that ISO 8601 is the relevant standard, but understand that this does not define whether 00:00 or 24:00 should be used and appears to actually recognise both as valid times. 00:00 however, appears to tbe the prefered display for midnight, and although some argue 24:00 is not a vlid time, there seems no reason (other than "convention" why we cannot use 24:00.
Frankly, I suspect that this is nothing more sinister than the fact that this is simply the way the clock was configured originally. I would be surprised if there was any attempt to differentiate the clock and the timer functions (why would there, as for every other minute of the 24 hour day, the settings are identical). Put simply, I think this is just the way it is.
As for Siemens and Bosch, [Borb] will no doubt be pleased to learn we stick with his prefered 00:00 display at midnight.
I think we can safely ignore this tiny minority in the article. −Woodstone (talk) 08:17, 12 February 2011 (UTC)
Agreed. I have deleted that row from the table. Now can we please get back to the original dispute, which is whether the digital clocks rows in the table should show "00:00 (start of next day) in the "Midnight (end of day)" column.

The "midnight confusion table" shows the meaning of the various notations in the selected styles. Again here, we should recognise that physical digital clocks simply never display end-of-day. They always show (now the exception is gone) the start of the next day. So in my view a dash is the best value. If you prefer we can also put a text like (shown as start of next day).−Woodstone (talk) 15:13, 12 February 2011 (UTC)

