1179: "ISO 8601"

This forum is for the individual discussion thread that goes with each new comic.

Moderators: Moderators General, Prelates, Magistrates

User avatar
tetsujin
Posts: 426
Joined: Thu Nov 15, 2007 8:34 pm UTC
Location: Massachusetts
Contact:

Re: 1179: "ISO 8601"

Postby tetsujin » Wed Feb 27, 2013 9:44 pm UTC

Area Man wrote:
Quicksilver wrote:it should just be DD/MM/YYYY. Makes the most sense.

No! Would you seriously write time as ss:mm:hh? or mm:ss:hh? No. The date is just an extension of that. L-R is big to small (more precision further right, like normal numbers).


Although I strongly prefer YYYY-MM-DD, I think DD/MM/YYYY does make a certain amount of sense (at least in some contexts).

Basically, it puts the most useful information up front.

"What is the date?"
"It's the 27th." ("Of what?" "February.")

Often, that's all you need. The year and the month are usually the same as they were yeaterday, and it's mainly the day-of-month we have to keep track of.

With time-of-day, it's the opposite situation. The hour is more often the most useful piece of information, followed by some measure of our progress through that hour (it's a little after four). Seconds go by so quickly that they're often not worth keeping track of.

But for datestamps, back-reference, I think Y-M-D is a good system. For the full timestamp it puts all the fields in consistent order of decreasing magnitude/increasing precision, and when you're looking at a collection of files, for instance, that goes back years, starting with the year makes more sense.

For sorting file names, there are a couple extra wrinkles... Filenames aren't necessarily sorted in a strict lexicographical order. Windows tries to sort numbered files smartly (like "porn_image_8.jpeg" will sort before "port_image_11.jpeg") - that approach could take dates into account as well, though personally I think the whole concept is kind of a mess - if there's fields that should be used for sorting, those should maybe be separate from the concept of a singular "name" for the file... But that's not the world we presently live in, so we make do...
---GEC
I want to create a truly new command-line shell for Unix.
Anybody want to place bets on whether I ever get any code written?

User avatar
Pfhorrest
Posts: 4872
Joined: Fri Oct 30, 2009 6:11 am UTC
Contact:

Re: 1179: "ISO 8601"

Postby Pfhorrest » Wed Feb 27, 2013 9:45 pm UTC

phlip wrote:in the US they'll typically say it as "February 27", matching their MM/DD layout.

I never hear anyone in the US say "February twenty-seven". It's always an ordinal: "February twenty-seventh", or "the twenty-seventh of February" (also perfectly intelligible to an American even though we don't write the numeric dates that way). Whatever order you speak it in, the day should never be just a plain number with no ordinal suffix; and if the number comes first, there should be a "the" and an "of" in there. "Twenty-seven February" is just an abomination. "Twenty-seventh February" is even worse, as that's grammatically correct but means something different: that refers to the twenty-seventh whole month of February in some sequence of Februaries, e.g. February of 2009 was the twenty-seventh February after my birth.
Forrest Cameranesi, Geek of All Trades
"I am Sam. Sam I am. I do not like trolls, flames, or spam."
The Codex Quaerendae (my philosophy) - The Chronicles of Quelouva (my fiction)

Gargravarr
Posts: 74
Joined: Mon Dec 21, 2009 8:34 am UTC

Re: 1179: "ISO 8601"

Postby Gargravarr » Wed Feb 27, 2013 9:50 pm UTC

jc wrote:It's a bit surprising that such a thing would exist in any computer language so widely used, because it would totally fail for most east-Asian dates. Thus, in China and Japan, today is 2012年2月27日, which is ISO format with the Chinese characters for year, month and date as separators. This is legal ISO format, because ISO basically says that only the digits (and the +- if there's a time zone) are significant, you can use any separators that you like, and those are the date-field separators in several of the world's major languages. If a date-parsing routine in any widely used programming language fails to parse something so simple and widely-used, the question "What were they thinking?" comes to mind.

jc, I suppose you're not familiar with PHP, the most popular server-side web language, and the one that renders the question "what were they thinking" moot, since they weren't thinking at all. JavaScript is one big messy afterthought, yes, but you haven't seen true, metaphysical ugly until you've tried to write something substantial in PHP.

Edit: Actually I have no idea what I'm arguing about. A bit drunk, will shut up now.

Bromskloss
Posts: 7
Joined: Mon Nov 14, 2011 10:50 am UTC

Re: 1179: "ISO 8601"

Postby Bromskloss » Wed Feb 27, 2013 10:44 pm UTC

da Doctah wrote:Both / and - characters should be avoided because of spreadsheet programs that will interpret them as division and subtraction respectively. Sorry, ISO guys, but you dropped the ball on that one.


I'm pretty sure there is an ISO standard that says that you shouldn't use spreadsheet programs.

