Are we springing into a Y2K-class nightmare? The US Senate has passed legislation aimed at making Daylight Saving Time permanent, leaving the country in the "spring forward" state from 2023. The practice of changing the clocks twice a year in the US dates back to the agrarian heyday of the early 1900s and has endured, despite regular grumbling. Clocks go forward an hour …
Yep. The world of computing is plagued by shoddy, incompetent and lazy programmers.
I think I first saw this in action in person when the government changed the VAT rate and lots of really shitly designed software promptly crapped itself because it was hard coded. Followed by far too many shit updates to the same software which did nothing more than change the hard coding rather than actually implement it as a time sensitive value.
This was a different failure to Y2K which was about data encoding rather than solely about stupid short-sighted assumptions.
It keeps us in work though...
The very first database I wrote in the early 1980s hard-coded VAT. Because, well "vat" *means* 15% doesn't it? What do you mean it's changable, it's a constant, like pi or e, how can you redefine what "multiply by 1.15" means?
But do you remember the chaos in the 1970s when we had two VAT rates (in addition to the zero rate)?
The standard rate was 8% along with a "luxury" rate of, initially, 25%, which was later reduced to 12.5%. I was in the electronics industry at the time and the two rates meant that a component destined for a luxury item carried the luxury rate but if that SAME component was destined for a non-luxury item it carried the standard rate. Confusing or what?
Definitely a case of those imposing this madness not thinking things through properly - but when do they ever - sigh...
"But do you remember the chaos in the 1970s when we had two VAT rates (in addition to the zero rate)?"
No I don't, I wasn't conciously there, which was the problem. I thought "vat" was just another mathematical constant like "pi" or "e" or "1/SQR(2)" or the golden ratio. There was nothing in the knowledge culture around me to say otherwise. "vat" *meant" 1.15 just as 1/SQR(2) *is* 0.707.... If 1/SQR(2) was variable, the A-Series paper sizes wouldn't work, surely vat is also just as non-variable, it's a mathematical function. In fact it was probably used as an example in maths classes about percentages as "15%", not as "a variable amount which this year just happens to be 15%".
Of course, now it's more knowable, with 5% and 20% rates and regular news reports about arguments about changing it.
I've just checked Wiki, and just before VAT came in Purchase Tax had *four* rates, all the way up to 55%! And then during the '70s VAT bounced all over the place, changing almost yearly between 8% and 25%.
I'm a lazy programmer! That means I don't want to go back and sort out messes caught by short cuts in the first implementation so stuff like this is never hard-coded.
Though I'm happy to admit to being lazy (it is one of the three virtues after all) I do my best not to be shoddy and, well, you'd have to ask my peers about competency.
"Hopefully permanent DST is not coded in December by people who can't wake up properly without a bit of morning sunlight."
If it was up to me, we should (in higher latitudes where winter days get noticeably short) have shorter working days during the winter, starting later and finishing earlier, so that there is at least some chance that we aren't travelling to or from work entirely in the dark (I'm trying not to forget about unsociable hours shift workers, who get a pretty tough deal of it anyway, and I can only hope that they do get paid at premium rates to try to compensate for that).
I'm sure I'm not the only person who operates at (at best) 75% functionality between about November and February, we might as well formalise that!
Keep going north and you'll realise the big problem with this suggestion.
In Luleå, up in the north of Sweden, there are three hours of daylight per day in December and 23 in June. Describing the light in December as "daylight" is probably stretching the definition.
When choosing whether to go to permanent GMT or permanent BST you have to remember that Edinburgh, on the east coast, is west of Bristol on the west coast.
When I were a lad we did have permanent BST (British Standard time they called it) for a few years. The Scots, in particular, hated it because they are so far west that it puts them almost two time zones away from "local" time.
On the other hand, over the last couple of decades, there have been regular updates to Windows, Linux, macOS, iOS, Android etc. as other parts of the world have changed their daylight savings time start and end dates or abolished them completely.
Most systems these days work off of UTC +/- Offset and they have a database of time changes for each country. It just needs that list to be updated. But, that assumes all computers are still in their support windows and will get the updates.
Interestingly, Hawaii and Arizona already don't observe DST, so the systems must already be able to cope with it, mustn't that?
Parts of Indiana stay on Eastern time all year. Other parts stay on Central time. Still other parts have daylight savings. And the time line between Central and Mountain time runs right down the middle of Main Street in Pierre, North Dakota. And all of Alaska and all of Hawaii are one time zone, despite being rather large. There’s going to be lots of programming to be done, and I guarantee that mistakes will be made… and not detected for a while.
Yeah, there's nothing new here. Countries all over the world have been messing with their DST rules non-stop since - well, since long before Y2K - and people have kept up somehow. Maybe there are a few problems, maybe some people miss meetings or automated emails get sent on the wrong days, whatever, but the world hasn't ended yet.
Having worked in international car hire and seeing 86400 liberally splattered throughout multiple code bases, I whole heartedly concur. We used to dread clock changes.. customers paying 3 days for 2 or vice-versa was not uncommon.
