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

End Daylight Savings

petitions.whitehouse.gov

Post reply on HN
Print view View original

101–110 of 149 posts

Re: End Daylight Savings

#101

Post by tehwebguy »

Firehed wrote: post #56
ghshephard wrote: post #33

Earlier quoted context omitted.

Great for Unix Log files, bad for pretty much everything else. I work in emerging markets, and in the last two weeks I've had conference calls in Brazil, Singapore, Australia, Luxembourg, Portugal, and the UK. Whenever I'm scheduling a call with any of these people, I need to know what time it is. For whatever reason - every company we deal with seems to work from 8:00sh (sometimes 9:00) to 5:00sh (sometime 6:00) - L…

UTC would eliminate the confusion - just say "We're in the office from 17:00-02:00" (currently 9-6 PST) A nontrivial amount of my time is wasted responding to emails with "3pm pacific or mountain?". Asking "does 23:00 work for you?" avoids the variable. This is twice as bad if you ever have to work with anyone in Arizona during the summer, as they already skip DST. Fortunately calendaring software does a decent job a…

That makes a lot of sense, but it seems like there would still need to be some sort of qualitative way to consider the "effective time zone" of someone you're communicating with.

This may not be typical, but when I set up a call with someone on PT (I'm on ET) here's my current thought process: "OK they're on PT so whatever time that works for me is 3 hours earlier for them"

If we switched to UTC I would still have to think something like: "OK they probably start / end the working day 3 hours later than I do"

Re: End Daylight Savings

#102

Post by harshreality »

CJefferson wrote: post #86
zanny wrote: post #83

Earlier quoted context omitted.

There are 2 use cases for UTC: 1. People who don't leave their regional time zone very often. 2. Those that do. The first group of people would have a week where they aren't used to getting up at 1:00 or eating lunch at 23:00. But that won't last long, because the schedule maintains consistency and people would just shift their hours accordingly. The second group can now plan a flight, report the time they are landin…

I'm trying to decide if you are being sarcastic :) For people who don't leave their own timezone very often (which is most people), there is thousands of years of background of the concepts of midnight, midday. There is a shared language around the world as to what happens at certain times, which would be destroyed by this. If you want to organise meetings by UTC, why not just say "I'm available at 12-2 UTC"? See if…

Exactly. It won't catch on because most people will not know what UTC means, and will think you're crazy and eccentric. Most people do not even know their timezone offsets from UTC, and if they did they would consider the strain of applying that offset too taxing. They prefer that the meeting organizer state the meeting time in the attendees' local time. If it's not, and the meeting is important, they'll try to convert it, but they may get it wrong (especially with daylight saving time in the mix).

A great example of this is that most British think London time tracks UTC even during the summer. You talk to them about meeting times in UTC, and they'll be an hour off as long as London is observing DST.

Re: End Daylight Savings

#103

Post by riffraff »

zanny wrote: post #79
riffraff wrote: post #49

Earlier quoted context omitted.

> Morning or night isn't important actually morning or night encodes most of the things. For example, when visiting city X I won't check whether public transport is running from/to the airport if I know I'm arriving at 11:45, as I can safely assume it's there, while I will if I am arriving at 23:45 I'll have to find out whether public transport is running, or if there is taxi service I can pay with my currency, or if…

You are just moving complexity though, and transitioning to UTC globally eliminates broad classes of unneeded communication overhead. Right now, while you can assume at 11:30 AM public transportation is available, you have to look up what actual time zone wherever you are going is in and do math to figure out when that is for you. If everything were UTC, you would just look up the average operating hours of the day i…

> You are just moving complexity though

changing to UTC everywhere is moving complexity, I stand by the "keep the complexity where we have it already" :)

> Right now, while you can assume at 11:30 AM public transportation is available, you have to look up what actual time zone wherever you are going is in and do math to figure out when that is for you.

well, no. If I book a flight/bus/train it says that I get somewhere at 11:45 local time, they don't give me zulu time.

But assuming they'd tell me the hours in _my_ time, the complexity of doing a 2 digit sum is, IMVHO, not a major complexity over looking up what the time zone is (or equivalently, what the "beginning of the day" is locally).

