Letters Who would have thought that time would tick you off so much? Daylight saving - do we like it? No we don't!
Bryan, you wrote:
"It wouldn't be the first time that servers have got their clocks out of step, either. After all, the US and Europe only managed to synchronise their shift to DST a few years ago, after decades of being a week or so out of sync."
That is still the case today. Europe switches back to their standard time on the last weekend in October, the US on the first weekend in November. Nothing's changed in that respect. And it still drives people insane for a week every year.
Has the USA ever matched Europe? Or has it just been that the 4th Sunday is often also the last Sunday of a month?
This is going to be a bigger than usual discrepancy, but it's not uncommon.
When OPEC first jacked up prices in 1973, the US responded with winter daylight savings time. I was in large car pool, and often on the road at 6 to be at school for an 8 am class--mine or someone else's. A winter of such experiences destroyed any enthusiasm the public may have had for the measure, and in the fall of 1974 we fell back as usual to standard time.
In those days, of course, the usual computer was room-sized, the minicomputer about the size of a refrigerator, and networking not very far along. Fred Brooks could mention as an example of the "second-system effect" the dedication of 26 bytes of the permanently resident date-turnover routine to leap-year transitions, which "might have been left to the operator".
Here's a crazy idea for you. Why don't we change the hours worked? Start earlier, stop earlier.
Off the wall, I know...
Oh, yes Sir. We do like that suggestion.
I don't see any huge problem, even for systems that are timing critcal.
Firstly, if it's only logging, then most decent systems log in "UTC+offset" so the fact that they log something as "10:23 -0500" instead of "11:23 -0600" isn't that much of an issue - the time might be wrong but it's visible that it's wrong and can be corrected for. I doubt if there are any serously important systems that don't log in this manner.
As for control, well does it matter if (for example) the lifts are an hour late switching to "morning influx mode" ? Similarly, a bank vault opening an hour late may be an inconvenience but is unliekely to be life threatening. Also, if something is too inconvenient, the administrator can always alter the clock manually !
In control systems where there is a link between "X must happen before Y, and Z must happen within 30 minutes" will almost certainly a) be running from one system, and/or b) have other interlocks than just the time on the controller.
In comparison with the Y2K issue, it's going to be insignificant. With Y2K there was a real possibility of systems doing all sorts of strange things, thankfully nothing major happened BECAUSE of the effort that went into dealing with it - perhaps as an industry we should have left at least one high-profile system to "die in a headline gabbing fashion" to demonstrate that the effort was worthwhile !
The real risk of DST is all about the things you don't think about, but cost you money. Credit card authorizations (or failures in this case), package shipments, wireless key based security (usually on a time based key check), energy management systems (it's 9 AM, why aren't the lights on?), cell phone communications, and on and on and on.
The problems, and expense, of preparing for DST changes are all about the things that people take for granted, but are required to make it through the day. It's not about disasters, and havoc, and the end of the world. It's about someone having to pay for the several BILLION it's costing retailers, communications corps, etc. to prepare and deal with the changes.
Think of it like global warming. It has nothing to do with the doom and gloom that some looneys go on about; it's you paying extra every year with nothing to show for it. You can bet no one even cared to calculate that cost when they came up with this mess.
"However, come 1 January 2000, most of the work had already been done."
I seem to remember that countries like Russia which took virtually no action against the millennium bug also survived unscathed, which begs the question whether the bug was ever really a threat to anyone.
So, the week between the last Sunday in March and the first Sunday in April doesn't exist? Or did somebody not check their facts? For as long as the difference in the start of clock-buggering has mattered to me, it's been the case that Europe changes a week before the US and Canada.
Frankly, it's about time DST was abandoned and something close to solar time was stuck to. When will our political masters learn that you can't make a piece of string longer by cutting a bit off one end and tying it to the other?
