We all know ISO date format of YYYY-MM-DD, and the advantages of it like easy sorting. But for some reason local inconsistent and confusing date formats continue to be in use. Since people will always get the fields wrong, lets standardize a correct separator with each of the 6 permutations. Hopefully no one has got the idea to start writing the year in the middle, which reduces the number of permutations to 4, but if someone still measures in body parts and sports venues, how can I be sure?
The date format in the article header was adjusted accordingly to be consistent with the proposals in this article. Back then Blogger probably had a lesser selection of date formats. It's an abhorrent discrimination that there's still no custom format support.
AMD announced 2 announcements, where the date of the 1st announcement was given "10.08.2020", which I thought was August, therefore I had missed it. It was only after I had read the date of the 2nd announcement given as "10.28.2020" when I realized they misused the date format. Please, cut the salary of the guy who is responsible for this. Dots are a Central to Eastern European practice coming from ordinal number notation, and that practice is *always* D.M.Y, and *never* M.D.Y. You should have used slashes (like 10/08/2020) or written the year 1st (like ISO-standard 2020-10-08).
YY-MM-DD -- 20-09-02
Shortened from the ISO format, still retains its advantages, albeit not very Y2.1K compliant. However if management and elected officials cannot fathom the consequeces of their (in)actions even in 5 years, there's no need to bother. No file in filesystem of current lineage could have been created prior to 1960s anyway. Similarly, the UNIX timestamp being signed doesn't make sense, especially since it's not defined precisely for dates before 1972-01-01 (UTC and GMT seconds used to be longer instead of leaping), and with unsigned one we are good through 2106. Furthermore, humans can't live longer than 127 years without being exposed to relativistic effects or incrementally replaced in a ship of Theseus style. That means you don't need more than 7 candles on a birthday cake.
DD.MM.YY -- 02.09.20, 2. 9. '20, 2. sep 2020, or 02sep20
Popular in Central and Eastern Europe. Western Europeans use /s, which gets confusing with American field order. Usually there are spaces after the dots, but that just wastes space. Also zero-padding isn't that common and the month is sometimes spelled out without the dots. The DDmmmYY variant is quite popular on signs on 2b2t.org. Nevertheless this has no impact on the order of fields. Dot is also not a good character to use in filenames, but at least it's not a /, which is illegal even on Linux.
MM/DD/YY -- 09/02/20, or Sep 2, 2020
This one is popular over the ocean. Extra confusing when the year is left out, since the other order is widely used in Western Europe. Some have misread 9/11 as 9th of November, never mind the confusing month names. The ISO standard for this is --MM-DD, not MM/DD. Also / as a separator is a bad choice for filenames, since it's used as a directory separator on Linux, thus is the only illegal character, and is arbitrarily forbidden in Windows. You could argue the date is supposed to go hand in hand with a directory structure, which doesn't hold water with this crap field order, so you get YYYY/MM/DD (or YYYY\MM\DD if you still use Window$), which is the ISO version. Is there even any OS that uses something else as a directory separator?
YY\DD\MM -- 20\02\09
This is MM/DD/YY but in reverse, so it gets reverse solidi, or backslashes. Good for Muslim Americans, I guess. Same stuff as above for filenames on Windows, but on Linux it can at least pass with a bunch of escaping, like 20\\02\\09.
MM|YY|DD -- 09|20|02
DD*YY*MM -- 02*20*09
These 2 formats are so retarded I give them retarded separators too. The | descends from overseas (mis)practice of MM/DD, and the * descends from DD.MM.. ~ was deemed too similar to -, besides I use it for denoting ranges, like 2020-07-04~2020-07-14, instead of the confusing ISO nonsense. Thus _ remains reserved for programming languages restrictive on identifiers.