It is possible that we are undergoing a shift towards more unified time. But for what is worth, I can tell you that the time shift in Italy over the last twenty years has been in the same direction as yours (e.g. TV prime time used to be 20:00, then 20:30, now it's 21:00), so maybe europe's being pulled from east asia being pulled from americas being pulled from europe ad infinitum.

Re: End Daylight Savings

#104

Post by anonymouz »

jzwinck wrote: post #99
anonymouz wrote: post #55

Earlier quoted context omitted.

They are not exactly the same, but at least roughly: If I make a work appointment at 9:00 or 10:00, I can be pretty sure that it will fall into usual office hours. If I go to a shop at 16:00 it will probably be open.

Again, not in Spain, where 16:00 is siesta time for many if not most shops and a closed period for restaurants. You seem to expect what you're accustomed to, all over the world. But it doesn't work that way, even for seemingly simple cases. And it isn't just Spain--for example, Hong Kong's electronic trading closes from 12:00 to 13:00. You'd have thought you could trade at mid-day on any exchange, but you can't!

That's why I said "probably". Obviously there are local differences and customs. Nobody claims you can magically predict exact opening times for where you happen to be at the moment, but you'll find that the day to day activities generally are aligned with the local solar day, and in fact, most probably the siesta has it's reason in the extreme heat during the afternoon in Spean.

Re: End Daylight Savings

#106

Post by daigoba66 »

I would also be happy if people in the US would adopt 24hr time in everyday life. It's a lot less ambiguous than 12hr. And less stupid. For example, in the US, 12:30am happens before 11:30am on the same day.

Re: End Daylight Savings

#107

Post by zanny »

riffraff wrote: post #103
zanny wrote: post #79

Earlier quoted context omitted.

You are just moving complexity though, and transitioning to UTC globally eliminates broad classes of unneeded communication overhead. Right now, while you can assume at 11:30 AM public transportation is available, you have to look up what actual time zone wherever you are going is in and do math to figure out when that is for you. If everything were UTC, you would just look up the average operating hours of the day i…

> You are just moving complexity though changing to UTC everywhere is moving complexity, I stand by the "keep the complexity where we have it already" :) > Right now, while you can assume at 11:30 AM public transportation is available, you have to look up what actual time zone wherever you are going is in and do math to figure out when that is for you. well, no. If I book a flight/bus/train it says that I get somewhe…

> changing to UTC everywhere is moving complexity, I stand by the "keep the complexity where we have it already" :)

Like I said, it eliminates the class of complexity that matters. Scheduling appointments and synchronizing people across time zones, something more common today, and will become more common forever into the future, and will become absurd once we have regular space travel, is a much larger cost set than when people move across what are currently time zones permanently, and need to adapt to getting up or doing things at different numerical times habitually.

> well, no. ..... so maybe europe's being pulled from east asia being pulled from americas being pulled from europe ad infinitum.

They won't tell you the hours in your time, and if they do, they are doing time zone conversion math already. Right off the bat, you have undue overhead in communicating time. The question is that people would be used to having the "day" be between a certain set of hours in one place, and then by migrating across what was a zone boundary what they would expect is now an hour off from their internal clock, because most people still behave in some synchrony with the sun.

1. You won't avoid someones internal clock being off by moving into an area of different daylight hours. To note, time zones only make the number match by longitude as well, latitudinally crossing the equator or going extreme distances north or south produces the same effect (different areas have different patterns of awakeness based on the availably of the sun) so that will already offset regular operating hours of various things, just moving north or south. Using UTC universally makes that concept ubiquitous as well, rather than having the disparity that moving east or west "changes' the time, but going north or south doesn't, but going anywhere societies change the operating schedules.

Like I said, I think it is much easier on anyone considering business to have a unified time standard, and accept that the hours of public services and resources will be different wherever they go, instead of converting time across zones for purposes of communication or meetings.

Re: End Daylight Savings

#108

Post by dalke »

T-hawk wrote: post #98
dalke wrote: post #59

Earlier quoted context omitted.