Yours pedanticly (and grumpily), --Silas
My experience is that every normal DST weekend causes 2-3 times as many actual software failures as did 1st Jan 2000. That is not surprising, because the fact that the year after 1999 is 2000 is a lot less unexpected than the idea that the minute after 0:59 is 2:00. Since this year's US change is happening according to a different schedule, then it can be expected to be worse than normal.
Clearly we should bite the bullet of one more bad year, and abolish DST / BST etc. totally.
Lots of people think (apparently) that the Y2K issue was over-hyped; afterall, not much happened on 1-1, did it? The reason is that all the work WAS done to prevent all those bad situations. If everyone involved THEN had taken the attitude many have NOW (really no problems in the Y2K crossover; just a lot of hype) then there really would have been a mess to deal with, and the solutions would have taken a long time in many cases to fix. Tons of money (that's a technical term) was spent to ensure that the problems would be minimal - and that was a success. At the university where I worked our IT staff did almost nothing else for over a year but to fix and test the systems for compliance.
Talking spreading FUD, will you and the rest of the press stop saying that the Y2K bug was way over hyped and never materialized. No shit it never materialized, we spent the money and the man hours to fix the thing! I'm so sick of reporters 1/2/7 years later going "oh it was all a big push just to get more money for projects...". It was a push to get more money to FIX something so there wouldn't be a Y2K issue. I was a sysadmin at the time (salaried/not consultant) and I spent MANY an hour fixing or finding work around's for old systems. And know what happened to the fixes they weren't willing to spend the money on? They had to hire someone specifically to hand STAMP the correct date on environmental report pages being printed out. Yeah, that looked professional, and sure hiring someone part time for around 15k a year is cheaper the first year and spread out vs paying the 20k to replace the system. But that looked damn short sighted to me!
So stop with the FUD saying the Y2k was nothing. The money was well spent, and that's why you think you can say it was over hyped 'cause nothing happened.
Urgh just ticks me off.
And keeping with the with ticked off sentiment, NHS plans to slap barcodes on patients went down as well as...well, a cup of cold sick.
Whoever thought this one up doesn't sound to have been in hospital lately.
The nursing staff coming around with the drugs trolley, checking the bracelets, and talking to the patients, are going to spot more problems with treatments than somebody waving around some bit of electronics. And this is going to do nothing to stop the other mistakes.
They'd do better to spend the money on teaching Doctors to write clearly. Garbage in, garbage out, as they say.
Um. What process will be used to reduce the probability of misidentification at the bar code enrolment stage to infinitesimal? Since the bar code will subsequently be used as a cast-iron guarantee of identity (otherwise why bother?), then a mistake at the enrolment stage could be catastrophic.
Hmm, reminds me of the Prisoner show.
"Where am I?" "In the Hospital." "What do you want?" "Your Money." "Whose side are you on?" "That would be telling.... We want your money. Money! MONEY!" "You won't get it." "By hook or by crook, we will." "Who are you?" "The new Number 2." "Who is Number 1?" "You are Number 6." "I am not a number — I am a free man!"
Interesting the discussion about the use of RFID in hospitals. Just recently there has been an article here in the Netherlands about the dangers of using RFID in hospitals. As someone who understands RFID then I am also aware of the strength of the fields that have to be generated to get the chips to respond. Not that strong, but can affect delicate medical apparatus it would appear.
<1984>Surely it would be easier to put one of those new grain-of-snad sized rfid chips under the skin on the back of each patients hand?</1984>
However, not all of you thought it was all bad:
I find the resistance against using barcodes misguided.
Hospitals HAVE to clearly identify their patients even for basic safety like ensuring a patient doesn't get the wrong medicine. At the moment most of them fit temporary wrist bands, and if those would contain a barcode it would be no different to the name printed on the current version.
RFID, however, is a can of worms as it can be read from a distance without patient knowledge and consent. A barcode can only be scanned by physical presence, so the identity holder knows it is being checked and is (mostly) in a position to verify that the individual doing the checking is authorised to do so. RFID does not discriminate who it gives information to.