PlayMp1
Posts: 3
Joined: Wed Feb 27, 2013 10:54 pm UTC

Re: 1179: "ISO 8601"

Postby PlayMp1 » Wed Feb 27, 2013 11:08 pm UTC

I'd like to note that a large number of languages have dissimilar month names, even Western European languages. January in Spanish is "enero" (note the lack of capitalization, and its complete non-relatedness to other languages - apparently it's thanks to Andalusian Arabic), August in French is "août" (same first letter, that's it), and February in Portuguese is the similar, but still different "fevereiro."

User avatar
walkerm930
Posts: 69
Joined: Wed Apr 07, 2010 3:53 am UTC
Location: Canada: Ontario: Toronto

Re: 1179: "ISO 8601"

Postby walkerm930 » Wed Feb 27, 2013 11:47 pm UTC

This has been said before, but not enough people are saying it.

YYYY[-/]MM[-/]DD makes the most sense when writing things down, because the reason we write the date is so that when we read it later, we know when something happened. And reading is fast enough that it doesn't really matter what standard is used, so therefore the most logical (and sortable) is best. I independently decided to do things this way five years ago, and probably should have guessed there was an ISO standard... Anyway, when other people don't use the standard, it's kindof annoying because there might be ambiguity (especially in Canada because there doesn't seem to be a generally used format). With the year first (as long as you use a separator), there is no ambiguity, because there not exists any other standard with the year first.

Now, talking, on the other hand, people generally converse about close dates, and omitting the year, or year and month, is perfectly fine. I live in Canada, and I'm not sure about other people, but when the month is needed I say "February 27th", and then if the year is not apparent, then "2013". Because, seriously, who is ever confused about the year (there are exceptions, yes). I don't remember my younger (7 years) brother ever being not sure. I would say, that as soon as you are able to understand what a date is, then you will be rarely confused (because they are so darn long and the schools really drill it into your head).

However, there are a number of things that I do because I believe they are "better", despite what others do. Usually things that make more "sense", are more regular, or deterministic (eg. this, dvorak(don't go off topic please), putting punctuation outside of quotes, LibreOffice, Linux ...)
In the gospel according to trig there are 3 primary rules: sin θ = x/h , cos θ = y/h and tan θ = x/y. These rules are not open to interpretation and are to be treated as law.

User avatar
New User
Posts: 657
Joined: Wed Apr 14, 2010 4:40 am UTC
Location: USA

Re: 1179: "ISO 8601"

Postby New User » Thu Feb 28, 2013 12:58 am UTC

When I was in the US Army, I noticed that there was confusion about the "official" way to write dates. It was my understanding that the official policy was to use a system like the one in this comic, but without hyphens. 20130227 for example.
Still, there were many people in the organization that preferred the use of another system, in which the date is written with a two-digit number, followed by a three-letter abbreviation for the month, and then a four-digit number for the year. Today's date would be written 27 FEB 2013 in this system. It is my hypothesis that this alternate system of writing was an outdated official system for writing dates, which had been superseded by my previous example, but still adhered to by the people who had been introduced to it in years past.
I prefer the second one. It seems easily readable, and difficult to misinterpret. The only issue I foresee with it would be communicating dates to people who use different words for the months, because of language differences. I have not personally encountered that problem, though.
Many official government documents have a space intended for a written date, and include a format such as YYYY-MM-DD. But the suggested format isn't the same for every document. Whenever the "official" method of writing dates was introduced, every document used by the Department of Defense was not updated accordingly. This detracts from uniformity in writing dates.

User avatar
gmalivuk
GNU Terry Pratchett
Posts: 26453
Joined: Wed Feb 28, 2007 6:02 pm UTC
Location: Here and There
Contact:

Re: 1179: "ISO 8601"

Postby gmalivuk » Thu Feb 28, 2013 1:27 am UTC

Pfhorrest wrote:
phlip wrote:in the US they'll typically say it as "February 27", matching their MM/DD layout.
I never hear anyone in the US say "February twenty-seven".
I don't believe you. I believe you don't notice it, because the 'th' is easy enough to miss or imagine, but I don't believe that you never hear it, because I sometimes say it myself, and I've definitely heard other people do the same.

Whatever order you speak it in, the day should never be just a plain number with no ordinal suffix; and if the number comes first, there should be a "the" and an "of" in there. "Twenty-seven February" is just an abomination. "Twenty-seventh February" is even worse, as that's grammatically correct but means something different
And if English were meant to be Lojban, your argument might carry some weight.

