![[personal profile]](https://www.dreamwidth.org/img/silk/identity/user.png){kind=small}
The two primary modern justifications for Daylight Savings are that it saves energy and that it encourages more retail shopping since people tend to be out doing stuff later if the sun is later.
Neither of these justifications were party of Ben Franklin's original proposal, where he merely took the pragmatic issue of why waste having the sun out in the morning hours at 5am if nobody's awake to do anything with it. His view was much more oriented towards its obvious advantage for farmers. Energy had nothing to do with it though it would indeed save candle consumption. His proposal was actually more in humor than seriousness and he didn't actually suggest changing the clocks themselves. The first clock-changing proposal was from a golfer wanting that little extra time to play before dusk hit. And while Franklin suggested it might be better for farmers as well as merchants, it was the farmers that first led the repeal of DST in 1919, with enough support in congress to override a Wilson veto. Farmers hours are set by the sun and nature itself, no clock will ever change that.
Taking advantage of light was the exact reason the Germans adopted it during WW1, which then (like all WW1 innovations) was immediately adopted by the allies merely for their own survival.
The trouble with the modern justifications is that they are, in fact, completely bunk.
In the 50s, yes, it was more likely that a company had fewer lights on in the office building or factory floor and would take advantage of daylight as much as possible. This is no longer true today. All offices are flooded with florescent bulbs, on across the entire floor even if only one worker is there. No "energy" is saved by changing light bulb usage because light bulb usage simply is never changed. We're in, its on, that's that. Indeed union safety rules have also led to factory floor lights being on practically 24/7.
In fact, more energy is would be used because the modern A/C systems (which also didn't exist in the 50s) have to work harder when the sun is out than they would if darkness started to kick in.
A study in 1975 found that DST saved a whole 1% of electricity costs. In terms of nuclear power output, contrasted with trying to reduce coal plant production and emissions back then, 1% is negligible. Mexico's DST introduction only reduced energy by 0.7%.
So much for saving energy. It doesn't, and really it never has.
Now what about businesses getting more business?
Or really, the question is "so what if its true"?
Because the reality is that while retail may be having a better time at it, all of the support structures that retail depends on are having a *really* crappy time of it. Estimates are that the stock market losses the Monday after the switch, due to the lack of sleep of investors and floor traders, can run into the billions.
This current change has caused the major software makers to have to rush patches to their systems without adequate time to test them, and certainly there is much software in use that is not being properly maintained and will never be right and have to be hand-managed. One of the really insane difficulties of this one is that many small localities chose not to adapt the DST change, which now means that software that has to know about the many time zones has had the number of time zones to track *triple* to account for all of the little variations that have popped up as local legislatures have voted to not adopt the federal proposal. This has naturally led to a large number of bugs in the patches themselves that are trying to fix the problem.
Microsoft has put out 3 separate patches to its Exchange email/calendar system and still hasn't got it right. Meeting schedules have bounced back and forth to an hour off in both directions as a results of each patch, many of which assume that the prior patch had NOT been applied and so overcompensated. If my own company has had this disruption causing every worker to have to correct their calendars, how about a large Fortune 500?
Microsoft also formally did not support fixing Windows 2000, trying to use that as the final straw to get people to upgrade to XP or Vista. Trouble is, otherwise Windows 2K *still works* (as far as "works" means to Microsoft, of course). Most people don't need to, and in fact their fully functioning hardware would not handle the upgrade well since the minimum requirements for XP are higher than a 2K box. They did eventually give in and sell a patch to fix 2K, but at $4000 a site (apparently a little cost and price fiddling to have the 2K patch, which is no bigger than the XP one, cover the expense of fixing XP as well).
