Ok, so do the math... let's assume (i.e. I calculated it before, but for argument) that electricity costs $0.11/KWh:
$0.11/KWh * 1 year = $0.11/KWh * 8766 H = $964/KW = $0.96/W
In other words, a 1-watt device run 24-hours a day for a whole year costs about $1. Thus, your 5-watt clock-radio in the guest bedroom is $5, the 22W stereo standby costs $22, the cell-phone charger, the battery chargers, the microwave clock, the standby on the TV, the UPS on the computer -- they all add up to a modest but avoidable cost.
If this scaremongering were true, accident rates would be skyrocketing along with cell phone use.
I think the point is not so much "scaremongering" as it is to identify another likely cause of accidents. If police are not trained to identify specific distractions as the cause of an accident, such distractions will not be reported. If they're not reported then they never get added to the statistics, and people like you can say it's scaremongering.
The trouble is, there's a bunch of checkboxes on a police report revolving around alcohol, weather, and vehicle condition but only a few about other reasons for driver impairment. If we added "cell phone in use" to the list of checkboxes, we'd be on our way to determining whether this is a real problem. In the 1950's and 1960's, it took such research about alcohol impairment to get it put on the accident reports and it has proven to be a contributing factor in a lot of accidents. Whether this is another significant contributing factor is yet to be seen, but to at least begin analyzing it is a start.
Would it be inappropriate to suggest that at least one woman might make "Gnome Bob", where tasks are divided by "rooms" so you'd play games in the parlor and do finances in the "den", and with animated helpers like "Scuzz the Bastard Operator From Hell"?
Re:Sector encryption
on
Online Revenge
·
· Score: 2, Funny
Encryption takes space and with porn, every byte counts. I thought everybody knew that.
The first thing that needs to be looked at is, "is it a benefit to society?" That is, for the patent system, the length of patents, who gets patents, and what is being patented -- all these factors should be weighed against "what would be best for everyone?"
Point-by-point:
The patent system: yes, it's good to provide a temporary monopoly because it fosters invention. It's currently bad because instead of fostering invention, it is chilling it -- if I invent something, it may already be patented. If I try to build one, I could be trounced by any number of companies, therefore I'm going to just keep my mouth shut -- society be damned.
The length of patents: is currently fixed, but it greatly depends on the industry. A 20-year patent on a novel algorithm would chill innovation in the computer industry whereas a 20-year patent on a novel shipbuilding technique may barely be adequate to benefit from the temporary monopoly.
Who gets patents: currently both individuals and corporations can get patents. That means that if you can make a profit, it's worthwhile to make a corporation yet making money is not the same as benefitting society. For example, a patent-holding company makes money but is a drain on society. My take on it is that only individuals may hold patents, patents themselves may not be sold or transferred, and the transfer of exclusive rights on patents should be limited (including inheritcance to next-of-kin.)
What is being patented: you can get patents on just about anything. I'd like to see "novelty" defined a little more succinctly. A patent based on another patent -- one that builds on the techniques of the other -- is okay but less desirable than a patent that stands alone. A patent that transposes a method from one industry to another should be thrown out or severely limited -- "safety caps" on bottles, for instance, are little more than glorified ratchets. Patents should never be given for things that apply something in the public domain or in nature in a new manner -- Velcro, for instance, is just burdock.
A reliable source has also noted that it's easy for anyone to spot a pedophile. The abstract outlines its scienticfic validity:
Ever see a guy at work or school who sends off creepy vibes, and you say to yourself "man, I know that guy rapes children"? Some mental health doctors claim that there's no way to tell a pedophile apart from anyone else just by looks alone. Wrong. I scoured the FBI's most wanted list and found some examples that confirmed my theory.
Often these old machines are highly inefficient in terms of the computing power they provide vs the electricity they consume.
Yes, computing power-per-watt is less, but:
Sometimes that's all the computing power someone needs. Further, using an older machine that requires less power (i.e. my Mac LCIII ran at about 20 watts versus 70 watts for my PowerBook) would therefore be more efficient.