But start of next day is, well, the next day. The row should be for one day only, as indicated. Ohgddfp (talk) 04:20, 8 March 2014 (UTC)
There appears to be some confusion about the purpose of the table. (Remember that the purpose of the table is not stated in the article.) In my original question 1, I asked (03:28, 5 February 2011) "What is the purpose of the table? Ie what exactly is it intended to show?". Your original answer (13:27, 8 February 2011) was "The table should show how times within the day are stated (in a document) or shown (on a clock).", which is much the same as what I had assumed all along. However now you are saying that the table "shows the meaning of the various notations in the selected styles". This is not the same purpose.
For the first purpose (P1), a reader of the article should be able to take an existing clock/document/style S (eg a 12-hour digital clock) and a valid expression of time T (midnight end of day), look at the intersection of the corresponding row and column, and read off what value V the clock/document would show for that combination. As I have previously stated my digital clock will show V as "12:00 a.m."
"will show as 12:00 AM  ?? No it won't. It can on the next day. But this table row is not for the next day. It's for the day indicated for that row, which is today, not the next day. (Back to our program). Ohgddfp (talk) 04:20, 8 March 2014 (UTC)
so that is what I expect to see in the table. For P1, there should generally be exactly one applicable row (S = 12-hour digital clocks), one column (T = midnight end of day) and thus once intersection cell with one value (V = 12:00 a.m).
For the second purpose (P2), however, a reader should be able to look up (ideally find in a row and/or column) a style S (specific clock or document, eg 12-hour digital clock) and a specific "notation/value" V that appeared on a clock or in a document (eg 12:00 a.m.) and find in the table (ideally at the intersection of row/column - but not in this case) what that value means, ie find T expressed in "plain unambiguous English". For P2, there may be multiple values that satisfy some input criteria (S, V), because we know that some notations (eg V = "midnight" with no qualification) are ambiguous. For my case of S = 12-hour clock, V = 12:00 a.m, there is currently only one column with V = 12:00 a.m. and that is T = start of day. This is because (as you rightly point out) the clock's time does not ever reach end of day, it rolls over from 11:59 to start of day. So from the clock's point of view, it is start of day - regardless of my perception that it is end of day (eg as I'm going to bed).
Both P1 and P2 are useful purposes. The table appears to be currently structured for P1, in that row = S, column = T, and V = intersection of row/column. However it works for P2, because of a limited number of columns (3). I find row = S, then I check all columns for the value V and my output T is the title of the columns(s) in which I found V. In ambiguous cases, I find V in multiple columns, so I have multiple outputs T.
Now we could add some text to the article explaining that the purpose of the table is P2 (and not P1, despite being structured that way), and thus why the clock rows only have one midnight entry. Or we could change the structure of the table somehow so that the inputs S and V are row and column, or combinations of S, V per row with a single output column for T, to better suit purpose P2 - but this would probably be impractical. But the existing table can easily serve both P1 and P2 with little modification - all we have to do is add V = 12:00a.m/00:00 (start of next day) to the intersection of the "clock" row and the midnight end of day column.
Does this make the table any more confusing for readers using the table as P2? I think not. In fact, I think it adds value. It more clearly indicates the fact that when someone asks the question "what is the meaning of the display 12:00 a.m. on my digital clock?" (in your words "the meaning of the ... notation on the selected [clock]"), the answer is both midnight at the end of one day and simultaneously at the start of the next day. Surely it is more useful to give the "meaning" in human terms (midnight is both end of one day and start of the next), rather than clock terms. When I ask "what does this value mean", I don't care that it means for the clock, or that the clock can only go up to 11:59 p.m. - I want to know what it means to me.
To avoid going once more in the same circle, consider a digital clock that displays both date and time. On 13 February at midnight at the end of day it would not show 13 February 0:00 a.m.", but 14 February 0:00 a.m.". The clock does not show end and start simultaneously, it only shows the start. The end of day is never shown on such a clock and the table entry should be empty. In other words, the table takes on its meaning in the context of a specific date. Relative to that date 0:00 a.m. can only be shown at the beginning. −Woodstone (talk) 14:24, 13 February 2011 (UTC)
This is a valid point - for a digital clock that shows the date as well as the time (eg videocassette recorder, digital video recorder, wristwatch with date). However it does not apply to a digital clock that does not have the date (eg the alarm clock in my bedroom, the typical alarm clock pictured in this article, the digital clocks in my ovens, air-conditioners, car).
If dateless clocks and dated clocks show different times then one has to be wrong. They only way for dated clocks and dateless clocks to display the same as each other is to make the dateless clocks follow the same rules as the dated clocks. Ohgddfp (talk) 04:20, 8 March 2014 (UTC)
And even for clocks that include date, I would still maintain that for purpose P1 a reader could legitimately ask "what will this clock display at midnight end of day". I would answer that question - and expect the table to answer the question - "12:00am / 00:00 (start of next day)".
No, because an entry start of next day is not the same day as indicated by the row. There are not two dates on the same row.
How would you answer the question - surely not "the clock will not display anything" (which is what the current table says)
Actually the clock indeed displays nothing for that day. Lucky that it displays something for the next day, which can be found on the row for the next day. Ohgddfp (talk) 04:20, 8 March 2014 (UTC)
or "your clock only counts up to 11:59pm, so don't look at it when you're going to bed at the end of your day".
Fortunately, I cannot look at it at the end of my day and see 00:00, because then I'd be looking at it at the beginning of the next day. Ohgddfp (talk) 04:20, 8 March 2014 (UTC)
Possibly we need separate rows for clocks with/without a date - but I think that will just complicate matters unnecessarily.
As mentioned above, since dateless and dated clocks must follow the same rules, no need for the separate rows. Ohgddfp (talk) 04:20, 8 March 2014 (UTC)
It would help convince me of your case if you could point out what's actually wrong with my reasoning. To do this, could you please:
  • State explicitly what the purpose of the table is, ie my original question 1. You originally answered this as purpose P1 described in detail above, but then changed it to a different purpose P2. Do you think that table exists for purpose P1, or P2 or both? Or something else? Specific examples may help, just as I did in my description of P1 and P2. If there is some flaw in my description of the purposes P1, P2 tell me exactly what is wrong. (How can we agree on what should be in the table if we don't know what the purpose of the table is, or how we expect a reader to use it?)
  • Explicitly answer question 3, is "midnight end of day" a meaningful time for me refer to?
Not mathematically meaningful. But since "midnight end of toItalic textday" is not ambiguous, that might be an OK substitution for midnight beginning of new day. But since a clock cannot display both, only the mathematically correct one, 00:00 begins new day, should be displayed. And the table is about what is displayed. Ohgddfp (talk) 04:20, 8 March 2014 (UTC)
(If the answer is "no", life will be much easier because we can delete that whole column from the table!)
  • Explicitly answer question 4 (for a clock without a date). Imagine a reader asks the question - as I have done several times now - "when I look at my 12h clock at midnight end of day, what does it display". In plain english, how would you answer? If it's not a sensible question, tell me why? There's no point telling me it's not a meaningful question because of how the clock works - I assure you I can look at my clock at midnight at the end of day and it does display something! If you don't believe me, try it yourself.
I've tried it but it doesn't work. Every time I look at the clock at midnight, it's never at the end of the day. It's the next day. Remember that dated and dateless clocks alike must both show the time the same as each other. Otherwise, one would be showing the wrong time, and clocks showing the wrong time are not part of the table. Ohgddfp (talk) 04:20, 8 March 2014 (UTC)
You keep explaining how the clocks keep time, but the problem is now about how clocks keep time, it's about what's in the table. Obviously the two are related, but I can't see (in your statements) any logical progression from the former to the latter. I've presented some specific scenarios and linked them directly to specific usage of the table, with (what appears to me to be) a clear logical progression from scenarios to table contents. If you think my reasoning is incorrect, please point out where I'm going wrong. Mitch Ames (talk) 13:45, 14 February 2011 (UTC)