Wouldn't it be fun if, in a few years, someone has the great idea to advance clocks by an hour in the spring "to take advantage of all that sunshine" and then revert them by an hour in the autumn?
And then, a few years after that, someone has the great idea to advance clocks by an hour in the spring...
Within a couple of generations, we might end up getting up when it's dark and going to bed in the daylight.
I think we should change the clocks every day during winter, they should run at 1/2 speed at night and at double speed in the daytime, then we could all (apart from the night-workers but nobody cares about them in daylight saving discussions!) get up in daylight and have long sunny evenings even in the winter! Works for me. Sunshine icon.
When I was a child the UK experimented with Double British Summer Time, i.e. advance the clocks 2 hours ahead of GMT / UT in the summer. I remember being sent to bed and sometimes waking up at 10 in the evening when it was still full daylight outside. Very disconcerting.
>They occasionally talk about having a single European (including UK) timezone
It was tried but there was significant anti-europe feeling in the UK at the time and so the attempt to extend it to the UK was abandoned. in favour of trying to extend it to the east.
The problem with the change, for me, here in Northern Europe, is that it changes from bright mornings back to darkness, just as I've gotten used to walking with the dog at dawn and the sun has just gone over the horizon and stops blinding me, when I drive to work, so I have to go through the whole process again. The reverse in the autumn.
https://www.heart.org/en/news/2018/10/26/can-daylight-saving-time-hurt-the-heart-prepare-now-for-spring
But of course Americans "prefer" to be different just for the sake of being obstinate. They're the "greatest" country in the world, so they couldn't POSSIBLY consider adopting standards that someone else created; that would be admitting they couldn't do the job themselves. Can't have that...
It was France who gave them that loan to back their colonial scrip with something fungible.
It was France who sent them General Lafayette.
It was France who sold them what they now call the Midwest, Cajuns included.
It was France who gifted them their Statue of Liberty.
And it was France who gave them their Freedom Fries.
You might be on to something here.
Just to set the record straight ...
... so-called "Freedom Fries" was the brain-child of disgraced Republican congressman and convicted felon Bob Ney (served 17 months of a 30 months sentence at Club Fed (Federal Correctional Institution, Morgantown)).
Nobody in the US paid all that much attention to the supposed name change, other than the Press and the usual rabid Republican sycophants, who are a very small, if vocal, subset of the population as a whole. I can quite honestly say that I never saw the item on a restaurant menu, even when it was supposedly an in thing.
A rather impressively large number of naval military ships were reconditioned spoils of war. After they were checked for improvements worthy of copying, captured ships were typically just patched up and given a new flag. I understand that some ships were lucky(?) enough to change nations quite a large number of times!
There are/were a significant number of ships named Temeraire in the Royal Navy; the famous one was actually the second of the name in the RN. There were also a few Redoubtables, Belleisles, and more; famously there were three Neptunes at Trafalgar, one each British, French, and Spanish. (One of the French, soon to be British, Redoubtables was at Trafalgar; she was one of two French ships of the line closely engaging Victory when the famous Temeraire piled in, and with one double-shotted broadside inflicted over 200 casualties, or about half of the crew, on Redoubtable, including much of the boarding party about to board Victory. Temeraire sent a boarding party of her own, took Redoubtable, crossed Victory, and assaulted the other French liner.) The French had a Swiffsure or two, over time, as well.
The USN has had ships named Canberra, de Grasse, and Churchill, though for different reasons.
Sounds like another case of "If it's good for America, it's good for the goddam world!"
Case in point: liquid measurement. Not only do they reject the logical Metric system by retaining gallons, etc - but their "gallons" are a bit short of the real thing, leading to continuing confusion.
Sorry to tell you that the Imperial gallon and the US liquid gallon are two completely different measurements, with the former being about 25% larger than he latter[0]. Perhaps you grew up close to the Canadian border? Or maybe yer DearOldMum found a sale on mis-directed gas cans at the local hardware store?
[0] Note to the inevitable pedant(s): I said "about" for a reason. Find something else to bitch about. Ta.
Sorry to tell you that the Imperial gallon and the US liquid gallon are two completely different measurements,
Indeed , so your atrocious vehicle mpg figures are not quite as horrific as it seems at first glance.
yet you still call your antiquated system of inches , yards , pounds and whatnot as "imperial measurements"
Why the hell is the gallon different? isnt it a complicated enough system as it is without the surprise of "oh , we changed the gallon , its now different to everyone elses!"
any others? is a fathom still a fathom?
… of inches, yards, pounds and whatnot as “Imperial measurements”
No, we call our system of measurements “US customary”. The Imperial system of measurements was a replacement for the traditional English and Scottish units in the UK and its possessions; the Imperial system was not adopted in the US.
Why the hell is the gallon different?
The US customary gallon (3.785411784 l = 231 cubic inches) was originally the pre-Imperial English wine gallon. (There were also ale gallons, corn gallons, and coal gallons in the pre-Imperial English system.)
The Imperial gallon (4.54609 l) is closer in size to the pre-Imperial English ale gallon (4.621152048 l = 282 cubic inches). The Imperial gallon’s size was not based on a set number of cubic inches, but (originally) on the volume of 10 pounds of distilled water weighed in air with brass weights at a barometric pressure of 30 inches of mercury and at a temperature of 62 °F.
any others?