You already have to do a latitude adjustment, and that's much more complicated than longitude. I got quite disoriented once going to the southern hemisphere because my northern hemisphere-trained instincts though that the sun should be towards the south at noon, when it's actually to the north. In Sweden, I had to really adjust my calculations based on the season. If it's summer then the sun sets almost in the north.…

You don't have to do a latitude adjustment if you anchor your expected position of the sun to time rather than sunrise/sunset. The mean sun is always due east at 6 AM, due south at 12 noon, due west at 6 PM, due north at midnight (usually below the horizon except for arctic summer.) And you can interpolate between those, south by southwest is 3 PM and so on. Just shift the time points an hour for daylight savings if…

I'll grant you that I'm in a somewhat unusual circumstance. Today, sunrise was at 8.49, with azimuth 133 degrees, and sunset was at 15.49 at 227 degrees. It's easier to remember that it rises in the SE and sets in the SW this time of year than to estimate where under the horizon it would be at 6AM or 6PM.

In summer, at 22.00, it's also easier to think that there's another 30 minutes to sunset, which will be in the NNW than it is to figure out where it was 4 hours previous. The knowledge of where the sun will rise at 03.30 is helpful when setting up a tent, since it's easier to sleep when in the shade.

Though since I've found I'm rather bad at estimating angular width, I pull out a compass instead of eyeballing it.

Re: End Daylight Savings

#109

Post by eric_the_read »

T-hawk wrote: post #98
dalke wrote: post #59

Earlier quoted context omitted.

You already have to do a latitude adjustment, and that's much more complicated than longitude. I got quite disoriented once going to the southern hemisphere because my northern hemisphere-trained instincts though that the sun should be towards the south at noon, when it's actually to the north. In Sweden, I had to really adjust my calculations based on the season. If it's summer then the sun sets almost in the north.…

You don't have to do a latitude adjustment if you anchor your expected position of the sun to time rather than sunrise/sunset. The mean sun is always due east at 6 AM, due south at 12 noon, due west at 6 PM, due north at midnight (usually below the horizon except for arctic summer.) And you can interpolate between those, south by southwest is 3 PM and so on. Just shift the time points an hour for daylight savings if…

That was exactly the problem that started time zones in the first place! Every town had their own definition of noon, which made it impossible to schedule trains across long distances. Time Lord, about Sanford Fleming is a great book about how time zones came to be: http://www.amazon.com/Time-Lord-Sandford-Creation-Standard/d...

Re: End Daylight Savings

#110

Post by lazerwalker »

zanny wrote: post #78
anonymouz wrote: post #48

Earlier quoted context omitted.

But when scheduling the meeting for time x you'll have to look at the time offsets of all the participants to figure out if x falls into their work day or into the middle of the night. If you schedule a meeting at 13:00 UTC, is the Chinese guy still in his office? Is the one from the US already in his office? Instead of knowing their timezones and their offset you'll now have to know their office hours in UTC, which…

You still have to ask for times they are available. That never changes. With UTC as the global time standard, you don't have to try to figure out time offsets. You just say "I'm available from 8 to 12" and they say "I'm available 5 to 9" and you make the call at 8. The alternative is "I'm available 3 - 7 London time", the other guy is available 12 - 4 Pacific Time, and you have to get out a calculator to figure out w…

For prescheduled meetings, you're absolutely right. "I'm available 8-12, and you're available 5-9, so let's chat at 8" is a clear improvement.

But what if it's not a scheduled meeting? If it's the middle of the day where I am, and I need to give you a quick call to verify something, I need to figure out if it's appropriate to call and if you're likely to be in the office.

Currently, that means converting my time to your time, as easy as looking up the offsets and doing a simple sum. Context clues make it easy to figure things out: if it's 10:30am your time, there's a high chance that you'll be in the office. If it's 3:30am your time, you're probably asleep. Sure, there are still cultural fudge factors at play (do you come into the office late, do you take a siesta, etc), but a ballpark estimate isn't hard. Assuming it's not an emergency, I don't care if I get your answering machine if you're in a meeting; I do care if you're offended that I woke you up.

If we're both operating in UTC time, this conversion now requires cultural knowledge of what your working hours are before I can even get a big-picture idea of whether it's appropriate to call.

Post reply on HN