Can we please have some input from other editors on this. Surely someone else out there must be able to confirm my view, or point out what's wrong with my reasoning! Mitch Ames (talk) 13:47, 14 February 2011 (UTC)

OK, each row is for a day. There are not two days in the same row.Ohgddfp (talk) 04:20, 8 March 2014 (UTC)

I still think the fundamental disagreement is rooted in the unspecified purpose of the table. Perhaps the solution is to add a title to the table stating what the table contents are actually intended to be (rather then what the purpose of the table per se, ie how we expect the reader to use it). Then for the the missing entries, a note explaining why they are missing (since it is not necessarily obvious). For example:
Time as denoted by various devices or styles
Device or style Midnight
start of day
Noon Midnight
end of day
Digital 24-hour clock displays 00:00 12:00 N/A*
12-hour digital clocks
with a.m. and p.m.
12:00 a.m. 12:00 p.m. N/A*
* Digital clocks never reach midnight at the end of the day. Instead they wrap from 11:59 p.m. or 23:59 to midnight at the start of the next day.

The wording might need a bit of tweaking, the asterisk should perhaps be a numbered note using proper wiki syntax, but this illustrates the general intent. The important parts are that:

  • The table has a title indicating that it shows the time as denoted by various devices/style (eg it's from the clock's perspective, not the person looking at the clock at midnight end of day)
  • We explain why some entries are missing.

Mitch Ames (talk) (not logged in, so this edit appears from an IP), 00:03, 15 February 2011 (UTC) —Preceding unsigned comment added by (talk)

I was tempted to write a nearly essay-length response to this about how digital clocks work and why 24:00 is never a valid displayed representation of time on a clock (even though it's valid as a representation of an exact instantaneous moment) but in the end it didn't have much to do with my conclusion as it relates to this topic - consider yourselves to have dodged a bullet =)
It's clear that while it's possible to define 'midnight end of day' and 'midnight start of day' as different conceptual representations of the same thing, and while some systems allow this when referring to zero-length instants, in practice it is almost never seen. Do the few exceptions then justify an entire extra column in the table? My feeling is no. This table can be more cleanly represented with 'midnight' and 'midday' columns, where the 'midnight' column contains multiple values where appropriate (e.g. '00:00/24:00') and attach a table footnote if explanation is needed (e.g. '00:00 refers to midnight at the start of the day and 24:00 refers to midnight at the end of the day, though in practice 24:00 of one day is the same moment in time as 00:00 of the next day' though this could use better wording, obviously). This essentially makes the table built around the rule rather than the exception, but still allows you to make mention of the unusual cases as necessary.
On a related note, I don't agree that the 'de facto US legal standard' should appear in the table. The legal standard uses 00:01 and 23:59 explicitly to avoid making reference to midnight, it doesn't define midnight as one of those two values. TechnoSymbiosis (talk) 05:33, 16 February 2011 (UTC)
Avoid or define? Are there any legal precedents where a car accident took place at 12:59:30 p.m. on the last day of validity? If the insurance has to pay, the notation "defines" (for legal purposes) midnight. If they don't have to pay, they "avoided" midnight. My guess would be that any time on the last date is included in the contract in such cases. −Woodstone (talk) 08:45, 16 February 2011 (UTC)
I disagree that midnight is defined as 23:59 (or 11:59pm), but I don't have any US contracts on hand to serve as demonstration. My understanding is the reason 23:59 or 11:59pm is used is expressly to avoid using the term midnight, which is understandable ambiguous. Ambiguity is a Bad Thing in legal contracts, since it tends to rule in favour of the interpreter (if I said a bill was due by Wednesday, I may have intended to mean the beginning of the day but the other party could reasonably interpret it to mean the end of the day) so they try to avoid it where possible. In this sense I believe the common usage in legal contracts is to avoid midnight and use an unambiguous nearby figure than to define the term midnight itself as something it's not. TechnoSymbiosis (talk) 20:29, 17 February 2011 (UTC)
It's a fact that contracts in the U.S.A. sometimes state validity till 12:59 p.m. on the last day. Question remains what would happen if you have a car accident at 12:59:30 that day. Will your insurance have to pay up? Literally no, according to intent yes. −Woodstone (talk) 05:23, 18 February 2011 (UTC)
I'd like to improve the article, because I still think it is wrong as it stands, but I can't seem to get any specific responses to my specific reasoning or suggestions. Does anybody object to my changing the table to be something like this?
Time as denoted by various devices or styles
Device or style Midnight
start of day
Noon Midnight
end of day
Digital 24-hour clock displays 00:00 12:00 N/A*
12-hour digital clocks
with a.m. and p.m.
12:00 a.m. 12:00 p.m. N/A*
... ... ... ...
* Digital clocks never reach midnight at the end of the day. Instead they wrap from 11:59 p.m. or 23:59 to midnight at the start of the next day.
Or does anyone have any specific better ideas? Or shall I just give up in disgust take a break for a while? Mitch Ames (talk) 12:50, 18 February 2011 (UTC)
Yes, I know the feeling, WP editing and "Talk"ing can be quite unnerving. I mostly agree with your proposal above, but would like to combine with proposals in the section on ISO below as in:
Time as denoted by various devices or styles
Device or style Midnight
start of day
Noon Midnight
end of day
written 24-hour clock
(including ISO 8601)
00:00 12:00 24:00
24-hour digital clocks 00:00 12:00 —*
12-hour digital clocks
with a.m. and p.m.
12:00 a.m. 12:00 p.m. —*
written 12-hour clock 12:00 a.m. 12:00 p.m. —**
... ... ... ...
* Digital clocks never reach midnight at the end of the day. Instead they wrap from 11:59 p.m. or 23:59 to midnight at the start of the next day.
** Likewise the written 12-hour clock wraps immediately to the start of the next day
Woodstone (talk) 17:34, 18 February 2011 (UTC)
OK, I've updated the table, mostly from your version above, but with some variations. I've put "written ... time" (or "" in the ** note at the bottom) instead of "written ... clock". I think it is important to clearly distinguish between an actual clock and the time when written down on a bit of paper - ie by not using the word clock unless referring to an actual clock. Also ISO 8601 doesn't use the word "clock". Mitch Ames (talk) 05:58, 19 February 2011 (UTC)
Fine. Looks good. Seeing the result, it occurred to me that we might move the daggers to the last column, to have all notes there. −Woodstone (talk) 06:19, 19 February 2011 (UTC)