You also need to consider the energy and resources to dispose of the old machine. Don't forget about the non-dollar costs such as environmental damage -- even though it's "free" (i.e. my area; albeit through taxes) to put a computer in the garbage, there are other costs to consider.
You also need to consider the energy and resources it takes to build the new machine. Plus you need to consider the resources to dispose of the new machine.
Taken together, I would bet that continuing to use an old machine would be more cost effective.
If you went to get your car repaired and left your CD's and photo album on the seat, you wouldn't expect them to clean your car so it looked just like knew -- you'd expect that you'd get your music and pictures back. In the case of computer warranties, it's like they would strip the car down, reinstall new upholstery, and throw out everything you had.
It struck me reading about "loaner computers" -- imagine if you swapped out your current computer with one that was identical but without any of your data.
Obviously they can't guarantee they will save your data, but they should at least try. I mean, if your car burns up under warranty, you can't expect that they'll be able to save your CD's and photos. Likewise, barring hard drive failure, the data should all be intact and able to be moved to a new system.
There's no way to get rid of duct tape. It has been so well marketed that people actually believe it is superior to other tapes. There's no force stronger than a belligerently held opinion.
I mean, the duct-tape has a gummy glue that dries out, the fibrous tape tears easily, has poor high-temperature properties, and is not waterproof. What more could you want?
I've got a Kill-A-Watt tester and I thought the same thing about my wall-warts for things like USB hubs, my PDA charger, cell-phone charger, etc. I plugged them all into a power strip and they use 16 watts total. I then wired up an empty PC case with a switching supply to power most of the devices. I just used diodes to drop 0.6V at a time from the various taps (12V, 5V) to get to the levels I needed for the oddball devices (the few that need something other than 12V or 5V).
I wired it all up and: 16 watts again.
It was exactly the same between using all the individual supplies and using the centralized PC supply. Admittedly, 16 watts isn't exactly ideal for a 90-watt supply (hmm... maybe I'll try a smaller source supply...) but at least I get a nice solid 5 volts going to the USB hubs.
If you get one of those Kill-A-Watt (or equivalent) meters, it's a great help in figuring out what you might want to put on a power strip and switch off manually. My stereo components when off drew a total of about 50 watts so I started switching them all off. The battery chargers in the basement used about 10 watts total, but since I was only using them to keep batteries topped-off, I could reduce it by putting them on a timer and running them an hour a day instead.
In essence, do your experiments and figure out how much you'll really save.
This seems to go a long way to protect children from Lifetime movie stereotypes. A quick Internet search reveals
lots
of
sites
that
indicate
a
child
is most likely to be
abducted
by someone they already know.
How exactly will this system help? By providing technology that few understand, people will be tempted to lean on it and begin to ignore their instincts. If a seldom-seen uncle appears out of the blue to pick up a child a few hours early, even if they pass the scanner, it would still warrant a call to the parents — just to check. The temptation of an administrator would be to think, "well, if they passed the scanner, they must be okay," so I'd think it might not be all that useful. After all, a policy change and an ID check at entry doors would do the same thing without so high a cost.
Second, and no less important: how reliable is the system? Can it false-positive? Can I hack in to add an malicious record? Umm... yeah... can I see the source code just to make sure it's secure?
I set up an X-10 based home automation system using an old computer (programmed by me in AppleScript running on a PowerMac 7100, if you must know) that has a "home" and "away" mode -- for instance, any time I'm away, the vacation timer automatically runs, whereas when I'm home, the driveway IR sensors will activate the plug-in chimes. (My friend has a similar system which she refers to as "turning on" or "turning off" the house.) The gist is, in both our cases, that when we leave or arrive, we hit a button to change modes.
Among the next steps in my own project is to do a better job of temperature control. I too have a varying schedule and the programmable thermostat just doesn't cut it. My solution will be to have two thermostats that get switched with an X-10 controlled relay. One will be my existing programmable one for the "high" temperature and the other the "away"/always low temperature. The idea is that while I'm away (which can range from an hour to over a day) the computer would eventually (i.e. using a time-lag to ignore short trips) switch to the "always low temperature" thermostat and avoid wasting heat.