Of course! Both gallons contain eight pints, but for liquids, the US customary pint contains 16 US customary fluid ounces, and the Imperial pint contains 20 Imperial fluid ounces; thus, the Imperial fluid ounce is slightly smaller than the US customary fluid ounce.
There’s another unit called the bushel, which is a larger unit of volume. The Imperial bushel contains eight Imperial gallons. There are two US customary bushels: one for liquids (eight US customary gallons), and one for dry goods such as grains (approximately eight English corn gallons).
The US still uses both avoirdupois pounds and Troy pounds for measuring mass; Troy weight is used mostly for precious metals. The Imperial system has only retained Troy ounces (but they are identical in both systems).
The US used to define certain SI units in terms of US customary units. This changed in 1959, when the US customary units were redefined in terms of SI units. However, the “survey foot” is still based on the pre-1959 US definition of the meter in terms of the yard. This wasn’t redefined because of the amount of survey data that was based on the survey foot.
And it's still in common use as a term in the UK. Just not an official one for trading in. As noted, we buy fuel priced in litres but measure its efficiency of consumption by our cars in miles per gallon.
We also buy our bags of flour or sugar and think of them as 2lb bags when they're sold as 1kg (2.2lbs) bags.
It sort of works.
Though measuring stuff in American style "cups" does my head in. WTF does "2 cups of.." mean? Which cups?
If you only have cups in a recipe it's easy, just replace them with "1 volume measure of choice", problems turn up though when you have compactable items like flour or brown sugar. Do you sieve the flour first to start at a known state? Do you get the spoon out to press it hard into the measure?
As for measuring butter in teaspoons, that's just stupid if you want accurate measures in any way.
… that’s just stupid if you want accurate measures in any way.
Over here, butter is generally sold in one-pound boxes, approximately 4¾″ × 2½″ × 2½″ (12 cm × 6.4 cm × 6.4 cm). Each box contains four individually-wrapped four-ounce “sticks” of approximately 4¾″ × 1¼″ × 1¼″ (12 cm × 3.2 cm × 3.2 cm) each. Each wrapper has tablespoon and teaspoon markings on it; each stick has eight tablespoons or 24 teaspoons. if a particular number of teaspoons or tablespoons of butter is needed, all it takes is lining up a knife with the appropriate mark and cutting the desired amount of butter from the stick.
The metric system has some uses butter is sold in blocks of 250 grams with markers on the wrapper every 50 grams. A Pound is 500 gram, 2 pounds a kilo(gram)
recipes are mostly in grams or milliliters just use a kitchen scale or measuring cup but we do have set of
slightly increasing measuring spoons for older recepies.
WTF does “2 cups of…” mean? Which cups?
In American recipes, a cup is a unit of volume equal to eight fluid ounces. Over here, this cup is ½ pint; in Imperial land, it would be ⅖ pint, so it’s easier to think of it there as equal to eight fluid ounces (the difference between your fluid ounce and our fluid ounce is only about 4%, which should be a negligible difference for most recipes).
The 'Imperial' system was defined after the US measures were standardised.
Both the Imperial, and US system, were based on an earlier English system.
There were multiple Gallons in the earlier English system, depending on what you were measuring.
The US decided, quite rightly, that having multiple gallons was daft, so when they implemented their standardised measures after independence, they adopted the Wine Gallon as the single Gallon measure.
A few years later, when the UK also decided to standardise, and so created the then new Imperial standard, they adopted the Beer Gallon instead, which is a bit bigger than the Wine version. (This is also why things like Pints are different, as they are derived from the gallon).
If you had something in the US labelled as "Imperial Gallons", it would likely to have been an import (such as from Canada), as Imperial units have never been used in the USA (not officially anyway).
How about also adopting the same paper sizes as the rest of the world ?
Never mind paper sizes, how about pretty much every other unit? While the rest of the world trades its wheat in tons, a unit of weight, the US trades in bushels, a unit of volume. US recipes talk of cups, everyone else uses weight for the majority of dry ingredients. They've even got their own version of the meter, which is subtly different from the SI one.
In Canada, butter is often measured in "cups" or tablespoons, which while being a volumetric measure is singularly unhelpful, unless you happen to be melting it.
The choice is pretty arbitrary because a lot of recipes here have measurements in cups which is handy if all you have a a cup measure. Butter packaging has measurement indications on it in cups and tablespoons and you can even get a separate "butter measure" rule.
Had a casual discussion this week with the editor of a magazine about the length of a famous American landmark. Different sources attribute different lengths to it, but they all used miles as the measuring unit. The trouble with that, arguably, is you need to specify which kind of mile you are using (English Statute mile, Nautical mile, Irish Country mile, etc.) whereas the kilometre suffers no such ambiguity, therefore suggesting quoting that for accuracy purposes.
We inherited all that crap from the British. Damn them!
But then you don’t use it all. One of the points of the imperial system is that there are always nice numbers to work with and if they start getting too large or small you move up or down to the next unit. So for weight you have ounces, pounds, stone, hundred weight and tons.