Of course we must not forget about our East Asian comrades, which use the genius unambiguous separators of 年 for year, 月 for month, and 日 for day, even though when we look at it, it's basically good old YY-MM-DD, which may happen to be displayed as YY☐MM☐DD☐ or YY�MM�DD� when your eurocentric font repository is lacking. Also we must not forget about those East Asian comrades who decided to make their own alphabets or abugidas or customized other's to their liking, instead of borrowing Chinese characters, possibly squiggling them so badly it became ひらがな, or simplifying them so much it became カタカナ or ㄅㄛㄆㄛㄇㄛㄈㄛ AKA ㄓㄨˋ ㄧㄣ.
There is 1 extra bad habit of writing the month in Roman numbers, which is deadly in combination with some people writing 1 as a mere |. A math teacher once told us, someone mistook 5. ||. 19xx on their certificate for 5. 2. 19xx and it took years to have this corrected in all the systems it has spread trough.
Another separate but related by far not as bad habit is writing years in Roman numbers, further augmented by the use of overlong forms like MCMXCIX over MIM. When the credits of 90s movies scroll too quick, I can't read the year at the end when written in this way. I presume this practice will have to go in the year MMMIM or MMMCMXCIX, as there is no standard way to deal with values above that, other than violating the standards and stacking more than 3 Ms in row. Some promising consistency and range can be found in putting lines above (vinculi) for each 3 additional zeroes, up to M̿M̿M̿IM̿ (3'999'999'999) if it can render. Etruscan apostrophus method needs ((((|)))) to reach a million and (((((((|))))))) to reach a milliard, starting to look like a LISP program.
On the same page in my excercise book, there's a polemic that you can't just be covering your face niqab-style (especially in 35°C heat or nights, that's suss, in winter it may be OK), since you are then unidentifiable, thus a policeman can request you to identify yourself, revealing the face for comparison against the ID and probably even charge you a fine. Now that gets very confusing when the same government that forbids face covering orders mandatory face masks, which means covering face in certain public places is now required, in contradiction to previous policy about covering faces. Reportedly Chinese have an AI that can identify people based on just their eyes and hair, which reminds me of a history test, when there was not enough space at the bottom for full faces of dictators and presidents, so we had to guess just based on the eyes in black and white. Luckily, mongoloid eyes are different from both europoid and negroid ones, the black skin is clearly identifiable even in black and white, and old people have more wrinkles. Also, retina checks are also used for secure access, which doesn't impede wearing niqabs (burqas are still problematic), so we may be here onto something. If only that overblown American police budget was spent on equipping officers with retina readers, in conjunction with biometric IDs with all sorts of data, instead of excessive militarization.
YY-MM-DD -- 20-09-02
Shortened from the ISO format, still retains its advantages, albeit not very Y2.1K compliant. However if management and elected officials cannot fathom the consequeces of their (in)actions even in 5 years, there's no need to bother. No file in filesystem of current lineage could have been created prior to 1960s anyway. Similarly, the UNIX timestamp being signed doesn't make sense, especially since it's not defined precisely for dates before 1972-01-01 (UTC and GMT seconds used to be longer instead of leaping), and with unsigned one we are good through 2106. Furthermore, humans can't live longer than 127 years without being exposed to relativistic effects or incrementally replaced in a ship of Theseus style. That means you don't need more than 7 candles on a birthday cake.
DD.MM.YY -- 02.09.20, 2. 9. '20, 2. sep 2020, or 02sep20
Popular in Central and Eastern Europe. Western Europeans use /s, which gets confusing with American field order. Usually there are spaces after the dots, but that just wastes space. Also zero-padding isn't that common and the month is sometimes spelled out without the dots. The DDmmmYY variant is quite popular on signs on 2b2t.org. Nevertheless this has no impact on the order of fields. Dot is also not a good character to use in filenames, but at least it's not a /, which is illegal even on Linux.
