6 ms·
It is only in the last century or two that humanity has reached the technological and economic level that a single standard global year numbering made any sense
by skissane 1y ago
It is only in the last century or two that humanity has reached the technological and economic level that a single standard global year numbering made any sense-before that, year numbering systems were highly culture-specific. And it just happened by accident that Western European-derived cultures were globally ascendant at that point in history, so that culture’s calendar became the de facto global standard. And if history turned out a bit differently, it easily could have been another calendar instead - e.g. the Islamic chronology (AH).
But now we have a single global standard, I think there is a huge amount of inertia against changing it - it is baked into untold millions of computer systems and business processes now.
I think the most likely way it might change would be (a) if humanity collapsed back to a premodern civilisation, and later recovered; or (b) some new culture/religion became globally dominant which demanded the calendar be changed.
Personally, I’m sceptical (a) is going to happen in the next few thousand years. I think the most likely scenarios are (i) technological modernity survives, (ii) humanity goes extinct completely, (iii) a more moderate collapse in which things get very messy but don’t go all the way back to the premodern era. I think all three are more likely than the kind of complete and extended collapse then eventual recovery which would be most likely to reboot the calendar into a new and different global standard.
I don’t think odds of (b) are high-it would require not just a new dominant culture, but also one which felt very strongly about wiping out all traces of the old calendar. Suppose 1000 years from now, 99% of humans are devout Muslims-I personally think that’s rather unlikely to happen, but anything is possible-would that trigger the current year numbering to be replaced by the Islamic one? I’m sceptical-all Muslim majority countries currently heavily use the Gregorian calendar for business use, computer systems, etc, and they don’t have a theological issue with that, so I’m sceptical they’d feel the need to change even if Islam became the globally dominant culture. And this isn’t a new thing in history-many historical Islamic empires continued the use of pre-Islamic calendars in parallel with the Islamic, especially since the Islamic calendar, being purely lunar, was less than ideal for agricultural use.
- dmurray 1y ago> But now we have a single global standard, I think there is a huge amount of inertia against changing it - it is baked into untold millions of computer systems and business processes now. We have two global standards, both from the Western Christian tradition. The Gregorian calendar, and Unix time. Order of magnitude, they're probably baked into the same number of processes.
- kragen 1y agoWhat's Christian about Unix time? I don't recall ever reading that either Thompson or Ritchie was theist at all, much less Christian. And the division of the solar day into 24 hours of 60 minutes of 60 seconds is not Christian but, mostly, Sumerian: both pagan and Eastern.
- dmurray 1y agoIt's an offshoot of the Gregorian calendar, being offset from it by a whole number of years. And Thompson and Ritchie grew up and worked in a majority-Christian country: their individual faith doesn't have much to do with it.
- skissane 1y agoThe Gregorian calendar is essentially non-Christian in origin. Yes, it is named for Pope Gregory, but he just tinkered with the leap year rule, other than that it is essentially the same as the Julian calendar, named for Julius Caesar Now, the common AD/CE year numbering is obviously Christian in motivation, but that’s technically distinct from the Gregorian calendar-the Julian calendar historically used other year numbering systems too (ab urbe condita, anno Diocletiani aka anno martyrum, anno mundi), and the Gregorian calendar itself doesn’t care what numbers you assign to years beyond modulo 400 - you could replace 2025 with 2425 or 12025 and it would still work fine. But that’s the thing, Unix time doesn’t care whether 1970 is called 1970 or 1570 or 2370 or 11970, only how distant you are from it - so it isn’t really tied to our Christianity-inspired year numbering, only to 1970-01-01T00:00:00Z as an instant I think CE + 10,000 - called “Holocene Era” (HE), also sometimes expanded Human Era or Historical Era - is a good proposal - high compatibility with CE (you can just pretend 2025 is missing the leading digit), makes all dates from recorded human history positive. Okay, is ambiguous in that 2025 CE might get confused with 2025 HE (= -7975 CE = 7976 BCE), but in practice that’s unlikely since that’s prehistorical and essentially nothing (outside astronomy) from that long ago can be dated to a year’s accuracy anyway
- DonHopkins 1y agoUnix time is explicitly anti-Christian! https://en.wikipedia.org/wiki/BSD_Daemon https://en.wikipedia.org/wiki/BSD_Daemon >The copyright on the official BSD Daemon images is owned by Marshall Kirk McKusick, a very early BSD developer who worked with Bill Joy. McKusick has freely licensed the mascot for individual "personal use within the bounds of good taste (an example of bad taste was a picture of the BSD Daemon blowtorching a Solaris logo)."[ Any use requires both a copyright notice and attribution. Blowtorching a Solaris logo sounds like it's in good taste to me! Just not blowing it without a torch.