So why when you ask the weight of the Saturn V rocket, do you always get an answer in pounds?!
This was always one of the strange points about UK's decimalisation.
Moving to 100p=£1 was always sensible.
Determinedly killing off the intermediate units (shillings- 5p) was just annoying and pointless. Especially at the time when say 5 shillings (25p) was a useful sum of money ( about £4 current values). If you can change the subdivision of the £1 to a hundred (new) pence, you can equally retain the shilling and change its subdivision to 5 new pence. In the great scheme of things it doesn't make any difference ( though the older folk might have found it easier, perhaps) but the determination to do something so pointless seems, even decades later, to be more about some kind of bureaucratic dogmatism.
"Determinedly killing off the intermediate units"
You lot finally managed that, eh? That's sad, in the old meaning of the word.
I was in the UK just prior to, on, and immediately after Decimal Day. From what I recall, a pint was still twd a half bob on Feb 15th 1971 ... By 1980ish, it was 50p, but my favorite beertender was still asking ten bob ... his wife asked for "ten shillings please, luv". Likewise in all the local shops near where I lived.
And we all bitched about it ... imagine, half a quid for a fuckin' pint! What was the world comin' to?
I was in school then. The old "shillings" there and out in the world were, pretty much over night, officially declared non-existent and could only be used by those at the entrance to God's waiting room. Of course, some of us persisted, even though in our early teens, to use the term ( and I welcomed decimalisation and later metrication- not being good at rote learning). But that was pure mischievousness.
"Determinedly killing off the intermediate units (shillings- 5p) was just annoying and pointless."
But why (seriously)? That was the whole, well half, the point of the process: 1 unit, with 100 sub-units in it (a sensible round number), and no further complexity.
Shillings were just an unnecessary extra 'layer' between pounds and pennies that, by the 1970s, you didn't need any more. I get that even then £1 was worth substantially more than it is now (10p Sparkle ice lollies were the pinnacle of my childhood luxuries), but it's not like it was the first half of the 20th century any more when, for most people, something costing more than a few shillings would have been really quite expensive, and a few pounds a whole week's wages (and so pounds would rarely have been in use for everyday shopping, an occasional new suit perhaps, or perhaps a good pair of shoes).
But I did think it was quite cute how (slightly funny looking, and often quite old) 1 shilling and 2 shilling coins remained in circulation (and valid) for many years afterwards, although they only ever meant 5p and 10p to me. It was a little sad when they were finally withdrawn when the new, smaller, 5p and 10p coins were introduced, a little bit of living history finally passing away.
That's an illusion. "25p" takes (very, very slightly) longer to say than "five bob", but it's easier to comprehend and less subject to error.
Imagine you're in a noisy environment. The "five" will probably get across, because the listener will be on the lookout for numbers, but the "bob" could take them by surprise. They might be expecting "shillings", "florins", "sovs", "pounds", "guineas". Or you could have said "a crown", for even more brevity and confusion.
"25p" removes a whole layer of potential ambiguity and misunderstanding.
>So why when you ask the weight of the Saturn V rocket, do you always get an answer in pounds?!
We have 1st stage separation at a vehicle mass of 100 tonnes, 4 hundredweight, 6 stone 3 lb and an altitude of 100 nautical miles, 4 furlongs, 6 chains, 3 yards, 2feet
Simple really
"They've [the USA] even got their own version of the meter, which is subtly different from the SI one."
Are you sure it's not perhaps the US survey foot that you are thinking of?
That is the previous US foot, because the current definition of the international foot (which is defined in metric terms) is a compromise value between the then US and UK foots, which were very slightly different from each other. If measuring distances in feet, over long distances, the difference between the old US foot and the international foot would start to become quite noticeable. Of course, if they switched to the metric system, they wouldn't have that problem (although obviously a lot of land ownership maps would need to be updated).
Citation needed. I'm an engineer in the US and my international employer has used SI since the 1980s at least. The only difference I know of between the meter we use in the US and the metre used by my UK colleagues is the order of the "r" and "e" at the end.
Oddly, for a company that makes some very large things (tens of meters long) our standard is to use centimeters on all drawings.
It ( and other anomalies) are, I assume, because Americans tend to start with speech from way back and work forward to modern usage. It's a kind of conservatism that seems to be part of American culture.
So, they say " March 16th" and write 3/16 then add the year as an afterthought 3/16/2022.
In a similar vein, I assume, they had a paper size used for letters. So they've stuck with it. And having measure stuff in cups full in the farmhouse kitchen they've copied the recipes into the recipe books.
Being fair, we do some similar things in the UK. We buy fuel in litres, but measure the distance it takes us in miles per gallon . We routinely measure sizes in cm, except for our own, which are in feet and inches. Most people still weigh themselves in Lbs and stones even though they weigh the food they eat in Kg. And for the most part we're happy to do it like that- there was surprisingly little resistance to changing the statutory weights and measures.
Yeah, I used to think this. But then I discovered that, as recently as the 1970s, "March 16th" was the most common way of saying it in British English too. The whole "16 March" convention didn't really take hold until after computers were embedded in life.