Sun has also had to fix Java implementations that, because of compatibility issues, can't be upgraded to more recent versions. Even then, they didn't get it right the first time, now causing 10 people in my company to waste another 2-4 hours each (napkin guess: $100/hr * 10 * 4 = $4000 we'll lose, minimum, all because Sun screwed up - and that's not counting the amount of work our customers will have to do, with a requisite downtime necessitating after-hours work, to apply the patch from Sun that we forward to them, added onto the amount of work we did figuring out and running the first patch we've already sent). I am, of course, one of those 10. My total time has already been 16 hours arguing with and testing this problem, and I'm about to throw another 4 onto that. 20 hours of my time wasted.
And so you know I have now even less respect for the asshole that made me waste 20 hours of my time better spent doing real work.
I hated this administration before for general reasons.
But this time, its personal.
That's just 1 small company, out of thousands. If we have had to waste $100 * 10 * 20 = $20,000 over this issue (and that's being generous), multiply that by the 500,000 companies using Java and you have $10,000,000,000, yes $10 BILLION, wasted.
No amount of "increased retail sales" will ever make up for that.
Ever.
And that's just in dealing with Sun's Java problem. Now figure in Oracle/Sybase/MS SQL/MySQL, MANY Linux versions, often different and rarely upgraded (again, don't knock a system that's working), Solaris and HPUX, the tons of software and libraries used by IBM customers, and it *really* adds up. I wouldn't be surprised if total estimates on this problem actually exceed spending on the Y2K problem 10 years ago. Much of this problem is because the software now can no longer assume that the underlying operating system has been upgraded - Java and Oracle and all of that has to run on Windows 2K or Unix even if the box hasn't been patched, so it has to recognize whether or not the OS is right and adapt accordingly. This is not an easy task, nor is testing it by the end consumer.
To be perfectly honest, we've spent $20,000 and we actually honestly have no idea if any of it worked. We really don't.
And that uncertainty is more costly than any piece of software we've ever written.
It should have been obvious when this was proposed that any time this administration is certain of something, we should all be certain that the only thing that will result is more uncertainty.
You want to know why the economy has *never* really rebounded since 2001? Uncertainty. Wall Street hates it and always will, but unless you're Halliburton or Exxon/Mobile, uncertainty is all this administration has ever given you.
Neither of these justifications were party of Ben Franklin's original proposal, where he merely took the pragmatic issue of why waste having the sun out in the morning hours at 5am if nobody's awake to do anything with it. His view was much more oriented towards its obvious advantage for farmers. Energy had nothing to do with it though it would indeed save candle consumption. His proposal was actually more in humor than seriousness and he didn't actually suggest changing the clocks themselves. The first clock-changing proposal was from a golfer wanting that little extra time to play before dusk hit. And while Franklin suggested it might be better for farmers as well as merchants, it was the farmers that first led the repeal of DST in 1919, with enough support in congress to override a Wilson veto. Farmers hours are set by the sun and nature itself, no clock will ever change that.
Taking advantage of light was the exact reason the Germans adopted it during WW1, which then (like all WW1 innovations) was immediately adopted by the allies merely for their own survival.
The trouble with the modern justifications is that they are, in fact, completely bunk.
In the 50s, yes, it was more likely that a company had fewer lights on in the office building or factory floor and would take advantage of daylight as much as possible. This is no longer true today. All offices are flooded with florescent bulbs, on across the entire floor even if only one worker is there. No "energy" is saved by changing light bulb usage because light bulb usage simply is never changed. We're in, its on, that's that. Indeed union safety rules have also led to factory floor lights being on practically 24/7.
In fact, more energy is would be used because the modern A/C systems (which also didn't exist in the 50s) have to work harder when the sun is out than they would if darkness started to kick in.
A study in 1975 found that DST saved a whole 1% of electricity costs. In terms of nuclear power output, contrasted with trying to reduce coal plant production and emissions back then, 1% is negligible. Mexico's DST introduction only reduced energy by 0.7%.
So much for saving energy. It doesn't, and really it never has.
Now what about businesses getting more business?
Or really, the question is "so what if its true"?