Which leads to a very simple question: why didn't we use barcodes in passports and ID cards?
/// P ///
The idea of bar-code IDs on patients' wristbands, and on their prescription/drug charts, and on their actual packages of medication is one which has been tested, and not surprisingly found good. The most obvious demonstration is in the US Veterans Administration hospitals, where they have been doing it for some years and convincingly report savings - saved lives and treasure.
The VA has been using software called VISTA for 20 years or so, and this achieves one of the other aims of the NHS, that of keeping electronic records and actually working. It runs on M, is available in a gratuit version from the VA, and a libre version from worldvista.org and indeed, if you only need X86 processors, a version of M - GT.M from Sanchez - is also available libre and yes, Worldvista and Sanchez's Bhaskar have put the stack together on a CD. All you need in the way of software to computerise a national health service, should you notice any lying around needing that doing.
A substantial chunk of Mexico did notice, and is doing it - I suspect they may make it ahead of the England & Wales NHS.
I'm a member of OSHCA - open source healthcare alliance www.oshca.org - and so is I think everyone working in WorldVista. Our next meeting, to which health services are of course welcome, is in Kuala Lumpur on 8-11 May 2007.
FLOSS, you know it makes sense.
Regarding wrist bands on patients at NHS hospitals, the "no ID" camp has clearly no experience in the medical field!
I work in the USA for a hospital that has implimented such a system. Every patient gets a wristband with a barcode when they enter the hospital.
It's not so much about patient identification (if you need to know who a patient is, ASK THEM) but it's about severely reducing the number of medication administration mistakes.
Here's how the process works... A nurse is to give a patient some medication, let's say for ease of argument a blood thinner. Said nurse also has a small wireless barcode scanner. Nurse scans THE MEDICATION, then the patient, and waits for the check.
On the back end, the system checks to make sure 1) the patient has orders for this medication and 2) that the *dosage* is correct and that 3) the *timing* is correct. All three of these areas (wrong medication, wrong dosage, and wrong timing / duplicate dose) are severe areas that any competant *health care* worker will tell you they need help with. That's what the system is intended to do!
Obviously, if someone is writhing on the bed with a code, the nurses chuck the scanners and simply administer what they need immediately. But that's not the situation this type of system was designed to address. It's designed to make sure that the nurse that's at the end of a double shift doesn't wind up killing someone by giving them 10mg of the *correct* medicine instead of 10mcg! It happens A LOT more often than health care professionals care to admit, for obvious reasons.
Whether the technology is RFID or barcodes kind of doesn't matter. Occasionally the paranoid part of me runs away and says someone might clone my patient ID number and steal my meds in the hospital. Yeah, *that's* going to happen. :)
Regarding bar coded patient wrist bands.
Here in Sydney, each patient has a set of stick on identification labels. Each label has their name, date of birth, address, and crucially, their unique Medical Record Number. One copy of the label goes into the patient's plastic wrist band. Another copy of the label goes onto every piece of patient documentation. When a nurse gives a medication, she checks the patient's medical record number and name on the wrist band against the label on the medication order. For some classes of medication, the patient's identity is always checked by two nurses.
In practice, the reason for checking the patient's wrist band is not to identify the patient. The nurse has a good idea of who her patients are, and most conscious patients can tell you their name and date of birth. The danger is in picking up the medication order for Mrs. Smith, preparing the drugs, and then giving them to Mrs. Jones. That's why the nurse checks the label on the order against the wrist band. Not to ensure that she has the right patient, but to ensure that she has the right medication chart.
Safe medication administration means right patient, right drug, right dose, right time, right route of administration, checking for allergies, and right documentation that the dose has been given. I'm sure many more errors are caused by illegibly written medication orders than misidentification of patients.
- Mark A. Bell, E.N.
The medical profession did come up with a favourable conclusion this week, when it emerged that chocolate
does could protect your brain. Girls have known this for centuries, fellas...