Hi. I just read this article and saw "Digital clocks never reach midnight at the end of the day." which is simply not true. I've used several cookers over the years that have displayed 24:00 for a minute at midnight. I would be the first to say it shouldn't happen, that it's wrong, etc. but this statement should not be as bold as saying digital clocks "never" reach midnight. Perhaps "typically" or "usually" but the word never is just not correct. I was going to correct it but having seen all the talk here I'm just going to suggest that the word "never" be replaced with "usually". squish (talk) 14:06, 12 February 2012 (UTC)

: Agreed, fixed. Mitch Ames (talk) 10:37, 13 February 2012 (UTC)

Does anyone else have a concern regarding the Japanese entry in the table? I have a concern given the fact that it cites a blog page written by chiyon? I am also concerned regarding statements such as, At the moment, {Japanese characters} seems to be the correct way to indicate noon in Japanese law. Besides the word seems, the author links no authoritative site of Japanese law. Without the meanings of the Japanese characters with their historical context explained, I don't see how this blog page can be accepted by Wikipedia as a valid source reference (not to mention the fact that it appears to be original research with no cites). I recommend this table entry be deleted unless this information can be validated by a reasonable source reference. JackOL31 (talk) 18:09, 20 March 2012 (UTC)

Agreed. It should be removed until it can be confirmed. Betaneptune (talk) 04:17, 3 September 2012 (UTC)
The Japanese FAQ referenced doesn't even have the letters "a.m." or "p.m." or any variation thereof. I'd recommend updating this table so it includes the specific Japanese characters mentioned in the referenced article rather than proclaiming that midnight is written as "12 p.m." and noon is written as "12 a.m.". This is clearly an absurd assumption with no references to back it up. dctucker83 (talk) 23:32, 3 November 2015 (UTC)

The purpose of the table, as best I can tell, is to show how each entity, be it a clock, organization, or what have you, represents midnight, noon, and midnight end of day. In other words, the table answers the question: "At the given time, what will we see on the clock or in a paper that represents that time?"

For clocks, we're talking here about those that show only hours and minutes. Since such clocks cannot show the day, the third column should show the same as the first column. Note that this works in only one direction. Time --> Representation. At noon, for example, a 24-hour clock will show 12:00. But the clock shows 12:00 for an entire minute, the time period [12:00, 12:01). (This means the period from exactly 12:00 to exactly 12:01 including the 12:00 instant but excluding the 12:01 instant.) This is what results from truncating the exact time, which is what digital clocks do. Only the first instant of this interval is noon. So it works differently in the other direction. That is, when you see 12:00, it is already past noon. The table doesn't indicate this crucial difference.