It would be OK if the shoddy US developers noticed that the rest of the world used a different size.
It was only a couple of years ago that Microsoft word stopped fucking up the margin sizes in documents by insisting on storing them in antiquated obsolete measure units and doing a conversion time. It was annoying AF setting a margin or spacing to something like 0.15cm only to find it changing to 0.14 or 0.16 when checking back on it. Something like that anyway...
But talk about 'noticing the rest of the world' !!!!
If Richard was not a London-Based London Journalist covering a foreign country, he probably would have noticed that the USA had it's daylight savings Y2K event several years ago, when, for the first time since introduction, the set-in-stone start and end date were shifted.
What they are looking at now is a Y2010 event. "Same as last time, only less important".
You wouldn't be saying that if you tried to print something and got an annoying error message, or the formatting went wrong. Which for many years was a constant problem. At one point, when I was supporting the introduction of computers in schools, this was the most common problem I had to help staff with.
Wait. Let me get this straight ... you are blaming the Yanks for ThoseInCharge of procurement and installation at British schools (and other places) being unable to read and understand the specs for the equipment and software they are ordering?
Wow. Things must be worse over there than I thought.
I'd be happy if the incompetent US coders actually understood that most of the world is NOT North American and has a different date format and uses standard measurement units and not antiquated Imperial measurements that are currently only used by three backwards regimes in the world.
Showing information in different formats suitable for the target audience is not difficult and not doing smacks of gross incompetence and laziness. Sometimes it's easy to spot that, for example, we are not currently the 3rd day of the 16th month of the year, however other times it's annoying to have to do the mental gymnastics to have to guess whether or not I'm looking at the real date or an incompetent North American developer's output of the date. It was only thirty fucking years ago that it was agreed that if one must display dates in a localised form, it's best to use day number and month name because that's at least translatable compared to two arbitrary numbers.
Thirty years later, I sign into Microsoft 365 admin centre and all the formatting is in American and not English format. As a result, it's very easy to mis-read the mangled dates as something else. Apparently this bullshit is "progress".
/grumpy... just don't get me started on the fact that 12:00am and 12:00pm categorically DO NOT EXIST. The solution to this was created over 100 fucking years ago.
This is to avoid confusion over which midnight you're talking about, the one at the start of the day (00:00) or the one which really belongs to the next day (24:00).
It probably would have helped the USAian who explained to me why I was wrong for referring to midnight on Monday as midnight on Monday. I should have called it midnight on Sunday. And this was coming from a programmer who should really know there's only one midnight per day and it's at the start.
At that point given this explanation and the US software which I have the misfortune to use (calendars beginning on a Sunday, formats like dd-mmm-yy which just aren't used in my country outside of US software, even Adobe Sign which uses mm/dd/yy on legal documents of all things) I'm afraid to say I decided there and then all USAians are genetically predisposed to mis-understand times, dates, timezones, and anything else about localisation, which is odd because they have enough of their own timezones.
Midnight and noon aren't actual times, they are just markers between the old Roman notion of ante meridiem (before midday) and post meridiem (after midday). They both have zero duration, and as such are logical constructs, not actual times.
Thus "midnight" marks the time when the prior day stops the new day starts. As it is time of zero length, it doesn't actually belong in either day.
Put another way, there is no "midnight on Monday", but there is a "midnight between Sunday and Monday" and a "midnight between Monday and Tuesday".
It follows that the time 24:00:00 doesn't actually exist, and is an illogical construct.
You're confusing the beginning of the minute with its length. The day is divided into 1440 minutes, so it follows the minute beginning at 00:00:00.000000 is exactly the same length as any other minute in the day.
24:00 is not ideal, but a convention to try and remove ambiguity. Neither better nor worse than Saturday/Sunday or Sunday/Monday, but shorter, and perfectly understood in those countries which use a 24-hour clock more than English-speaking countries.
> USAians are genetically predisposed to mis-understand times, dates, timezones, and anything else about localisation, which is odd because they have enough of their own timezones.
Timezones are a little confusing, but we can figure them out. It's the concept that there are other countries (Canada doesn't really count) that throws us for a loop.
Lock the clock, yes absolutely. But lock it to match the sun FFS. Don't think you're ordering me to get up early.
Since working at home I've realised that I don't need to play along with that 'spring forward, fall back' nonsense. The clock goes forward, the alarm goes later. Cancels out.
I only wish I'd thought of it all those years I was on flexitime.
I'm starting to think that the idiot senator, deep down, doesn't want his own bill to pass. Free publicity and in the end no real harm done.
DST is an abomination. As the sun approaches the apex, it is noon. Some deviation is expected where the time zones meet, but more than an hour? A line must be drawn.
As for software -- we do strange stuff with time all the, ahem, time. Yes, some of it will break, but most will be because of a lack of updates. Better find out this way than await pwnage by some random Russian hacker.
>DST is an abomination. As the sun approaches the apex, it is noon.
We could do away with timezones altogether.