One of my own goals (which may or may not be as high on your list) is low cost. An X-10 controlled relay (with parts from the scrap pile) will cost less than $40 — the system advised in the parent post is already $240, and it's not clear that it is exactly what you want.
I haven't seen mention of the sort, but I quit NetFlix when they patented their business model — http://ir.netflix.com/ReleaseDetail.cfm?ReleaseID= 112258.
At that point, any dissatisfaction with their service was moot for (as far as I was concerned) competition was illegal.
Let me just run through some numbers to see what makes sense.
Assume a vehicle that weighs 2000 pounds (2000 pounds of force downward.) Based on the pictures, I'm going assume the "ramp" is 4 feet long and vehicles are travelling 60 miles per hour. For simplicity sake, I'm assuming the "ramp" is just an area of roadway that will sink as a vehicle moves over it.
The vehicle, therefore, will be on the ramp for 45 milliseconds. In 45 milliseconds, it will drop (from gravity) 0.389 inches.
The energy to move 2000 pounds 0.389 inches is 87.9 joules or 0.02 watt-hours.
I think I messed up the problem somewhere... so where does this 10 kilowatts come from?
D'oh! I'll concede this specific example... my argument was largely based on two things:
I tried a Java-based office-bundle package a year ago and it was so resource-intensive and intolerably slow that I assumed Java just hasn't changed much.
A quick glance at the old resume reveals that my Java experience ended in 1999 or so and hence is woefully out of date.
However, I think part of my point is still valid: that efficiency and elegance are not correlated in object-oriented programming, and particularly in Java. Your argument (and most others in this thread and in the original article) is that the optimizing compilers allow elegant code with good OO practices to work efficiently.
I'm using a particularly trivial example, and because of that, it would require an inordinately convoluted trick to fool the compiler. For instance, if the copy constructor had additional code that the compiler was unable to fit into a pattern (I don't know... make a call to an external C function) then it couldn't validly optimize out the constructor call and allocation of a new object.
What I'm getting at is that in a procedural language, if an algorithm is elegant, it is efficient -- a programmer can rely on direct assumptions like, "the library calls I use in this algorithm are explicitly written to be fast." In this Java example, there is a further assumption that "the compiler will be able to optimize out the allocation and construction of the Point object that I have specifically requested."
Do you see what I'm getting at? That the good OO programming practice of coding "Point otherLocation = other.getLocation()" is done because the copy constructor for Point may have some side effects that are important -- this specific example is too trivial, but I think it's obvious that more complex object hierarchies can have issues like this (for instance, a serial unique ID [but since it is never used before the object goes out of scope, that would probably be optimized out too, right?]). However, the efficiency of Component.getDistanceFrom() now depends on the copy constructor being able to be optimized out, and that seems to be a rather non-obvious requirement.
On the other hand, it may be that all the bases are covered. For instance, maybe the rule for an object that claims to be efficient is that copy constructors shall only make copies of members. More likely than not, it's that I have degree in procedural programming but almost all of my OO training is on-the-job, and that modern OO computer programming courses cover that stuff just as the procedural courses I took did.
I've never believed that Java's garbage collection is the root cause for its slowness. I do believe that Java's GC is the cause for its random (and more notably, its inconveniently timed) stutters.
I think the more general Java slowness comes from the obfuscation of efficiency. In C for instance, ugly code correlated with inefficient code. This is no longer the case for object-oriented programming in general, and it is possibly worst correlated in Java.
The example in the article provides a starting point for what I'm saying. It's based on the algorithm for a point in some Cartesian graphics system:
public class Point { private int x, y; public Point(int x, int y) { this.x = x; this.y = y; } public Point(Point p) { this(p.x, p.y); } public int getX() { return x; } public int getY() { return y; } }
public class Component { private Point location; public Point getLocation() { return new Point(location); }
public double getDistanceFrom(Component other) { Point otherLocation = other.getLocation(); int deltaX = otherLocation.getX() - location.getX(); int deltaY = otherLocation.getY() - location.getY(); return Math.sqrt(deltaX*deltaX + deltaY*deltaY); } }
The Component.getDistanceFrom() algorithm is considered "good OO style." By using bad style, though, you can break the pretty encapsulation -- and make "bad OO style" -- but get a performance gain:
public double getDistanceFrom(Component other) { int deltaX = other.location.getX() - location.getX(); int deltaY = other.location.getY() - location.getY(); return Math.sqrt(deltaX*deltaX + deltaY*deltaY); } }
Both code blocks require the math (two subtracts, two multiplies, an add, and a square root) but the original block (unoptimized) also requires the allocation of the Point object and the two memory copies to store the (x,y) location.