The U.S. Govt. Printing Office 1953 is simply wrong. I guess this is useful for old gov't papers. They now have it right, but the table should have 12 a.m. in both midnight columns, as the reference sited simply says "12 a.m. (12 midnight)".

From what I read above, the Japanese entry is not confirmed (perhaps I missed it if it was) and so should be removed for the present.

NIST still insists that 12 am and 12 pm are ambiguous, but at least they finally say something useful w.r.t. digital clocks. Also, on they say, "To convert 24-hour time to 12 hour time, subtract 12 if the time is greater than 12. A toggle switch for this will be featured in future versions of the site." (They are referring to the site Hmmm, how do you think they are going to handle noon and midnight?

The part that says, "Digital clocks usually do not reach midnight at the end of the day. Instead they wrap from 11:59 p.m. or 23:59 to midnight at the start of the next day", doesn't make sense. It implies that there are two midnights per day. The clocks do reach midnight at the end of day. They just display it as 00:00 or 12:00 am (or 24:00 on obscure kitchen appliances!). Clocks showing just hours and minutes do not tell you what day it is, so the whole thing is meaningless. Technically, according to both time systems, midnight the first instant of the day. It can be considered as the last instant for convenience. And ISO seems to be okay with it. But there is only one midnight per day. Additionally, any clock that shows 24:00 for an entire minute is wrong for all but the first instant. Furthermore, no clock is exactly right anyway! So all this fuss about instants is -- well -- overkill, for lack of a better word that I can come up with right now. I mean, suppose an am/pm clock is running fast. Then it will show 11:59 a.m. at noon and a little later, perhaps only seconds later, will finally show 12:00 p.m. Clearly this happens after the "real noon". See, it gets a bit ridiculous sticking to the original Latin. (!)

Legal: I don't believe 12:01 a.m. is intended to be midnight. The idea is to avoid ambiguity and it assumes that a difference of a minute makes little difference. Besides, if 12:01 a.m. meant midnight, what would they write for 12:01?! Nowadays, when timing down to the second or less is common, this issue needs to be revisited. OK, this whole thing is a little fuzzy. Can any lawyers chime in on what happens if an event occurs between 12:00 am and 12:01 am one minute later? (The same of course goes for 11:59 pm and 12:00 am one minute later.)

It's not just legal that uses 12:01. The NYC subway uses 12:01 a.m. WITH A DATE in service notices. Of course, service changes don't start with such precision, so saying 12:01 a.m. instead of 12:00 a.m. makes no difference other than clarifying things for those who don't know that 12:00 a.m. is the beginning of the given date.

Encyclopedia Britannica: 12m for noon? Are they for real?

Bottom line: I think the table is simply not able to give the full story that's intended. It is just not up to the job. The issue is not simple enough for such a table. Clocks don't indicate the day, but a few of the conventions listed do. Also, hour-and-minute clocks show time truncated down to the minute, whereas the other entries mostly refer to instants. Clearly we've got a case of apples and oranges, at least to some extent.


BTW, just to be clear (!): I have no objection to use of the terms noon and midnight. I only object to the insistence by some that 12 a.m. and 12 p.m. are ambiguous or meaningless. The term "midnight" is ambiguous, not 12 a.m. and 12 p.m.

The ambiguity about midnight is usually resolved by the context. "What time is it? Half past midnight." That's clearly 12:30 am, or 00:30. "I'm going to be up until midnight". That midnight is the next upcoming midnight. If the statement is made at 19:00, the midnight being referred to happens 5 hours later. In either case, whether it's considered the end of the current day or the beginning of the next day is irrelevant. This is because which midnight is being referred to is clear from the context. Betaneptune (talk) 04:14, 3 September 2012 (UTC) (Moved my post to the correct place in the page. Betaneptune (talk) 04:55, 3 September 2012 (UTC))

This entry in Japanese by Japan's standards group has an FAQ on the time notation. Perhaps this can augment the reference in the table on conventions per country or authority. Hv8EEHpq (talk) 01:30, 24 November 2012 (UTC)

