Daylight Waste of Time

This post is a couple of weeks late, but I feel like it’s important for people to understand just what an enormous waste of time the whole Daylight Saving Time concept is.  It has literally zero good or useful qualities and everything it touches turns to total garbage.

Nicholas!
It is the Nic Cage of time tracking.

I’ve read a lot of blog posts (OK, I’ve read a couple (OK, I’ve heard about some)) criticizing the modern practice of DST begin by saying that it made sense when we were an agrarian society and most of the work and production in the country took place on farms.  This argument makes exactly zero sense — farmers are known for getting up at the crack of dawn.  That’s sorta their thing.  Did you know that the crack of dawn doesn’t magically shift forward and backward by an hour every spring and fall?  Same thing with sunset — I’m not making this up, you can look it up.  It’s probably in books somewhere.  Whether a farmer gets up at 4:30 and goes to bed at 9:00 or gets up at 5:30 and goes to bed at 10:00 is completely immaterial to the farmer, who is beholden to the daylight.  He’ll get up no matter when it happens.  And before you say, “Ah, but the [grain | cattle | wegetable] market opens at a certain time each day,” I would remind you that those markets are beholden to the farmers, and therefore also to the daylight, so there’s no reason they wouldn’t open earlier or later as the daylight grows and fades.  Maybe this is why historically, farmers have opposed the practice.

The other main argument for DST — and the reason it was extended in the Bush administration — is that it somehow saves energy.  The idea here is that a large portion of our electricity still goes to lighting the home (accounting for around 14% of residential energy usage and 3.5% of total US energy consumption), and we expend that energy disproportionately while we are awake after sundown.  So the reasoning goes that if we make the sun set an hour later, but keep our schedules the same, we’ll effectively be going to bed an hour earlier, using an hour less energy to light our homes.  This also seems ridiculous to me.  Not only do recent studies suggest that we’re probably not actually saving any energy by doing this, but the whole concept of only doing this for 8 months out of the year makes no sense.  Why not always be on DST?  Or better yet, why not just… move our schedules back an hour?  From now on, we work from 10-6 instead of 9-5.  Problem solved.  Go home.  Keep your lights on or don’t, I don’t care.

idea
A whopping 30% of commercial electricity usage in the Bay Area goes to illuminating people’s idea-bulbs.

Right now you’re thinking, “But Mysterious Internet Authority, what about that weekend where we get an extra hour of sleep?  That’s a pretty great weekend!”  First of all, it’s “Mr. Ious Internet Authority” to you, and secondly, what about that weekend where we get an hour less sleep?  We reap what we sow, and in this case it’s a day of traffic accidents and lost money from sleep-deprived laypeople trying to navigate a world in which the very concept of time has magically skipped over an hour they normally would have spent resting.  The only good that DST ever did for us — making it dark earlier so we could trick-or-treat at 5:30 — has been stolen from us, since fall back was moved out of late October and into November.

Goge
Thanks, Obama.

OK — so I’ve “conclusively” “proven” that DST has little-to-no upside.  But what about its downsides?  Check this guy out — it’s basically a list of crazy things that happened because of DST!  Mixed in are hilaaaarious cases of, for example, a terrorist plot foiled by the terrorists failing to know what the time on the bomb meant, along with more innocuous stories of twins where the older one is actually born “after” the younger one.  There’s also stories about how one year there were 23 different DST events in Iowa alone, all of which needed to be kept track of for things like train schedules, which cost an estimated $12 million more per year to maintain than if DST had not existed at all.

These problems are neither problems of the past, nor limited to terrorists or Iowa.  For me, professionally, DST is a nightmare.  In my line of work, it is important — actually, it is essential — to be able to store information with what is called a “primary key.”  This key is used to look up data; a key (hahaha) component of a primary key’s effectiveness is its uniqueness, which allows the user to specify a primary key value and return exactly one entry.  This is so key (hahahaha) that the software used to store data requires primary keys to be unique.

As a concrete example, let’s say that I’m storing information about what the temperature was in Baltimore at a given time.  The reasonable primary key for this data is time.  Now let’s say I want to know what the temperature was in Baltimore on, say, Sunday, November 2, 2014, at 1:30 AM.  Looks like I’m SOL — there were two 1:30 AMs on November 2.  So now I either need to include an additional piece of information in my key — “was this hour the DST duplicate hour or not,” which will be “not” in 8,759 out of 8,760 entries in that table — or I need to pick a single time zone to put the data in (EST or EDT, rather than EPT) and then keep track of that every time I want to tie that out with another piece of data (this problem is actually largely solved, but my company is so far down the path of not-best-practices that we actually choose to just ignore the extra hour and store everything in EPT).  This leads to a huge number of problems for the company as a whole (how do we treat the extra hour for products that trade on an hourly basis?  Can we tie out our data to the point where we can even have a view on that hour?) and me personally, as I spend probably ten hours every fall fixing things that were written back when time didn’t suddenly and inexplicably duplicate itself — you know, 99.98% of the year — and the fix is almost always to ignore any duplicates in the primary key of the data, which means if there are real duplication errors (e.g., they post two completely different temperatures for Baltimore at 2:30 AM on Sunday, Nov. 2, and we need to figure out which one is correct), we completely miss them.  Then of course there’s the data that gets posted on an hourly level — what do you do when one of those says that the hour is “hour ending 25” or “hour ending 2_2” …

exaspachu
I feel you, Pikachu.

So I of course propose that we eliminate DST entirely.  However, if we’re going to look at DST, we might as well look at time zones generally.  I further posit that these, also, are misleading and confusing (though not to the same degree as DST), and should be eliminated in favor of a single universal time (say, a Coordinated Universal Time).  There’s exactly zero reason that 12:00 PM has to occur in the middle of daylight (and in fact, it doesn’t — we already shift that by an hour in DST, and near the borders of time zones by up to two hours), and as already discussed there’s no reason that we have to operate on a 9-5 schedule.  In Virginia, we’re (currently) 5 hours behind UTC; why not go to work on a 2PM-12AM schedule?  We’d wake up when the sun rises at 12:00 PM and go to bed after it sets at 1 AM.  Midnight would be 5AM and noon would be 5PM.  The key is that it makes no difference what time the clock says — the only thing that matters is that everybody agrees on its meaning.  In today’s global marketplace, if the entire world can agree on a calendar*, the entire world should be able to agree on a time.


* It’s surprisingly hard (not to say that it’s actually all that difficult — I googled a bit and didn’t find much, but I expected it to be very easy) to find data on who all uses the Gregorian Calendar, but the wikipedia page on New Year claims the Gregorian calendar is in “worldwide use,” and this page cites the Wiki page on the Gregorian Calendar (which doesn’t currently appear to have a list of countries using it) in its calculation that over 96% of the world population currently observes the Gregorian Calendar.  Of course, the Gregorian Calendar has its own problems — even if we all agree that the best way to keep track of dates is to have 12 months and an extra day every four years (but not … every four years…), why have a month with only 28 days ever?   Why not make every month either 30 or 31 days and have it be 7 / 5 for three years and then 6 / 6 for the leap year?


Note: for a (presumably) less biased and more fact-based discussion of the goods and bads of DST, you can check out the wikipedia page (duh), which has a section devoted to this discussion.

Leave a Reply

Your email address will not be published. Required fields are marked *