Daylight Savings Change Could Spell Doom For The World (Well, For An Hour, Anyway)

from the come-on dept

As predicted a few weeks ago, stories of doom-and-gloom about the forthcoming Daylight Savings Time changes causing all sorts of computer chaos are starting to pile up. All the stories compare the potential for damage to the Y2K bug, which turned out to be largely overblown, and if anything, it’s hard to imagine anything too earth-destroying coming from some computers’ clocks running an hour slow. Indeed, the real problem here appears that it will be little more than one of annoyance, such as people forgetting about the time change, then turning up somewhere an hour late, realizing the problem, then changing the time on their watch or PDA or PC or whatever. But, somehow, if the scare stories started last month and are stacking up now, we imagine they’ll just get worse and worse in the run-up to the change in March.

Rate this comment as insightful
Rate this comment as funny
You have rated this comment as insightful
You have rated this comment as funny
Flag this comment as abusive/trolling/spam
You have flagged this comment
The first word has already been claimed
The last word has already been claimed
Insightful Lightbulb icon Funny Laughing icon Abusive/trolling/spam Flag icon Insightful badge Lightbulb icon Funny badge Laughing icon Comments icon

Comments on “Daylight Savings Change Could Spell Doom For The World (Well, For An Hour, Anyway)”

Subscribe: RSS Leave a comment
JP says:

I must apologise on behalf of other Aussies for comment #5.

“The computers change automatically” do they? How do you think that happens, moron? That’s what the article is about…!

The Y2K bug (y3k is still 993 years away) didn’t hit as hard as it could have, because a lot of people spent a lot of time preparing for it.

zcat (user link) says:

I remember y2k...

New Zealand saw a massive surge of web traffic because we’re in the earliest timezone and the world wanted to see how much y2k affected us before it hit everyone else.

There were a few amusing glitches; Auckland airport announced (after waiting a few hours to be sure) that Midnight had passed without incident. The posting was dated “02:58 1 Jan 100”

charlie potatoes (profile) says:


so..dorpus is from Alabama…hmm that explains a great deal.
but now that i have your attention…i propose that since we can save daylight, why not go all out?.. save temperature…we raise the thermometer reading by ten degrees in late fall…..a freezing day in january becomes an almost balmy 42 degress. then come spring we subtract ten degrees.. the 98 degree july day would be lowered to a comfortable 88…im going to call Bush.

raggi says:


I’m sure everyone feels this is being overblown due to the ‘overblown’ Y2K issue, however….

Whilst the issue is WELL known in areas where the change is occuring, it is NOT well known in other areas.

The issue will come with international time based billing services, and I can be almost certain that issues will occur, and it will cost people money.

Furthermore, there isn’t such a good fix for this problem – time servers have fixed slotted time zones, they will not be aware of the change, and with many time dependant services out there, one will have to:

– patch the time server, and client
– stop time correction services
– correct all interactions with all other proprietary systems

I know of at least 20 telecommunications networks globally that due to lack of support for the networks (X.25 based paknet, some Smartzone systems and many others) these software issues will not be fixed, and I also know of at least a few areas where major billing mistakes could arise as a product of the time alterations.

The biggest issue will occur in the instances where ‘automatic time correction’ occurs, as it is more like to become ‘automatic time mis-correction’ from now on – leading to a loss of ‘base-truth’ in the logs – which means the billing information cannot be reliably recovered.

In this event the carriers loose legal standing to bill for that usage, as most would require ‘base-truth’ proof that calls occured, which is now no longer held to be reliable.

Life or death issue? I hope not…

Expensive? Most probably.

Dave Thewlis (user link) says:

Re: Extended DST Effects

Life and death – no. Complications – certainly, if people don’t take appropriate action or automated updates don’t get made (and they are not always available). Nor in some cases will they have been done in a sufficiently timely fashion.

CalConnect (The Calendaring and Scheduling Consortium) is trying to assemble as much information about EDST as possible, especially advisories/patches/etc. for calendaring systems, OSs, and related products, at a which would be worth a look for anyone concerned, and
especially those who feel this is a non-issue or strictly a consultant trick.

Dave Thewlis
Executive Director, CalConnect
(NOT a Consultancy)

Matt says:

Y2K overblown!?

The only reason Y2K didn’t destroy the world so to say is because of people like me who had to find and edit tons of ancient mainframe code to prevent it from happening! Thankfully, we were able to avoid the problem, and did anybody thank us? Noooo…

Just letting you know that Y2K was only “overblown” because we were able to fix it in time. The same thing goes for this DST thing: it’s already been fixed (for several months in some cases) in many affected areas such as the C library (GNU libc for example, as well as uClibc (used in embedded devices usually) and even Microsoft’s C library in some more recent versions of Windows) where the timezone files are maintained (see /usr/share/zoneinfo/ on Debian-based systems at least) and even Sun’s Java libraries (which also implement the timezones independently of libc).