We don't have watches anymore, we look at our phones for the time. Phones have GPS and so know their exact longitude. Obviously we should all have our own local solar time.
The only reason for standard time was railway timetables. The USA has almost abolished trains and has certainly freed itself from the tyranny of the rigid train timetable. Why not abandon this big-government standard time ? Freedum !
That's OK if you don't care about time scheduling. About 25 years ago I was writing scheduling software that had to take into account daylight savings time changes and accurately account for the hours in the schedule. It did this through asking the Operating System the real time (UTC) for the given time and using this to work out the duration of scheduled events. I suspect that it would probably fuck up royally with different time zones but it worked very well for what it was needed for (and no competitor systems could cope at all, many required tedious manual adjustments twice a year).
Seems like you are struggling a bit with the concept of time. The same 'fix' you used for DST would be the same you would use for time zones. UTC, Universal Time Coordinated (or Coordinated Universal Time, if you have trouble with acronyms), was designed just for this.
I can't think of any system that I have designed in the last 30+ years that will care if DST is abandoned. Not only can you ask the OS for the UTC time, it also gives it to you in local time for you to display to your users. So the OS gets to know all the rules for time zones and DST. (and the states that don't use DST)
I also go one step further, I let my database be my reference clock for my applications. Quite often my servers are not in the same time zone as the users.
You've misunderstood my exasperation about the state of most scheduling systems.
For example where an event starts at 22:00 one day and finishes at 04:00 the following day, the duration is normally considered to be 6 hours, and therefore billable as such. However when an event such as this occurs over the clock change over period the actual duration can be 5 hours (spring) or 7 hours (autumn). Calculating the correct time duration is one thing given the start and end date/times, setting the correct end point as a result of a desired number of hours (for example 6 hours of care) is another.
Luckily nothing I ever had to include in the system crossed time zones, just daylight saving time transitions. As long as event started and finished within the same time zone it would have worked. Where it would (probably) have failed is where the time zone changed between the start and finish times.
If you use a monotonic timescale, eg UTC, then accurate time periods fall out naturally on whatever date. Also logfiles should do the same. You can always get the time of day back after the fact from the TZ rules if that happens to be a factor in the log events. Now UTC does jump by a second occasionally, but so far it has never jumped backwards, so it's still monotonic even though a period measured over a leap second event will be out by a second. In practice who cares, except astronomers or spacecraft navigators.
Ah, yes, logfiles. There are morons who do logging in local time. So any time I have new data flows I have to wrangle that information out of some id-10-T person, and since we are not allowed to have real in person meetings I cannot use my LART on their lard-arses...
I'll have a beer now. Or something stronger.
It's about presenting the information in a format that is usable by the person with a schedule.
For example, overnight work starting at 22:00 and finishing the following day at 08:00. It wouldn't help them if the system was to output that the finish time was 07:00 GMT and they'd have to work out that this is really referring 08:00 BST. Billing always had to calculate the real duration regardless of event straddling over a clock change, and legal considerations had to be had when the duration of work was extended by one hour.
Simple stuff like
- working in the U.K.
- with colleagues on EST
- with customers across the globe means dealing with multiple time zones and DST changes. New York/Montreal, London, Milan, Mumbai and Manila just internal. Customers in pretty much every time zone.
Currently in the 2 week Vernal Equinox-ish DST hell of North America has shifted and EMEA is 2 weeks behind so my diary is like shit right now with temporarily overlapping meetings and NA colleagues dumbstruck wondering why some of their meetings have moved out an hour.
An old iPad 2 doing sterling service as a World Clock ‘what’s the time elsewhere’.. or the excellent timeanddate.com World Meeting Planner.
I think the North of England would object as well - last time this was trialled the RTA rate went through the roof, as did pedestrian casualties.
Now that all the streetlamps are on part-time lighting outside cities I suspect the same would happen again - probably with higher pedestrian casualties due to distractions like phones/earphoes etc.
North of Inverness/South of Watford people probably don't much care as it really doesn't make a lot of difference - one is mainly dark in winter anyway (lived up there myself) & the other is mainly illuminated on even minor roads. Its the bit in the middle where most people live that's contentious.
Also worth pointing out that the whole of the UK is north of any USA state bar Alaska so it actually does make sense to continue BST here.
I remember a bug like that where the software played up on the first day of the month, from midnight local time to midnight UTC. Only customers in Australia noticed when they started work 8am which was before midnight UTC.
QA had no chance to find this. Half a year it just worked in the UK, half the year if failed one day a month from midnight to 1am when no testing was done.
Programmers in countries that don't have a daylight savings system tend to forget that such things exist. At least I think that's why my Huawei watch sleep time tracker always crashes on the changeover dates. I wonder how long it will be until code written in America starts to have the same problem.
Well, Marco RubiNO conveniently forgets that back in 1974, during the Oil Embargo, the US tried year-round DST. Lasted 2 years before that naff idea got deep-sixed. If he really wanted to make a useful suggestion (Oops, I forgot who it was, sorry!) he'd suggest year-round Standard Time. I was just getting to leave for work in some semblance of light; now I'm back leaving in the dark - and I was VERY LUCKY to see the genius in a dark coat and slacks running across the middle of a 4-lane road before I hit her. I'd suggest that we shift the US time zones to start with Atlantic (GMT-4) so we get daylight more of the morning hours. Hawaii isn't connected physically to any other part of the US, so they can do as they please.