But English isn't perfectly logical, because it's a natural language rather than a constructed one. So some people say "half four" to mean 4:30 instead of 2:00, and all of us say o'clock even though "It's four of the clock" doesn't actually make grammatical sense, and if you can answer "four" for "what's the hour?", why not "February four" for "what's the date?" How does, "Today is February twenty-seventh," actually make any more sense, "logically", than saying, "Today is February twenty-seven."?
Unless stated otherwise, I do not care whether a statement, by itself, constitutes a persuasive political argument. I care whether it's true.
---
If this post has math that doesn't work for you, use TeX the World for Firefox or Chrome

(he/him/his)

Milnoc
Posts: 24
Joined: Mon May 07, 2012 4:43 am UTC

Re: 1179: "ISO 8601"

Postby Milnoc » Thu Feb 28, 2013 2:05 am UTC

That's the same format I've been using in my SQL date columns for over 20 years. Much easier to read and sort out in a report. All the other formats just confuse the heck out of me. Can't stand them at all.

User avatar
Klear
Posts: 1965
Joined: Sun Jun 13, 2010 8:43 am UTC
Location: Prague

Re: 1179: "ISO 8601"

Postby Klear » Thu Feb 28, 2013 2:16 am UTC

PlayMp1 wrote:I'd like to note that a large number of languages have dissimilar month names, even Western European languages. January in Spanish is "enero" (note the lack of capitalization, and its complete non-relatedness to other languages - apparently it's thanks to Andalusian Arabic), August in French is "août" (same first letter, that's it), and February in Portuguese is the similar, but still different "fevereiro."


Enero is quite clearly related to January (according to the Spanish wiki: del latín ianuarius--> janairo --> janero --> enero)
août begins to look like August when you realize that "û" signifies that there used to be "s" in the word - aoust (this is more apparent in other words - Île -> isle, être -> estre etc.

Of course, there are languages, such as Czech, where the names of the months are completely different, but you picked wrong examples.

User avatar
da Doctah
Posts: 902
Joined: Fri Feb 03, 2012 6:27 am UTC

Re: 1179: "ISO 8601"

Postby da Doctah » Thu Feb 28, 2013 2:17 am UTC

Diadem wrote:
Camahueto wrote:I have tried to set internet meetings with people from different countries at UTC times: I have failed miserably. All of them demands I set the time to their own time zones. For example, I need to propose hours listing the time in every time zone every time.

May I take this time to complain about people using UTC instead of GMT?

Everyone knows what GMT means, almost no one knows what UTC is, yet some people insist on using UTC, despite those two terms meaning the same thing (not quite, but the difference is never more than a second).


Why would I want to use a time system developed by the Zulus?

Anyway, we need an appropriate role model for this "day before month vs month before day" business, and I think I found one:
Image

User avatar
Djehutynakht
Posts: 1546
Joined: Thu Feb 10, 2011 1:37 am UTC

Re: 1179: "ISO 8601"

Postby Djehutynakht » Thu Feb 28, 2013 2:34 am UTC

Eutychus wrote:
Farabor wrote:Quick, someone post the xkcd about "There's too many standards, we need a universal standard!" Standard++ result. I would, but I don't do html or research :).


done.



Beat me to it.

User avatar
phlip
Restorer of Worlds
Posts: 7554
Joined: Sat Sep 23, 2006 3:56 am UTC
Location: Australia
Contact:

Re: 1179: "ISO 8601"

Postby phlip » Thu Feb 28, 2013 2:52 am UTC

boothby171 wrote:Also interesting: the forum-applied dates are not to Spec.

Go to Board Preferences, go down to the "date format" field, and replace all the gibberish in there with a single c. Bam, ISO-8601 dates. You can also enter r to get RFC 2822 dates.


Though, I just remembered my favourite date format: the one used by the "date" Unix command line tool to set the system date:

Code: Select all

date [-u|--utc|--universal] MMDDhhmm[[CC]YY][.ss]
Yes, that's month, day, hour, minute, optional year, optional second. With no delimiters, except the dot before seconds. So in order to set the date right now, you'd use

Code: Select all

date --utc 022802522013.20

Code: Select all

enum ಠ_ಠ {°□°╰=1, °Д°╰, ಠ益ಠ╰};
void ┻━┻︵​╰(ಠ_ಠ ⚠) {exit((int)⚠);}
[he/him/his]

User avatar
ysth
Posts: 182
Joined: Wed Apr 11, 2007 7:21 pm UTC

Re: 1179: "ISO 8601"

Postby ysth » Thu Feb 28, 2013 3:34 am UTC

Jean2 wrote:This system was good because at least there was a consistent date system, that didn't reset to year one every time a new king was crowned (like before), so Napoleon couldn't just declare it to be Year One again.