So chocolate's good for girls, beer's good for guys, and red wine's good for everyone.
Re: More work, they say, is needed.
So they're looking for volunteers to eat chocolate then perform small mental miracles? Count me in. Restraint from giving users a good beating is the mental miracle, and there's a Hotel Chocolat just around the corner. Mines a peppermint block or one of the 70%+ guest recipes.
Get in the queue, mate.
The suggestion was thrown up this week that future lunar explorers should do away with the traditional bunny-hopping technique and learn to propel themselves in the same fashion as cross-country skiiers. Cool or crazy?
Going somewhere with speed is certainly an attractive proposition anywhere, but on the Moon mistakes can very quickly have deadly consequences. A cracked faceplate, a broken air hose, even a small tear in the fabric can spell doom in just seconds. I like speed as much as anyone else, but it is already a hostile proposition on Earth, where the atmosphere guarantees that, if the windshield breaks, my chances for survival are still good. I would err on the side of extreme caution when mixing speed and conditions on the Moon are concerned - unless it's in an undergound, pressurized base, of course.
Skiing on the moon? What nonsense! Any fool knows that the moon is made of cheese and therefore the most efficient mode of transport would obviously be a boat in the form of an inverted fondue set allowing our plucky astronauts to skim around on a film of super-heated liquid cheddar.
Of course the French think it is brie but we know better in good old Blighty and besides, the skin on a brie would impede the melting process and snag both oars and propellors.
Ian McNee Wolverhampton
How do you solve a problem like old kit? Why, flog it online, of course. And this is just what the British forces have done with everything from old crutches to green goddesses....well, they would have done. Had their site been working.
It appears that their servers have fallen over, or at least their SQL Server: Microsoft OLE DB Provider for SQL Server error '80004005' [DBNETLIB][ConnectionOpen (Connect()).]SQL Server does not exist or access denied.
is all I get, shame as I really wanted to drive a yellow goddess in the full red Guards uniform.
Yet another classic UK Government site that falls over as soon it opens.
Re: the MoD flogging off its old kit.
Just tried to access the site and got this: "An error has occurred. Press home to continue"
Which of course just got me stuck in a loop! Just about sums up the MoD's grasp of technology, doesn't it?
-- Pete C.
A former IBM employee sacked for visiting an internet chat room "for a sexual experience during work" has sued the company for $5m:
He didn't log out, and a fellow worker subsequently "saw some chat entries, including a vulgar reference to a sexual act". He reported Pacenza, who was sacked the next day."
So apart from all the mental sexuality on the job, which must probably be totally unlawful and might be just another reason why people in chipmaking have to wear those bunny-suits, the true misdeed of this man seemingly is exposing a co-worker to sadly unspecified "vulgar" language and being in the low-luck situation of said co suffering from acute assholism. Assoholic vs. Assholist - When Worlds Collide.
What do you do if your bikini falls off during a photo shoot? Why, use your iPod to cover up, of course.
Re: iPod spares nude model's blushes
SI is already soft porn masquerading as sports coverage. Why stretch the point and make it masquerade as IT news too? If you really want to go full-blooded red-top why not start a 'Busty Vixens of IT' section and have done with it?
Lester's reply: Works for me.
Your story mentioned: "This caveat also applies to the iPod. " And you know this from "personal experience" in placing an iPod is a place such as this??
Lester's reply: No comment.
LogoWatch brought you news of TomTom's corporate rebranding exercise in the interests of ridding the world of such excessive marketing cruft:
Congratulations on LogoWatch's successful repulsion of the Strategy Boutiques and their Charlie-fueled twaddle. Can I suggest that you now launch a new campaign to get the world 'solution' banned from the corporate lexicon as it really, really gets on my tits!
We thought we'd hand that one over to you, dear readers. Should the word 'solution' be beaten down and banished from business forever? Letters to the usual address. ®