This is a trivial example, but my point is that in a complex Java project, readable and elegant code bears no correlation to fast and efficient code. I believe this is why Java is slow.
Slashdot Mirror
$0.11/KWh * 1 year = $0.11/KWh * 8766 H = $964/KW = $0.96/W
In other words, a 1-watt device run 24-hours a day for a whole year costs about $1. Thus, your 5-watt clock-radio in the guest bedroom is $5, the 22W stereo standby costs $22, the cell-phone charger, the battery chargers, the microwave clock, the standby on the TV, the UPS on the computer -- they all add up to a modest but avoidable cost.
Wait, so the White Rabbit is always in a hurry, but it's the Mad Hatter who has the accurate clock?
Would it be inappropriate to suggest that at least one woman might make "Gnome Bob", where tasks are divided by "rooms" so you'd play games in the parlor and do finances in the "den", and with animated helpers like "Scuzz the Bastard Operator From Hell"?
Encryption takes space and with porn, every byte counts. I thought everybody knew that.
Point-by-point:
Yes, computing power-per-watt is less, but:
Taken together, I would bet that continuing to use an old machine would be more cost effective.
It struck me reading about "loaner computers" -- imagine if you swapped out your current computer with one that was identical but without any of your data.
Obviously they can't guarantee they will save your data, but they should at least try. I mean, if your car burns up under warranty, you can't expect that they'll be able to save your CD's and photos. Likewise, barring hard drive failure, the data should all be intact and able to be moved to a new system.
I mean, the duct-tape has a gummy glue that dries out, the fibrous tape tears easily, has poor high-temperature properties, and is not waterproof. What more could you want?
Half the mission is accomplished! Now, can I have sex with it?
I wired it all up and: 16 watts again.
It was exactly the same between using all the individual supplies and using the centralized PC supply. Admittedly, 16 watts isn't exactly ideal for a 90-watt supply (hmm ... maybe I'll try a smaller source supply ...) but at least I get a nice solid 5 volts going to the USB hubs.
If you get one of those Kill-A-Watt (or equivalent) meters, it's a great help in figuring out what you might want to put on a power strip and switch off manually. My stereo components when off drew a total of about 50 watts so I started switching them all off. The battery chargers in the basement used about 10 watts total, but since I was only using them to keep batteries topped-off, I could reduce it by putting them on a timer and running them an hour a day instead.
In essence, do your experiments and figure out how much you'll really save.
"In Soviet Russia, cell phone serves YOU!"
Just saying it gives me tingles!
How exactly will this system help? By providing technology that few understand, people will be tempted to lean on it and begin to ignore their instincts. If a seldom-seen uncle appears out of the blue to pick up a child a few hours early, even if they pass the scanner, it would still warrant a call to the parents — just to check. The temptation of an administrator would be to think, "well, if they passed the scanner, they must be okay," so I'd think it might not be all that useful. After all, a policy change and an ID check at entry doors would do the same thing without so high a cost.
Second, and no less important: how reliable is the system? Can it false-positive? Can I hack in to add an malicious record? Umm ... yeah ... can I see the source code just to make sure it's secure?
dada21 wrote:
I have nothing against Monet, but I'd certanly like other options like boudinization, renoirization, or even rembrandtization.
Among the next steps in my own project is to do a better job of temperature control. I too have a varying schedule and the programmable thermostat just doesn't cut it. My solution will be to have two thermostats that get switched with an X-10 controlled relay. One will be my existing programmable one for the "high" temperature and the other the "away"/always low temperature. The idea is that while I'm away (which can range from an hour to over a day) the computer would eventually (i.e. using a time-lag to ignore short trips) switch to the "always low temperature" thermostat and avoid wasting heat.