Um, you do know that Napoleon actually revoked the French Revolutionary Calendar, which had reset counting of years to have year I (Roman numerals, yes) beginning 1792-09-22?
A math joke: r = | |csc(θ)|+|sec(θ)| |-| |csc(θ)|-|sec(θ)| |

EvanED
Posts: 4331
Joined: Mon Aug 07, 2006 6:28 am UTC
Location: Madison, WI
Contact:

Re: 1179: "ISO 8601"

Postby EvanED » Thu Feb 28, 2013 3:44 am UTC

Jorpho wrote:And filenames can be sorted algorithmically.

nowhereman wrote:As for those complaining about the necessity of sorting information correctly, programming is about solving problems, and not creating arbitrary guidelines to save you from having to do so. Whenever I hear the argument from laziness, I am reminded of websites that require my password be exactly 10 characters long, and contain only certain valid characters. Sure, I would rather use a 22 character password made up of random letters, numbers, etc... but security is secondary to simplicity. After all, we all know how long it takes to write fifty lines of code to convert a password into a sanitized 128 character array. Actually, I am being unfair to the security community as I am sure someone has already created a library to do this for you. I am sure that the security community at large are smart enough to not repeat redundant code when a call would suffice.

Klear wrote:Edit: I don't see why everybody things the stortability is such a big deal, btw. Do we want a format that's convenient for people or for robots?
I sort of agree. But not really, because reality intercedes.

Yes, it would be nice if software sorted correctly given DD-MM-YYYY or MM-DD-YYYY. But... it doesn't. Nautilus doesn't, Windows explorer doesn't, ls doesn't. I'd be surprised if any remotely-commonly used file tool did. (The most likely one would be Finder, but I don't have a Mac to try.)

It's hard enough to convince programmers that "10 - foo" should sort after "2 - foo" :-).

dp2 wrote:Well, no, there's one higher, and ironically, the ISO format doesn't allow for it: differentiation. Easily telling the difference between any two dates or times. Only "time units since reference date" provides that.

Depends what format you want that difference in. Seconds/minutes/days? Maybe; sometimes. Anything else? No. If you want something which is actually usable to humans, it's easiest to use normal dates.

User avatar
CatCube
Posts: 38
Joined: Wed Sep 21, 2011 5:28 pm UTC

Re: 1179: "ISO 8601"

Postby CatCube » Thu Feb 28, 2013 4:16 am UTC

New User wrote:When I was in the US Army, I noticed that there was confusion about the "official" way to write dates. It was my understanding that the official policy was to use a system like the one in this comic, but without hyphens. 20130227 for example.
Still, there were many people in the organization that preferred the use of another system, in which the date is written with a two-digit number, followed by a three-letter abbreviation for the month, and then a four-digit number for the year. Today's date would be written 27 FEB 2013 in this system. It is my hypothesis that this alternate system of writing was an outdated official system for writing dates, which had been superseded by my previous example, but still adhered to by the people who had been introduced to it in years past.


27 Feb 2013 is the AR 25-50 standard for writing dates used on memoranda, and is still very much a current standard. ISO 8601 is generally used on forms (as you stated). DTG (272315FEB) is generally used for orders/ALARACT messages and the like.

EDIT: Fixed the quote tags.

Fire Brns
Posts: 1114
Joined: Thu Oct 20, 2011 2:25 pm UTC

Re: 1179: "ISO 8601"

Postby Fire Brns » Thu Feb 28, 2013 4:53 am UTC

Klear wrote:
PlayMp1 wrote:I'd like to note that a large number of languages have dissimilar month names, even Western European languages. January in Spanish is "enero" (note the lack of capitalization, and its complete non-relatedness to other languages - apparently it's thanks to Andalusian Arabic), August in French is "août" (same first letter, that's it), and February in Portuguese is the similar, but still different "fevereiro."


Enero is quite clearly related to January (according to the Spanish wiki: del latín ianuarius--> janairo --> janero --> enero)
août begins to look like August when you realize that "û" signifies that there used to be "s" in the word - aoust (this is more apparent in other words - Île -> isle, être -> estre etc.

Of course, there are languages, such as Czech, where the names of the months are completely different, but you picked wrong examples.

If the concern is about alphabetization then we have to determine what alphabet to use. English Latin has 26 letters whereas Spanish or French will consider other letters into theirs such as á, é, and ñ. Cyrillic has 33. And then you have syllybaries and sinographs.

On second thought, lets just stick to numbers. But then you have non decimal number systems....
Pfhorrest wrote:As someone who is not easily offended, I don't really mind anything in this conversation.
Mighty Jalapeno wrote:It was the Renaissance. Everyone was Italian.