MM/DD/YY -- 09/02/20, or Sep 2, 2020
This one is popular over the ocean. Extra confusing when the year is left out, since the other order is widely used in Western Europe. Some have misread 9/11 as 9th of November, never mind the confusing month names. The ISO standard for this is --MM-DD, not MM/DD. Also / as a separator is a bad choice for filenames, since it's used as a directory separator on Linux, thus is the only illegal character, and is arbitrarily forbidden in Windows. You could argue the date is supposed to go hand in hand with a directory structure, which doesn't hold water with this crap field order, so you get YYYY/MM/DD (or YYYY\MM\DD if you still use Window$), which is the ISO version. Is there even any OS that uses something else as a directory separator?
YY\DD\MM -- 20\02\09
This is MM/DD/YY but in reverse, so it gets reverse solidi, or backslashes. Good for Muslim Americans, I guess. Same stuff as above for filenames on Windows, but on Linux it can at least pass with a bunch of escaping, like 20\\02\\09.
MM|YY|DD -- 09|20|02
DD*YY*MM -- 02*20*09
These 2 formats are so retarded I give them retarded separators too. The | descends from overseas (mis)practice of MM/DD, and the * descends from DD.MM.. ~ was deemed too similar to -, besides I use it for denoting ranges, like 2020-07-04~2020-07-14, instead of the confusing ISO nonsense. Thus _ remains reserved for programming languages restrictive on identifiers.
YY年MM月DD日 -- 20年09月02日
YY년 MM월 DD일 -- 20년 09월 02일
Ngày DD, tháng MM, năm YY -- Ngày 02, tháng 09, năm 20
Of course we must not forget about our East Asian comrades, which use the genius unambiguous separators of 年 for year, 月 for month, and 日 for day, even though when we look at it, it's basically good old YY-MM-DD, which may happen to be displayed as YY☐MM☐DD☐ or YY�MM�DD� when your eurocentric font repository is lacking. Also we must not forget about those East Asian comrades who decided to make their own alphabets or abugidas or customized other's to their liking, instead of borrowing Chinese characters, possibly squiggling them so badly it became ひらがな, or simplifying them so much it became カタカナ or ㄅㄛㄆㄛㄇㄛㄈㄛ AKA ㄓㄨˋ ㄧㄣ.
There is 1 extra bad habit of writing the month in Roman numbers, which is deadly in combination with some people writing 1 as a mere |. A math teacher once told us, someone mistook 5. ||. 19xx on their certificate for 5. 2. 19xx and it took years to have this corrected in all the systems it has spread trough.
Another separate but related by far not as bad habit is writing years in Roman numbers, further augmented by the use of overlong forms like MCMXCIX over MIM. When the credits of 90s movies scroll too quick, I can't read the year at the end when written in this way. I presume this practice will have to go in the year MMMIM or MMMCMXCIX, as there is no standard way to deal with values above that, other than violating the standards and stacking more than 3 Ms in row. Some promising consistency and range can be found in putting lines above (vinculi) for each 3 additional zeroes, up to M̿M̿M̿IM̿ (3'999'999'999) if it can render. Etruscan apostrophus method needs ((((|)))) to reach a million and (((((((|))))))) to reach a milliard, starting to look like a LISP program.
On the same page in my excercise book, there's a polemic that you can't just be covering your face niqab-style (especially in 35°C heat or nights, that's suss, in winter it may be OK), since you are then unidentifiable, thus a policeman can request you to identify yourself, revealing the face for comparison against the ID and probably even charge you a fine. Now that gets very confusing when the same government that forbids face covering orders mandatory face masks, which means covering face in certain public places is now required, in contradiction to previous policy about covering faces. Reportedly Chinese have an AI that can identify people based on just their eyes and hair, which reminds me of a history test, when there was not enough space at the bottom for full faces of dictators and presidents, so we had to guess just based on the eyes in black and white. Luckily, mongoloid eyes are different from both europoid and negroid ones, the black skin is clearly identifiable even in black and white, and old people have more wrinkles. Also, retina checks are also used for secure access, which doesn't impede wearing niqabs (burqas are still problematic), so we may be here onto something. If only that overblown American police budget was spent on equipping officers with retina readers, in conjunction with biometric IDs with all sorts of data, instead of excessive militarization.