The argument that it will be [too] dark in the morning [for] kids if we make #DaylightSavingTime permanent ignores the fact that we are already on it for 36 of the 52 weeks a yearIt’s time to #LockTheClock
— Marco Rubio (@marcorubio) March 16, 2022
During the 1973–4 oil crisis, the US went on year-long daylight saving time for a two-year period, but ended it early, in October 1974. Perhaps Senator Rubio might consider how winter daylight saving time affected the accidental death rate of school-aged children in Florida then (NB: the linked articles may have uncorrected OCR errors).
If school starts at a later time, that reduces the time available for afternoon/evening activities, cutting into the profit for all the businesses that aim to benefit from full-time DST. I'm surprised Rubio would go for that.
Up here above the Compromise Line [1], those winter mornings [2] are crazy dark enough already. I dread the "spring forward" not for the lost sleep but for the darkness (my own daily walk, raising the kids, etc.). Yes, evening car-on-car crashes are to be avoided, but there's a high school a mile away (plus 3 other schools nearby) and the combination of vehicles + pedestrians + dark has already been dangerous, even deadly; why make it worse? (--> icon just for this)
I'd rather go on full-time STANDARD time. If Rubio's bill becomes law, I'm petitioning the state bods in Lansing to put Michigan on US Central Time instead of Eastern. [3]
1. The Kentucky-Tennessee border, NOT the true Mason-Dixon line between Maryland and Pennsylvania.
2. Including those extra weeks ever since 2007 [3], some of which are actually "spring".
3. Ideally, with 15-degree time zones centered at 0, 15, 30, etc., then Michigan already falls too far west of where "US Eastern Time" should be.
4. Speaking of history, full-time DST shouldn't be much more difficult than when the dates changed back then, right?
>Is time considered a Federal matter or not ??
It was upto individual towns, so especially in the south either for obvious reasons that they are in a sunny desert near the tropics, or because it isn't in the bible.
There was a recent law saying that any further changes had to be agreed by the feds, tied to bridge maintenance funding or similar chicanery.
The government probably cannot mandate this change, but it can force all federal government bodies to use it and potentially anybody getting money from the feds.
The argument that it will be [too] dark in the morning [for] kids if we make #DaylightSavingTime permanent ignores the fact that we are already on it for 36 of the 52 weeks a year
Yes, you are on it for the 36 summer weeks, but the issues occur in the 16 winter weeks. How do people without a brain ever get elected?
Jc
Well, he's from Florida and Sinema's from Arizona, so they don't consider issues for the northern half of the country. Phoenix can have up to an hour or so more daylight in winter than Minneapolis, and Miami 30-45min more than that. In the vast swath where there's less than 9hr of daylight, one of the day's work commutes is going to be in darkness. School is around 6-hr duration so opening and closing times might be adjusted, but there are other pre- and post-school activities, and businesses pegged to the instructional day so I don't think parents will find it any less hassle.
"How do people without a brain ever get elected?"
Well... if you took the idiots out of Congress, it would no longer be a representative body.
(Some years back, when Richard Nixon nominated a particularly hard-to-defend choice to the Supreme Court, one senator was reduced to arguing : "Okay, this guy is mediocre, but so are many Americans, and they deserve representation, too, don't they?" Somehow, nobody replied that mediocre Americans had ample representation via the Senate already.)
Perhaps it might be a good idea if Outlook and associated tools actually looked at the DST rules in calendar invites, rather than the TZ description, to work out what time an appointment should be. Lost count of the number of helpdesk tickets we got from that particular doozy.
Ah yes "focused inbox" or the "randomly hide important emails for no even remotely logical or discernable reason" option. I wonder if it's possible to disable this option (Group Policy) however the retarded thing is set by default therefore I suspect that Microsoft's marketing droids will have blocked any attempt at sense. Again.
They were planning recently (last year or the year before?) to stay on summer time. A lot of people support this (I'm one of them). But for <mumble mumble bullshit mumble> it got kicked into the long grass and, well, maybe we in the EU will stop buggering around with the clock twice a year after pretty much everybody else has given up on it.
The first is TheWife's monthly cycle. If you are married to a woman, you'll grok.
The second is the seasonal clock handily provided by the Solar Year & the Earth's axial tilt with respect to its orbit. It is totally out of my control, but I plant my fields & breed my critters by it, as humans have since time immemorial. Trying to change this is a fool's errand.
The third is the clock provided by the Master clock on my network, which syncs up to an atomic clock once per day (ntp.org works for most purposes ... I use something else), which all of my machines adhere to. This is for computer record keeping more than anything else.