User avatar
gmalivuk
GNU Terry Pratchett
Posts: 26453
Joined: Wed Feb 28, 2007 6:02 pm UTC
Location: Here and There
Contact:

Re: 1179: "ISO 8601"

Postby gmalivuk » Thu Feb 28, 2013 5:21 am UTC

Are there any groups of people who have a written language but who don't know the decimal system?
Unless stated otherwise, I do not care whether a statement, by itself, constitutes a persuasive political argument. I care whether it's true.
---
If this post has math that doesn't work for you, use TeX the World for Firefox or Chrome

(he/him/his)

Ashtar
Posts: 24
Joined: Sun Jun 03, 2012 5:56 am UTC

Re: 1179: "ISO 8601"

Postby Ashtar » Thu Feb 28, 2013 5:24 am UTC

Everybody is wrong. The current date and time is 8400108.1331, that is, one shock eight days since the start of the eight shock fortieth Ashtrian bimester, thirteen palas and thirty-one ghatikas.

Jamaican Castle
Posts: 151
Joined: Fri Nov 30, 2007 9:10 pm UTC

Re: 1179: "ISO 8601"

Postby Jamaican Castle » Thu Feb 28, 2013 5:25 am UTC

EvanED wrote:I sort of agree. But not really, because reality intercedes.

Yes, it would be nice if software sorted correctly given DD-MM-YYYY or MM-DD-YYYY. But... it doesn't. Nautilus doesn't, Windows explorer doesn't, ls doesn't. I'd be surprised if any remotely-commonly used file tool did. (The most likely one would be Finder, but I don't have a Mac to try.)

It's hard enough to convince programmers that "10 - foo" should sort after "2 - foo" :-).


On the topic of sortable file names - doesn't the Date Created (or Date Modified, if appropriate) attribute provide that already? Even for non-files, it seems like odd to store the date as plaintext, rather than some sort of automatically-formatted date object, unless it's in the name. And the name is an awkward place to put metadata to begin with, let alone when people don't agree on the format.

EvanED
Posts: 4331
Joined: Mon Aug 07, 2006 6:28 am UTC
Location: Madison, WI
Contact:

Re: 1179: "ISO 8601"

Postby EvanED » Thu Feb 28, 2013 5:31 am UTC

Jamaican Castle wrote:On the topic of sortable file names - doesn't the Date Created (or Date Modified, if appropriate) attribute provide that already? Even for non-files, it seems like odd to store the date as plaintext, rather than some sort of automatically-formatted date object, unless it's in the name. And the name is an awkward place to put metadata to begin with, let alone when people don't agree on the format.

No, not for a lot of things. For instance, I typically download several months' worth of bank statements at a time because I'm too lazy to log onto several different accounts once a month, and they get named things like 2012-12-31.pdf, but all have basically the same creation time.

annodomini
Posts: 1
Joined: Thu Feb 28, 2013 5:30 am UTC

Re: 1179: "ISO 8601"

Postby annodomini » Thu Feb 28, 2013 5:37 am UTC

Now we just need a comic for ITU-T E.123, and we'll be all set.

User avatar
orthogon
Posts: 2955
Joined: Thu May 17, 2012 7:52 am UTC
Location: The Airy 1830 ellipsoid

Re: 1179: "ISO 8601"

Postby orthogon » Thu Feb 28, 2013 7:59 am UTC

commodorejohn wrote:I say we start dating by Roman emperors again.

I totally overlooked the word "by" in that sentence the first time I read it.

Since somebody mentioned UTC, I know GMT is better known, but I tend to say "UTC" as a small gesture of internationalism in return for everyone else agreeing to use our solar time as a reference.

My pet issue is invitations that use "GMT" to mean "London/Dublin/Lisbon time", even during the summer. You're likely to get all the engineers turning up an hour later than everyone else. It's our own fault though. In the US you can say "E.T." to mean the time in NY on that day, and specify "E.D.T." or "E.S.T." if you need to be unambiguous. There's no equivalent notation to E.T. for London time.

Then there's "Central European Time", which is the time in western Europe, not central Europe, and may or may not take account of daylight saving.
xtifr wrote:... and orthogon merely sounds undecided.

User avatar
22tcp
Posts: 7
Joined: Thu Nov 17, 2011 11:37 pm UTC
Location: Midtown DA, DE

ISO - another naked emperor

Postby 22tcp » Thu Feb 28, 2013 8:04 am UTC

Morning.
Strange. Am I the only one that sees an abomination in even to try to set a world standard of time writing that is definitely based on a superstition about events in a fictional past? This is actually a lively act of neglection or even racism towards any people who do not use that sort of time reckoning at all. And you bicker about usability in programming... the code to handle different date formats is in place, is it not? Also. It is easily observable that ISO became a money generator for quite a bunch of people. Well done and thumbs up, that's one hell of a perpetuum mobile. All one had to do was to persuade some politicians to not accept contractors without certain certifications regardless of their actual quality so companies involved in ISO had an advantage. But I am strolling off-topic. Though, reading along the comments that was well deserved and needed. To the real question, Watson. How about a standard that uses some fixable point in time folks can agree on worldwide (so it should be based on science, shouldn't it?). And even then it will be a very slow process to get that kind of change for the everyday life. It will take generations, I guess. But it still has a better chance than using something based on a sole religion. That's exactly the kind of arrogance I expect from the roman empire - not from modern engineers.