Because the reality is that while retail may be having a better time at it, all of the support structures that retail depends on are having a *really* crappy time of it. Estimates are that the stock market losses the Monday after the switch, due to the lack of sleep of investors and floor traders, can run into the billions.
This current change has caused the major software makers to have to rush patches to their systems without adequate time to test them, and certainly there is much software in use that is not being properly maintained and will never be right and have to be hand-managed. One of the really insane difficulties of this one is that many small localities chose not to adapt the DST change, which now means that software that has to know about the many time zones has had the number of time zones to track *triple* to account for all of the little variations that have popped up as local legislatures have voted to not adopt the federal proposal. This has naturally led to a large number of bugs in the patches themselves that are trying to fix the problem.
Microsoft has put out 3 separate patches to its Exchange email/calendar system and still hasn't got it right. Meeting schedules have bounced back and forth to an hour off in both directions as a results of each patch, many of which assume that the prior patch had NOT been applied and so overcompensated. If my own company has had this disruption causing every worker to have to correct their calendars, how about a large Fortune 500?
Microsoft also formally did not support fixing Windows 2000, trying to use that as the final straw to get people to upgrade to XP or Vista. Trouble is, otherwise Windows 2K *still works* (as far as "works" means to Microsoft, of course). Most people don't need to, and in fact their fully functioning hardware would not handle the upgrade well since the minimum requirements for XP are higher than a 2K box. They did eventually give in and sell a patch to fix 2K, but at $4000 a site (apparently a little cost and price fiddling to have the 2K patch, which is no bigger than the XP one, cover the expense of fixing XP as well).
Sun has also had to fix Java implementations that, because of compatibility issues, can't be upgraded to more recent versions. Even then, they didn't get it right the first time, now causing 10 people in my company to waste another 2-4 hours each (napkin guess: $100/hr * 10 * 4 = $4000 we'll lose, minimum, all because Sun screwed up - and that's not counting the amount of work our customers will have to do, with a requisite downtime necessitating after-hours work, to apply the patch from Sun that we forward to them, added onto the amount of work we did figuring out and running the first patch we've already sent). I am, of course, one of those 10. My total time has already been 16 hours arguing with and testing this problem, and I'm about to throw another 4 onto that. 20 hours of my time wasted.
And so you know I have now even less respect for the asshole that made me waste 20 hours of my time better spent doing real work.
I hated this administration before for general reasons.
But this time, its personal.
That's just 1 small company, out of thousands. If we have had to waste $100 * 10 * 20 = $20,000 over this issue (and that's being generous), multiply that by the 500,000 companies using Java and you have $10,000,000,000, yes $10 BILLION, wasted.
No amount of "increased retail sales" will ever make up for that.
Ever.
And that's just in dealing with Sun's Java problem. Now figure in Oracle/Sybase/MS SQL/MySQL, MANY Linux versions, often different and rarely upgraded (again, don't knock a system that's working), Solaris and HPUX, the tons of software and libraries used by IBM customers, and it *really* adds up. I wouldn't be surprised if total estimates on this problem actually exceed spending on the Y2K problem 10 years ago. Much of this problem is because the software now can no longer assume that the underlying operating system has been upgraded - Java and Oracle and all of that has to run on Windows 2K or Unix even if the box hasn't been patched, so it has to recognize whether or not the OS is right and adapt accordingly. This is not an easy task, nor is testing it by the end consumer.
To be perfectly honest, we've spent $20,000 and we actually honestly have no idea if any of it worked. We really don't.
And that uncertainty is more costly than any piece of software we've ever written.
It should have been obvious when this was proposed that any time this administration is certain of something, we should all be certain that the only thing that will result is more uncertainty.
You want to know why the economy has *never* really rebounded since 2001? Uncertainty. Wall Street hates it and always will, but unless you're Halliburton or Exxon/Mobile, uncertainty is all this administration has ever given you.