I must admit I find this statement That leaves 12:00 a.m. to be used for midnight at the beginning of the day, continuing to 12:01 a.m. that same day, to be a tad misleading. The section belabors the same point (7 paragraphs) and this statement utilizes tortured reasoning to make that point. I would find it difficult to argue the fact that the beginning of a day does not fall within the ante meridiem timeframe of the given day. This does not leave 12:00 am to be used for midnight-beginning of day, but rather it IS midnight-beginning of day. The article tries to make a problem where none exists. I agree some people do not know which 12:00 is which, but the naming convention matches the condition. It seems to me the first sentence in the section is incorrect and the second sentence fully supports the fact that 12:00am is indeed denoted correctly!
Regarding noon, all of the 43,200 post ante-meridiem times (in seconds) have historically been denoted by the label post meridiem whereas only 43,199 are technically post meridiem. While important to note the mislabeling, I don't see the need to belabor the point that the 1 time (12:00pm) is post ante-meridiem and by convention denoted as post meridiemJackOL31 (talk) 19:18, 9 June 2013 (UTC)

Since the Associated Press and the "U.S. de facto legal" group (U.S. lawyers, i.e.) represent 12:00 a.m. as 12:01 a.m., how do they represent 12:01 a.m.? How would you distinguish between 12:01 a.m. meaning midnight one minute back and 12:01 a.m. meaning the actual time 12:01 a.m.? (Maybe with 12:01:01 a.m.? :-) Again I ask, primarily out of curiosity: If there are any lawyers reading this, how are cases handled for events that happen between 12:00 a.m. and 12:01 a.m.? And similarly for the case with p.m. (between 11:59 p.m. and 12:00 a.m. the next day)?

Again I say: It's time to move past the original medieval, but now obsolete and unworkable definitions of a.m. and p.m. and go with the eminently sensible, practical, logical, mathematical definition in which 12:00 a.m. is 00:00 ("midnight" at the beginning of the day) and 12:00 p.m. is "noon". This definition is used by virtually all digital clocks, be they standalone devices, embedded systems, or computers.

This is the year 2014, not 1700. Words evolve. That's how it is. Time to move on.

Betaneptune (talk) 20:53, 8 March 2014 (UTC)

If we're going to "move on", we should get past this concept of measuring things relative to noon, and all switch to 24-hour time, expressed per ISO 8601. Sadly, I don't think that will happen any time soon... Mitch Ames (talk) 13:56, 11 March 2014 (UTC)
I'm all for moving on to the 24-hour format, but if you're going to use the 12-hour format at all, it is best to shun overly literal interpretations of outdated archaic terms, and let 00:00 be represented by 12 a.m. and 12:00 by 12 p.m. Hey, if you park on the driveway, and drive on the parkway, you can (and should, regardless!) exempt midnight and noon from strict interpretation of ante meridiem and post meridiem. And furthermore I don't see anyone arguing against the current meanings of driveway and parkway. Betaneptune (talk) 00:15, 4 April 2014 (UTC)

Is Midnight 12 am or 12 pm?[edit]

Americans have style guides that state 12 pm is Noon. Where I grew up in Kenya, 12 pm was midnight. I've asked many people that I know in the UK, South Africa, Kenya, Australia what time of day 12 pm was? Every one of them said midnight. How do we prove this when there is no "style" guide? Government or standardisation organisations in English speaking countries dictate use of the 24 hour clock. So we're left with what people in general might express online or in conversation.

Here is an extract from Trip Advisor feedback on hotels,, the respondent states: "I prefer to stay on the Club floors (4th and 5th floors) where the club lounge remains open till 11 or 12 pm! A relaxing place to hang out after a long day". What I take from this is that this person uses 12 pm as the end of the day, as would I. Other examples would be appreciated. Avi8tor (talk) 22:21, 19 June 2014 (UTC)

Because of this confusion, the best advice is to avoid using a.m. and p.m. after "12". Just use "noon" and "midnight". It seems that the majority, world-wide, use 12 p.m. to mean noon, then 12:01 p.m. follows on, but I agree with you that there is no universality about this convention. I've never used it. Dbfirs 06:34, 20 June 2014 (UTC)
Certainly on computers or digital clocks 12:00 p.m. is always noon, never midnight. −Woodstone (talk) 16:39, 20 June 2014 (UTC)
There is an important difference between a static time, such as time written on a piece of paper, and the dynamic display of a clock, especially a digital clock. If we imagine that an ideal digital clock that's perfectly accurate and that it only displays hours and minutes, for an infinitesimally short time it displays whatever notation the clock designer things is appropriate for noon, either 12:00 AM or 12:00 PM. But the duration of that display is 0 seconds, so it cannot be seen. Then it is afternoon, and the clock displays 12:00 PM until a minute has passed, and then it displays 12:01 PM. Of course a practical clock will ignore the question of what the correct display would be for noon, and only display the correct notation for afternoon.