Jamaican Castle
Posts: 151
Joined: Fri Nov 30, 2007 9:10 pm UTC

Re: 1179: "ISO 8601"

Postby Jamaican Castle » Thu Feb 28, 2013 8:18 am UTC

I suspect that by the time anyone cared to devise a universal timekeeping schema, the current calendar and all its baggage were simply too widespread to be worth changing. A change in timekeeping would have to adjust billions of individual calendars and clocks to the new standard, not to mention all the past references that would have to be translated. It simply isn't worth going to that much trouble.

Particularly since the baseline for your new calendar would have to be chosen just as arbitrarily. You could settle on anything, of course. A particular celestial event. A holiday. Your birthday. The Day the Music Died. It literally does not matter; any day or year is just as valid as any other.

CharlieP
Posts: 397
Joined: Mon Dec 17, 2012 10:22 am UTC
Location: Nottingham, UK

Re: 1179: "ISO 8601"

Postby CharlieP » Thu Feb 28, 2013 9:05 am UTC

Diadem wrote:May I take this time to complain about people using UTC instead of GMT?

Everyone knows what GMT means, almost no one knows what UTC is, yet some people insist on using UTC, despite those two terms meaning the same thing (not quite, but the difference is never more than a second).


As a counterpoint, I live in the UK but always use UTC when addressing an international audience, because I think of GMT as just another local offset:

BST: UTC+1
CET: UTC+1
CEST: UTC+2
EDT: UTC-4
EST: UTC-5
GMT: UTC+0

So to my mind, it's right and proper for New Yorkers to talk about "12:00 EST" and for Londoners to talk about "12:00 GMT" in local contexts, but UTC remains the baseline that everything else is compared to (even if in practice GMT and UTC are "the same thing").

You also have the issue that we Brits only use GMT for five months of the year, and referring to GMT times when we're on British Summer Time can be ambiguous and confusing.
This is my signature. There are many like it, but this one is mine.

User avatar
ManaUser
Posts: 284
Joined: Mon Jun 09, 2008 9:28 pm UTC

Re: 1179: "ISO 8601"

Postby ManaUser » Thu Feb 28, 2013 9:13 am UTC

EvanED wrote:Yes, it would be nice if software sorted correctly given DD-MM-YYYY or MM-DD-YYYY. But... it doesn't. Nautilus doesn't, Windows explorer doesn't, ls doesn't. I'd be surprised if any remotely-commonly used file tool did.

One good reason why it's better do what they computer understands rather than trying to make it understand what we do is that every special case you program in becomes a potential trap. Say they made Windows Explorer automatically interpret dates in file names and sort accordingly. Then someone comes along and make files called "David-21.jpg", "Chuck-30.jpg", "April-26.jpg", "Marsha-22.jpg"... And where the hell did April's picture go?
Oh, there it is, right above "2032-10th St Inspection.txt".
Last edited by ManaUser on Thu Feb 28, 2013 9:44 am UTC, edited 3 times in total.

User avatar
Klear
Posts: 1965
Joined: Sun Jun 13, 2010 8:43 am UTC
Location: Prague

Re: 1179: "ISO 8601"

Postby Klear » Thu Feb 28, 2013 9:13 am UTC

orthogon wrote:
commodorejohn wrote:Then there's "Central European Time", which is the time in western Europe, not central Europe, and may or may not take account of daylight saving.


Seems like it goes right through the middle of Europe to me: http://upload.wikimedia.org/wikipedia/commons/thumb/4/40/Timezones2008_UTC%2B1_gray.png/800px-Timezones2008_UTC%2B1_gray.png

It's just that Spain, France, and Benelux are tacked on for convenience.

Smylers
Posts: 8
Joined: Wed Nov 16, 2011 12:18 pm UTC

Re: 1179: "ISO 8601"

Postby Smylers » Thu Feb 28, 2013 9:34 am UTC

xenotrout wrote: ...I'm in the USA) ...

maxmaxmaxmax wrote:any explanation on 1330300800?


Unix timestamp for 2012-02-26T16:00[1] 2012-02-27?

[1] GNU date interpreting input date in local and outputting UTC?


Isn't it actually doing the opposite of that? date by default uses your local timezone (for both input and output). But in this case the input was 1330300800, which is the number of seconds since the start of 1970 UTC; by giving an epoch time you're effectively specifying a UTC time: there isn't a 1330300800 in your local timezone. The output is in the default of your local timezone.

That also fits with your being in the USA (presumably the west coast): your original output was a time 8 hours before the UTC output — in other words, when it's midnight in Greenwhich, it's still only 16:00 the previous afternoon where you are.