One of my own goals (which may or may not be as high on your list) is low cost. An X-10 controlled relay (with parts from the scrap pile) will cost less than $40 — the system advised in the parent post is already $240, and it's not clear that it is exactly what you want.
I haven't seen mention of the sort, but I quit NetFlix when they patented their business model — http://ir.netflix.com/ReleaseDetail.cfm?ReleaseID= 112258.
At that point, any dissatisfaction with their service was moot for (as far as I was concerned) competition was illegal.
Let me just run through some numbers to see what makes sense. Assume a vehicle that weighs 2000 pounds (2000 pounds of force downward.) Based on the pictures, I'm going assume the "ramp" is 4 feet long and vehicles are travelling 60 miles per hour. For simplicity sake, I'm assuming the "ramp" is just an area of roadway that will sink as a vehicle moves over it. The vehicle, therefore, will be on the ramp for 45 milliseconds. In 45 milliseconds, it will drop (from gravity) 0.389 inches. The energy to move 2000 pounds 0.389 inches is 87.9 joules or 0.02 watt-hours. I think I messed up the problem somewhere ... so where does this 10 kilowatts come from?
They probably have a list of business practices like Google does. I'm pretty sure one of eBay's is "be evil."
The really strange part, though, is that while the book is automatically thrown in the garbage can, all my other books stop working.
D'oh! I'll concede this specific example ... my argument was largely based on two things:
However, I think part of my point is still valid: that efficiency and elegance are not correlated in object-oriented programming, and particularly in Java. Your argument (and most others in this thread and in the original article) is that the optimizing compilers allow elegant code with good OO practices to work efficiently.
I'm using a particularly trivial example, and because of that, it would require an inordinately convoluted trick to fool the compiler. For instance, if the copy constructor had additional code that the compiler was unable to fit into a pattern (I don't know ... make a call to an external C function) then it couldn't validly optimize out the constructor call and allocation of a new object.
What I'm getting at is that in a procedural language, if an algorithm is elegant, it is efficient -- a programmer can rely on direct assumptions like, "the library calls I use in this algorithm are explicitly written to be fast." In this Java example, there is a further assumption that "the compiler will be able to optimize out the allocation and construction of the Point object that I have specifically requested."
Do you see what I'm getting at? That the good OO programming practice of coding "Point otherLocation = other.getLocation()" is done because the copy constructor for Point may have some side effects that are important -- this specific example is too trivial, but I think it's obvious that more complex object hierarchies can have issues like this (for instance, a serial unique ID [but since it is never used before the object goes out of scope, that would probably be optimized out too, right?]). However, the efficiency of Component.getDistanceFrom() now depends on the copy constructor being able to be optimized out, and that seems to be a rather non-obvious requirement.
On the other hand, it may be that all the bases are covered. For instance, maybe the rule for an object that claims to be efficient is that copy constructors shall only make copies of members. More likely than not, it's that I have degree in procedural programming but almost all of my OO training is on-the-job, and that modern OO computer programming courses cover that stuff just as the procedural courses I took did.
I've never believed that Java's garbage collection is the root cause for its slowness. I do believe that Java's GC is the cause for its random (and more notably, its inconveniently timed) stutters.
I think the more general Java slowness comes from the obfuscation of efficiency. In C for instance, ugly code correlated with inefficient code. This is no longer the case for object-oriented programming in general, and it is possibly worst correlated in Java.
The example in the article provides a starting point for what I'm saying. It's based on the algorithm for a point in some Cartesian graphics system:
The Component.getDistanceFrom() algorithm is considered "good OO style." By using bad style, though, you can break the pretty encapsulation -- and make "bad OO style" -- but get a performance gain:
Both code blocks require the math (two subtracts, two multiplies, an add, and a square root) but the original block (unoptimized) also requires the allocation of the Point object and the two memory copies to store the (x,y) location.
This is a trivial example, but my point is that in a complex Java project, readable and elegant code bears no correlation to fast and efficient code. I believe this is why Java is slow.
You need smarter friends is all. Are you a hippie?
For more torque, maybe they should have used the motor from a LaserDisc player ... same stable operation, but able to spin heavier discs.