- rollcat 1y agoI see another possibility, on topic with TFA. We start colonising other worlds, and will need to coordinate timekeeping between them. Time dilation will also kick in (GPS already has to compensate). An Earth year is a year only on Earth.
- skissane 1y agoIf humans were to colonise Mars, I suspect they might adopt a hybrid calendar, combining the Martian day with the Earth year I think for humans on Mars, the local day-night cycle would likely be much more important than local seasonal cycles. Plus for a long time they’d be dependent on Earth (financially, politically, etc), and Earth years are a natural unit of Earth-centric planning, plus they’d see themselves as part of Earth-centric human history. And maybe eventually they’ll wean themselves off their dependence on Earth, but that would likely take centuries, by which time such a hybrid calendar may have become deeply embedded in Martian culture, and the use of the Earth year might endure long after the original motivations for it ceased to apply So I could see our current Earth-centric year numbering system being maintained as humans spread out through our solar system. And if eventually we spread to other star systems, we might take that with us too
- DonHopkins 1y agoIt would be interesting to have two sets of precessing holidays and weekends to enjoy!
- rollcat 1y agoI would be looking forward to the Landing Day celebrations every 780 days of 24.6 hours!
- rollcat 1y agoSourcing of my opinions mostly based on Wikipedia and random hear-say. The Antarctica already is an interesting case study. Different time zones are being used mostly due to territorial claims, but south of 80deg, UTC is dominant. Each base might be using their own, local equivalent of a day/night cycle. Connectivity is extremely limited; pastime Internet access makes 56k modems look like a luxury. Interesting reading: https://brr.fyi/ https://brr.fyi/ Near-term, we will most likely colonize the Moon long before Mars; shorter missions to lunar bases might happen in the upcoming decades. The Moon doesn't have a day/night cycle at all. It may have a similar effect on local, "human" timekeeping; people might talk with other local colonies as much as they do with mission control, and will rely on each other for immediate support. There are existing proposals for an LTC, which is both an engineering and a political challenge. One of the requirements stated in the current proposals already includes the handling of extended periods during which independence from UTC must be maintained - computers will run almost everything after all. Even access to the Internet will require a store-and-forward protocol, which will break the current webapp models. It won't affect how the years are being counted, but that's a lot of precedent even if just for technical and survivability reasons. Any kind of travel from Earth to Mars will have to be scheduled years in advance, and takes hundreds of days at least for unmanned missions. The colonies will need a significant degree of independence for basic survivability: food, oxygen, recycling, production of maintenance parts, etc. Some may specialize and rely on each other. Even information travels for ~3m to ~20m, and there's a blind spot when the two planets are on the opposites of the Sun. Staff involved in existing missions are already carrying customized timepieces to track both planets' solar cycles. Human missions will likely be one-way trips for a very long time; several new generations may pass. We will have to figure out how to deal with the lower gravity too, which may require new science. On the other hand, early colonies will likely remain subterranean (fun stuff like cosmic radiation), perhaps until we can figure out new materials. Big terraforming projects will require solving the unknown unknowns. I think it's too early to speculate, but eight millennia is a heck of a long time.