Anyone else annoyed it doesn't seem to be able to parse ISO-8601?


date -d 2010-10-23 works for me. There's even an example like that in its manpage.

marcgaston
Posts: 12
Joined: Fri Feb 18, 2011 8:18 am UTC

Re: 1179: "ISO 8601"

Postby marcgaston » Thu Feb 28, 2013 9:51 am UTC

Gargravarr wrote:
Oktalist wrote:
Gargravarr wrote:When the alien invasion fleet lands, official documents will be stamped with Planck time units, with the Big Bang as reference date, written in base-64. That will solve a few problems :wink:

They will also need a system for encoding inertial reference frames, or else any reference to a standard time would be meaningless.

Well, I didn't say it would solve all problems

Edit: Bleh, that was a really weak comeback :oops:



Yes I like the idea of Planck units: It's surprisingly usable; when my calculations are not off, you would need only 36 characters to encode todays date (should be around 8,019E+60 Planck times). Please, someone write a function to do that!

Of course we would have to agree on an epoch date nevertheless since we do not know the exact number of Plank time units since the big bang. :(

Protoform
Posts: 43
Joined: Sun Dec 30, 2012 1:09 pm UTC
Location: The Nether

Re: 1179: "ISO 8601"

Postby Protoform » Thu Feb 28, 2013 10:28 am UTC

gmalivuk wrote:So some people say "half four" to mean 4:30 instead of 2:00


Serious question: would that not be 3:30, or does English differ from Dutch there? When we say 'half vier', we generally mean 3:30.

User avatar
Diadem
Posts: 5654
Joined: Wed Jun 11, 2008 11:03 am UTC
Location: The Netherlands

Re: 1179: "ISO 8601"

Postby Diadem » Thu Feb 28, 2013 10:37 am UTC

Clearly we need to redesign dates entirely.

My suggestion: Days start when the sun comes up, and end when the next day starts. So 3 o'clock shall always mean 3 hours after the sun came up. There shall be 13 months of exactly 28 days each. Dates shall be giving DD-MM-YYYY. The year shall start at the exact moment that the sun is in its northern solstice - regardless of when this falls in the day. This particular day where the year changes shall be counted as part of the new year, and dated 00-00-xxxx. On most years, this spare days shall conveniently fall immediately after 28-13-xxxx, making the calendar add up nicely. Approximately once every 4 years though, the new year shall fall 2 days after 28-13, giving one spare day. This day shall be known as 'leap day' (written down LD-YYYY) and shall be a national day of reflection, where humanity shall reflect upon the stupidity of the creator for not giving the year an integer number of days.

This system is perfect. Not only will dates finally make sense, but so will times!


Protoform wrote:
gmalivuk wrote:So some people say "half four" to mean 4:30 instead of 2:00

Serious question: would that not be 3:30, or does English differ from Dutch there? When we say 'half vier', we generally mean 3:30.

when English people say "half four" they mean "Half past four" and are omitting the word 'past'. When Dutch people say "Half four" they mean "Half to four" and are omitting the word 'to'. So yeah, English and Dutch are different here.
It's one of those irregular verbs, isn't it? I have an independent mind, you are an eccentric, he is round the twist
- Bernard Woolley in Yes, Prime Minister

User avatar
yurell
Posts: 2924
Joined: Sat Nov 13, 2010 2:19 am UTC
Location: Australia!

Re: 1179: "ISO 8601"

Postby yurell » Thu Feb 28, 2013 10:55 am UTC

Most people I know say 'four-thirty' instead of either.
cemper93 wrote:Dude, I just presented an elaborate multiple fraction in Comic Sans. Who are you to question me?


Pronouns: Feminine pronouns please!

User avatar
Klear
Posts: 1965
Joined: Sun Jun 13, 2010 8:43 am UTC
Location: Prague

Re: 1179: "ISO 8601"

Postby Klear » Thu Feb 28, 2013 11:00 am UTC

Diadem wrote:Clearly we need to redesign dates entirely.


Image

The XKCD memetic completeness conjecture strikes again!

Ekaros
Posts: 32
Joined: Fri Jan 13, 2012 8:37 am UTC

Re: 1179: "ISO 8601"

Postby Ekaros » Thu Feb 28, 2013 11:05 am UTC

Diadem wrote:Clearly we need to redesign dates entirely.

My suggestion: Days start when the sun comes up, and end when the next day starts. So 3 o'clock shall always mean 3 hours after the sun came up. There shall be 13 months of exactly 28 days each. Dates shall be giving DD-MM-YYYY. The year shall start at the exact moment that the sun is in its northern solstice - regardless of when this falls in the day. This particular day where the year changes shall be counted as part of the new year, and dated 00-00-xxxx. On most years, this spare days shall conveniently fall immediately after 28-13-xxxx, making the calendar add up nicely. Approximately once every 4 years though, the new year shall fall 2 days after 28-13, giving one spare day. This day shall be known as 'leap day' (written down LD-YYYY) and shall be a national day of reflection, where humanity shall reflect upon the stupidity of the creator for not giving the year an integer number of days.