Once we hit daylight saving time this year, most things will continue to work without a problem, and people will wonder why this got “blown out of proportion”. Sure, fixing the timezone issue this time around is far easier than maintaining decades-old code and updating their use of year values (along with all the data associated with it in some cases, ugh), but the general public doesn’t know that…

a says:

The daylight savings issue is a problem. It really isn’t an issue that affects clocks and meeting times. Take a look at how many things are controlled by time. Phone systems, doors, lighting systems, heating systems etc. There are many things controlled by time and ensuring that everything that will be affected is not an easy task. People might say its a joke until they can’t get into their office for an hour, their automated phone system throws people into voicemail systems automatically during business hours or the light signal starts blinking red an hour early.

Those that comment on the Y2K issue being overblown probably have not thought through all of the issues either. Most companies didn’t even have an inventory of all of their systems before Y2K. When 911 happened, the work that was done for Y2K was invaluable in the recovery after 911. Companies had upgraded their systems, mapped out their architecture, improved their networks. Had that Y2K work not been done, recovering in the aftermath of 911 would have been much harder and taken a lot longer.

Old IT tech says:

Yet another person who spent much of 1999 in a lab testing code. Mainframe, UNIX, PC, didn’t seem to matter much… a lot of that code broke the first time we ran it through a simulated Y2K. That code was then fixed and retested, and therefore was a “non event” on Jan 1, 2000.

It would have been bad, if not for all that work.

Annoys me to no end when people say “Y2K was overblown” as a blanket statement. Certainly there were fear-mongering greedy scam artists out there… and there was a LOT of very good work to make it a non-event.

Is early DST as much risk as Y2K? I think not. Some risk? Yeah…

QED says:

Y2K overblown!?

Having run one of the largest Y2K remediation engagements anywhere (X.25? sure), I have some salient observations to impart:

1. Y2K wasn’t a century problem, it wasn’t a millenium problem, it was an epoch transition problem. My favorite demo was — and still is — showing all the flaws in Microsoft Excel in handling transitions across epochs.

2. Y2K was, as several have pointed out, a nonevent because of lots of hard work by lots of folks to make up for shortcomings in their forebears, the invariable charlatanage notwithstanding. We did our jobs well: by Sat PM, Jan 1, 2000, customer was racing to me begging for “at least one failure” to prove that the massive expenditures had been worthwhile. Fortunately, we’d obliged, in the form of a 286-based Sync Research X.25 PAD (R.I.P.) that had been soft-disconnected from the network and replaced online but not physically removed as the planned site visit on Dec 31 busted when the local contact with the TELCO closet keys was out sick. Said PAD failed hard at roughly 2am, but subsequent forensics were unable to discern the cause of that failure: the logs were complete gibberish. Yes, I suppose it _could_ have been coincidence. Right.

3. Y2K was a big deal because it wasn’t just an hour, it was a year, a day, all time forever (pick one). We saved folks’ houses from sheriff’s seizures that would’ve happened had preforeclosure payments not been properly recorded the fact that liens coming due had been satisfied. Etc.

4. “A non-event”? Would you consider a bunch of your checks’ bouncing because your deposit timestamp was off by an hour a non-event? There was (in Y2K) and is (with DST) lots of potentials like that, and more (some have been noted above in previous postings to this thread). One key area is that many of the patches only apply to current or very recent versions of software; anyone running the old stuff (e.g., earlier PeopleSoft versions) may well be on their own.

There IS a positive side of a sort to EDST. Those planning certain nefarious schemes (e.g., visiting the mistress, slaying the MIL) get a bonus hour to cover their butts with alibis, as long as the deed’s done on March 11. “That tollbooth timestamp showing my client was on the LIE right before his dear departed mother-in-law was murdered, your Honor, is a baldfaced lie, as it’s a fact that no computer system can have been trusted with correct timestamps on March 11 because of the confusion surrounding the early conversion to Daylight Savings Time. My client was home in bed. I move for dismissal with prejudice.”



Add Your Comment

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

Have a Techdirt Account? Sign in now. Want one? Register here

Comment Options:

Make this the or (get credits or sign in to see balance) what's this?

What's this?

Techdirt community members with Techdirt Credits can spotlight a comment as either the "First Word" or "Last Word" on a particular comment thread. Credits can be purchased at the Techdirt Insider Shop »

Follow Techdirt

Techdirt Daily Newsletter

Techdirt Deals
Techdirt Insider Discord
The latest chatter on the Techdirt Insider Discord channel...