If it happens to be an alarm clock, then the issue of the correct notation for noon and midnight comes into play when setting the alarm. Jc3s5h (talk) 21:27, 20 June 2014 (UTC)

The disambiguation of 12AM and 12PM provided in the article is useful, but might I suggest that we add the standards used by computing? 12AM has a fuzzy meaning in english, but it probably has a very specific meaning in Unix and Windows (And I hope the same one...) (talk) 12:24, 28 May 2015 (UTC)

Wikipedia - from a basic intelligence level you and or the American and Canadian International Convention is illogical and wrong!

00.01am is a minute past midnight - 11 hours and 59 minutes before 12 Noon.

12.01am doesn't exist because it is actually 00:01am; but 12.01pm (afternoon) does when using the 12 hour clock.

It is 59 minutes before 1pm.

Having reached the Twelfth division, the units return to zero, so technically 12:01 does not exist - especially just after Midnight and 12:01 only exists just after Noon in the 24 hour clock system.

Therefore the use of am and pm with the number 12 should be avoided at all costs, if confusion is also to be avoided.

Wikipedia is always right?

.....And all people should accept any and all absolutely absurd and confusing American and Canadian conventions which are 11 hours and 59 minutes adrift/apart.

Key Facts:-

example:- News Story reporting a sighting of a missing child.

11.59pm is 1 minute before midnight - so it's completely illogical that the missing kid was last seen between 11:00am and 12pm - so he was in the same place for 13 hours then? 13 hours or 1 hour in one place - I think most probably 1 hour between 11am and noon but the police didn't correct or clarify their tweet. Nor did they quote Wikipedia with all quotes from American and Canadian Sources.

....12pm is 11 hours and 59 minutes later!?

Similarly 12:01pm is 1 minute after 12 Noon and not 1 minute after Midnight - that is 00:01am

If you, the Americans, Canadians and Wikipedia are all right then the kid was seen/observed for 13 hours!

Think we need to use British English and sensibly avoid using the am or pm suffixes; but instead use the terms Noon, Mid-day or Midday and Mid-Night or Midnight.

It is therefore well advised to drop the use of 12am or pm because there is confusion and people can easily be 11 hours and 59 minutes early or equally, 12 hours and a minute late - or is it the other way round?

More intelligent to say 12 noon, Mid-day/Midday and use 12 mid-night/Midnight or just midnight for 24:00 if not sensibly using the 24 hour clock.

Digital Clocks need correctly reprogramming to avoid 12am and 12pm and convention needs amending to apply colloquial and general terms used in the every day universally accepted British English Language.

The difference is as plain to see as Night and Day!

Conclusion regarding the Time and Clock Convention.

General Analogue meter reading conventions.

When reading an analogue gas or electric meter - or any meter for that matter, it is read backwards and not forwards. That is to say that a Seven stays a Seven until the next unit/dial zero has passed through zero, which then simultaneously changes the Seven to the next whole number, namely an Eight. To read the Seven and Three quarters as an Eight for the whole set of dials would give a dramatically inflated reading and subsequently a very very high Gas or Electric bill.

A clock is a measuring device or meter and so therefore to read it accurately and correctly the same conventions need to be applied. Using this principal, in the case of a clock and time it is still in the Eleventh hour until it actually passes the twelve. There is no Twelfth hour time period and so therefore it stays am until precisely Twelve o'clock plus a fraction of a Nano second. When using a twelve hour system the next unit after Twelve is Zero and Zero plus an infinitely small proportion of a Nano second is 00:00:00:01pm. So am does not become pm until a very small fraction of time after Twelve o'clock and therefore up until the change to pm the time actually still remains am. Incorrectly reading forwards causes an increase of Eleven hours Fifty Nine Minutes and a large number of Nano seconds. The logical and analytical difference when explained correctly in this way is as plain to see as Night and Day!

A comment on the Twelve Hour Clock.

It is very important to remember that a Twelve Hour Clock has only Two divisions of Twelve hour periods and that the Thirteenth Hour does not exist. An am time period ends at 12:00 and a pm time period starts immediately the time has passed a split Nano second after the Twelfth hour and begins at zero. So technically speaking there is no true time of Twelve Twenty in the use of the Twelve Hour Clock for example. It is a misnomer. Twelve Twenty does exist, however, in the Twenty Four Hour Clock.