This system is perfect. Not only will dates finally make sense, but so will times!


Just one question is this one done on certain spot on earth? Or localy, if localy what about northern and souther regions where there is possibility that sun doesn't come up or go down and also there is certain drift of sun coming up and going down as such days would be different length?

Smylers
Posts: 8
Joined: Wed Nov 16, 2011 12:18 pm UTC

Re: 1179: "ISO 8601"

Postby Smylers » Thu Feb 28, 2013 11:16 am UTC

cwolves wrote:this is the wrong way to way to write a date. Why? because some implementations of common languages (such as JavaScript) won't parse `2013-02-26`


Which implementations?

JavaScript's Date.parse() is defined as accepting an ISO 8601 date.

User avatar
Diadem
Posts: 5654
Joined: Wed Jun 11, 2008 11:03 am UTC
Location: The Netherlands

Re: 1179: "ISO 8601"

Postby Diadem » Thu Feb 28, 2013 11:53 am UTC

Ekaros wrote:
Diadem wrote:Clearly we need to redesign dates entirely.

My suggestion: Days start when the sun comes up, and end when the next day starts. So 3 o'clock shall always mean 3 hours after the sun came up. There shall be 13 months of exactly 28 days each. Dates shall be giving DD-MM-YYYY. The year shall start at the exact moment that the sun is in its northern solstice - regardless of when this falls in the day. This particular day where the year changes shall be counted as part of the new year, and dated 00-00-xxxx. On most years, this spare days shall conveniently fall immediately after 28-13-xxxx, making the calendar add up nicely. Approximately once every 4 years though, the new year shall fall 2 days after 28-13, giving one spare day. This day shall be known as 'leap day' (written down LD-YYYY) and shall be a national day of reflection, where humanity shall reflect upon the stupidity of the creator for not giving the year an integer number of days.

This system is perfect. Not only will dates finally make sense, but so will times!


Just one question is this one done on certain spot on earth? Or localy, if localy what about northern and souther regions where there is possibility that sun doesn't come up or go down and also there is certain drift of sun coming up and going down as such days would be different length?

Yes, days will have different length depending latitude and time of the year. Makes sense, doesn't it?
It's one of those irregular verbs, isn't it? I have an independent mind, you are an eccentric, he is round the twist
- Bernard Woolley in Yes, Prime Minister

User avatar
Klear
Posts: 1965
Joined: Sun Jun 13, 2010 8:43 am UTC
Location: Prague

Re: 1179: "ISO 8601"

Postby Klear » Thu Feb 28, 2013 12:48 pm UTC

Diadem wrote:Yes, days will have different length depending latitude and time of the year. Makes sense, doesn't it?


I don't see how days stopping being a unit of time is useful.

User avatar
Diadem
Posts: 5654
Joined: Wed Jun 11, 2008 11:03 am UTC
Location: The Netherlands

Re: 1179: "ISO 8601"

Postby Diadem » Thu Feb 28, 2013 12:52 pm UTC

Klear wrote:
Diadem wrote:Yes, days will have different length depending latitude and time of the year. Makes sense, doesn't it?


I don't see how days stopping being a unit of time is useful.

Just what I'd expect from a hidebound reactionary :)
It's one of those irregular verbs, isn't it? I have an independent mind, you are an eccentric, he is round the twist
- Bernard Woolley in Yes, Prime Minister

User avatar
Carlington
Posts: 1588
Joined: Sun Mar 22, 2009 8:46 am UTC
Location: Sydney, Australia.

Re: 1179: "ISO 8601"

Postby Carlington » Thu Feb 28, 2013 12:59 pm UTC

Gargravarr wrote:To summarize, ISO 8601 is
- Unambiguous
- Sortable
- Understandable (unlike eg Stardate or "seconds since reference date")
- Language independant
- Context independant
- Scalable (need more precision? Just add a number at the end)

If that's not a check mate, I don't know what would be.

Needs to be standard in all reference frames, otherwise it's not going to cut it, I'm afraid.

According to phlip, this was posted 02/13/23, 02:21:38
Kewangji: Posdy zwei tosdy osdy oady. Bork bork bork, hoppity syphilis bork.

Eebster the Great: What specifically is moving faster than light in these examples?
doogly: Hands waving furiously.

Please use he/him/his pronouns when referring to me.


Return to “Individual XKCD Comic Threads”

Who is online

Users browsing this forum: NonchalantTurtle and 48 guests