Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

ISO 8601: a better date format

kirby.kevinson.org

Post reply on HN
Print view View original

21–30 of 437 posts

Re: ISO 8601: a better date format

#21

Post by CyberRabbi »

hannob wrote: post #9

There's another aspect you may not know about: It's close to how it's done in chinese, which is 2021年3月17日. They use different separators (the chinese words for year, month and day), but it's close enough. Given that china is the country with the largest population it may not be a bad idea to use something a large number of people will easily understand.

Maybe we should also start speaking Chinese since it may not be a bad idea to use something a large number of people will easily understand.

Edit: this is not sarcasm. It’s simple logic, unless there is something wrong with learning other languages.

Re: ISO 8601: a better date format

#22

Post by Someone »

tpetry wrote: post #17

The article is missing one of the best features of iso 8601: it‘s string form is naturally sortable. You don‘t need any specific logic for sorting, so e.g. your filesystem will automatically sort files correctly if you prefix them with a iso 8601 date.

FTA:

“The elements are always padded to the maximum number of digits. This not only makes all of the dates look equally nice, but also, coupled with other quirks of this format, allows the files with the date in the name to be sorted just by the filename.”

Re: ISO 8601: a better date format

#23

Post by wodenokoto »

People look at ISO-8601 and think year-month-day looks great! But don’t realize that it also includes thing like —month-day (yes, you don’t need to include year to have a valid iso date!) year-week, year-day and lots of other things.

So I would generally advise against accepting iso formatted dates and instead decree a single date format.

https://en.m.wikipedia.org/wiki/ISO_8601

Re: ISO 8601: a better date format

#25

Post by kirbykevinson »

eesmith wrote: post #18
kirbykevinson wrote: post #10

Earlier quoted context omitted.

I omitted this detail on purpose to not distract the reader who doesn't know much about date formats.

How would you describe the accuracy and veracity of > "Here there’s only one correct way to write a date," given that there are multiple correct ways, and the way shown wasn't actually ISO 8601 but RFC 3339?

Yes, technically there are some edge cases where you can write things differently (apart from what you mentioned, one is, for example, writing the UTC time zone by Z vs +00:00 and another is adding additional precision to the date), but for the end user it's basically the same. Let's not ruin the illusion here :)

Re: ISO 8601: a better date format

#26

Post by mjevans »

kitd wrote: post #2

The real purpose of endianness which is missed here is the ability to sort. Sorting a list of iso8601 dates alphabetically means sorting them chronologically too.

That's also the only time (full) European dates make sense, when writing them on the right hand side of a ledger page so that the exterior of the ledger / book is indexed with the largest components on the edge of the page.

US dates only make sense if you don't care about the year and are writing on the left most side of a page; that might make sense if ink or paper is at a premium.

Re: ISO 8601: a better date format

#27

Post by rollulus »

At the risk of nitpicking, I prefer RFC 3339, which is a profile of ISO 8601. ISO 8601 provides room for things you typically don't want, e.g. dropping time zone offsets.

While I'm typing this I realise that I've never actually read the ISO 8601 spec, like probably most people here, because it's not free. I cannot even be sure it really exists.

Re: ISO 8601: a better date format

#28

Post by systemvoltage »

No one here is talking about practicality for non-software engineers which is probably 99.9% of the population. It might be fine for sorting and no one gives a shit about endianess. Common person would argue - why do they need to know the year first? Most people encounter dates in the same year, they already know we're in 2021 and it is not as important. Having to laboriously parse YYYY before getting to the most important bits, either month or day depending on what they're trying to do; doesn't seem too practical. I hate writing ISO 8601 dates, even though I know it is logical.

Not everything logical works with humans. In my day to day life (not engineering use), I personally prefer inches and feet, they are just human scale dimensions. Conversion happens less often than you think - most people convey to others that this box is 12x8x8 inches. Conversion from inches to feet happens way less often. I hate fractions though. It would be great if a centimeter was twice as long in physical dimensions. It's just too granular and falls in an uncanny valley. And no one uses decimeter (10 cm) for some reason which is actually nice since it is roughly the size of the index finger.

Re: ISO 8601: a better date format

#30

Post by stochastimus »

The unmentioned benefits: it is fixed-width and follows most to least significant parts left-to-right. What this means is that the significance of the parts follows the same L-R convention as numbers, making it intuitively obvious which is the month and which is the day. Oh, and a simple string sort will sort timestamps correctly - very handy at the command line. Even db columns holding timestamps as strings will get the ordering right by default.
Post reply on HN