It is very important to remember that a Twelve Hour Clock has only Two divisions of Twelve hour periods and that the Thirteenth Hour does not exist. An am time period ends at 12:00 and a pm time period starts immediately the time has passed a split Nano second after the Twelfth hour and begins at zero. So technically speaking there is no true time of Twelve Twenty in the use of the Twelve Hour Clock for example. It is a misnomer.

Twelve Twenty does exist, however, in the Twenty Four Hour Clock.

HigherIntelligence. — Preceding unsigned comment added by HigherIntelligence (talkcontribs) 13:32, 21 June 2014 (UTC)

This is not an American versus British issue. The use of "12 p.m." instead of saying 12 noon seems to be common world-wide. I agree with you that it is best avoided, but we have to accept the world as it is, including accepting the figure 12, not zero, at the top of our clocks. (Posting your opinion on Wikipedia is not going to change the world, however high your intelligence.) Dbfirs 17:47, 21 June 2014 (UTC)
HigherIntelligence: This is a wikipedia talk page for improving the article, not a forum for generalized discussion for your (somewhat logical, I will admit) feelings about how time and clock conventions are used in various places. Your comments are completely off topic for this talk page. Utter wankery is a not a demonstration of higher intelligence. If you have suggestions for improving the article, by all means advance them. Hating the way that the world works and complaining about it are not a legitimate use of Wikipedia talk pages, sorry. (Reply at 04:29 on March 31, 2015‎ by User:ZeroXero(talk))

12 vs. 24 hour clock dial and centuries[edit]

The article contains this paragraph:

During the 15th and 16th centuries, the 12-hour analog dial and time system gradually became established as standard throughout Northern Europe for general public use. The 24-hour analog dial was reserved for more specialized applications, such as astronomical clocks and chronometers, and timetables, especially for railway and airline travel.

Neither railways nor airlines existed in the 15th and 16th centuries. This paragraph needs to be rewritten to make sense.

ZeroXero (talk) 03:33, 31 March 2015 (UTC)

Good point. I'll adjust the paragraph. Dbfirs 07:41, 31 March 2015 (UTC)


Is "meridian" just a misspelling, or is it a related concept? —DIV ( (talk) 05:57, 6 September 2016 (UTC))

A meridian is the collection of all places on Earth, where the (solar) meridiem occurs concurrently. −Woodstone (talk) 07:26, 6 September 2016 (UTC)

External links modified[edit]

Hello fellow Wikipedians,

I have just modified 2 external links on 12-hour clock. Please take a moment to review my edit. If you have any questions, or need the bot to ignore the links, or the page altogether, please visit this simple FaQ for additional information. I made the following changes:

When you have finished reviewing my changes, you may follow the instructions on the template below to fix any issues with the URLs.

You may set the |checked=, on this template, to true or failed to let other editors know you reviewed the change. If you find any errors, please use the tools below to fix them or call an editor by setting |needhelp= to your help request.

  • If you have discovered URLs which were erroneously considered dead by the bot, you can report them with this tool.
  • If you found an error with any archives or the URLs themselves, you can fix them with this tool.

If you are unable to use these tools, you may set |needhelp=<your help request> on this template to request help from an experienced user. Please include details about your problem, to help other editors.

Cheers.—InternetArchiveBot (Report bug) 00:31, 14 June 2017 (UTC)

Why does it start at 12 and not 1?[edit]

In this article, it clearly states "Each period consists of 12 hours numbered: 12 (acting as zero),[3] 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, and 11." Why does it start at 12 instead of 1? Shouldn't 12 be at the end? Why is this? Thanks. (talk) 00:39, 17 July 2017 (UTC)

Yes, logically, it should start at zero, but the zero symbol was a late addition to our number system, so 12 was used by the ancient clockmakers. Dbfirs 07:14, 17 July 2017 (UTC)
... later ... Sorry, you mean the order in the list in the article! Yes, perhaps it would be clearer with 12 at the end. What does anyone else think? Dbfirs 07:17, 17 July 2017 (UTC)
I think it's better as is, because the hours are listed in chronological order. Jc3s5h (talk) 11:40, 17 July 2017 (UTC)
In fact, a zero would be more logical than 12 for the 12-hour clock (0 AM, 0 PM). (talk) 20:14, 24 July 2017 (UTC)
Lets say it's Saturday evening, 9 PM, 10 PM, and 11 PM, and it's still Saturday. But "12" becomes AM and it's now Sunday? Why isn't 12 still part of Saturday? Why doesn't Sunday AM start at 1, back at the beginning of the counting cycle? Can someone answer this please? Thanks! (talk) 19:40, 4 September 2017 (UTC)