Context is the key. There is no "SingleTimeStandard[tm]", and never will be. With the exception of The Wife's, of course ;-)
As a side-note, I don't wear a wristwatch day-to-day ... and haven't in nearly half a century (since my HP-01, back in 1977). In my mind, they are completely pointless. Everywhere you look these days you can see something giving you a pretty good approximation of "local time". Humans living life to the second or minute (or even ten minutes!) is counter productive. Even when baking bread ... Relax, be patient, learn to make homebrewed beer :-)
I do have a dive watch, and wear it when appropriate. It's kinda important in that context.
Parts of the USA already don't observe daylight savings, Arizona for instance.
I would say the majority of the past 20 years of my engineering career has been to try and convince everyone to develop or store in UTC, and include an offset if local time needs to be recorded, and then only in the presentation logic do the conversion, potentially with a provided timezone if converting to a new local time. There is some business logic that are exceptions to this rule, generally around user specified scheduling, but its significantly safer, to use this approach.
Unfortunately I suspect many people don't see this problem, and considering how popular event sourcing has become, I do worry that off the shelf products have not thought this through, and will lead to Y2K2.
What many people don't know is timezone boundaries actually change over time, as hinted at by the registers reference to the tz-database. there are cases where a small town might find its timezone changes if it happens to be in the wrong place.
Federal law defines the boundaries between US timezones, and allows individual states (and populated territories) to decide on whether or not to stay on standard time year round. (The Navajo Nation’s lands overlap with parts of three states; Arizona is on year-round standard time, and New Mexico and Utah aren’t. The Navajo Nation was able to choose to be on daylight saving time on all of their lands, even in Arizona.) In the case of a state or territory that stretches across more than one timezone, the state or territory could make a decision for each timezone-specific part. Federal law does not currently give states and territories the option of going on daylight saving time year round.
I'm surprised a clever guy like Marco hasn't realised they can just move the continental USA 15 degrees west?
Hawaii will be closer.
Russia will be further away.
Puerto Rico will be out in the Atlantic and can become an offshore tax haven for Florida's politicians, without giving foreigners money for the privilege.
I found this entertaining rant last night. I think we've all been there to some extent.
The last time they piddled with DST and changed it. They were taking flak for not doing anything about energy shortages or something. And expanding DST let them say "We did something". Didn't matter if it was wrong or dumb. Pretty sure they can't tell the difference.
Well they want to say they did something again.
I wonder if I still have a copy of my last email to resend.
This isn't another Y2K. This is another 2006, a deja vu all over again moment.
A little history - in 2006, the US decided to alter DST definitions from 1st Sunday in April through last Sunday in October, moving the changes to 2nd Sunday in March to 1st Sunday in November. Windows 2000 systems were pretty easy to address. Using an application called TZEdit.exe, one was able to enter into the time zone tables and adjust the DST definitions from there for your time zone, restart the system and all good. However, Windows XP had an embedded database that was not able to be edited in the same way. Ditto for newer server platforms. In this instance, Microsoft released updates via Windows Update. When dealing with Process Controls, including HVAC and building security access systems, time is a critical thing - devices need to be accurate synchronized together, and actions taken based on time definitions, such as building occupancy schedules, but certainly not limited to that, suddenly malfunctioned in environments that had no internet access and disallowed Windows Updates due to the damage these can do to front end SCADA software. To address this condition, downloading redistributable patches for XP (SP 2 and SP 3) as well as for Windows 2003 server were necessary and field techs ran around like chickens with their heads cut off trying to apply these patches to supported systems prior to the change taking effect. Turned out the time zone patches from MS were ineffective. New patches were released, which worked... for one year. If memory serves, it took 3 or 4 patches to get it right, and new operating system had the new definitions baked in, but will require similar work to address. I am sure similar success will be realized.
Then you have tech and travel. We have many time zones across globe, let alone 4 across the USA alone. Not all of them will change to follow our lead. This will unnecessarily create huge process controls issues across the board. And many of these environments are still using Windows XP, which hasn't been supported in eons, especially the CE edition, which saw support drop in 2020. Many HMIs still run on Windows 6 CE.
This is not a good idea.
There would be six time zones in the US. Eastern, Central, Mountain, Pacific, Alaska, and Hawaii. If you count Puerto Rico and the US Virgins and Guam and Wake and American Samoa and Federated Micronesia, then at least two more. One will be on the other side of the International Date Line.
It could be worse. It could be Russia. 11 time zones. That’s insane.
... is October (an hour longer than the others with 31 days).
Incidentally, the USA runs a research station at the South Pole. Can anyone tell me which time zone it is in? (and does it have Daylight Saving Time?) ... given that in the southern summer they have 24-hour daylight.
Officially Amundsen–Scott South Pole Station has no time zone, but it could be almost[0] any of them for what should be obvious reasons. However, by convention, because most flights arrive from or depart for New Zealand, they run on New Zealand time. Yes, they follow DST per NZ.
;0' I'll leave why I say "almost" as an exercise for the reader.
The EU has voted to end Winter Time/Summer Time; the vote went through in 2018. It was due to come into effect in 2021, but it seems to have got stuck somewhere. Each country could decide if it would prefer to be on local summer time, or local winter time.
The UK has decided, post-brexit, that it will keep changing it's clocks
The Register 
Biting the hand that feeds IT
Copyright. All rights reserved © 1998–2024
