For a Bit of Colored Ribbon
codinghorror.com
codinghorror.com
I sat down and calculated it costs them roughly a million dollars per year in postage to print and send those out.
I emailed the managers insisting they just put it online and use the million dollars to fund low income people from getting their power turned off and avoid the $50 fee to reinstate power.
All they would offer to do is take me off the mailing list.
If they insist on snail mailing stuff, they should only mail the worst offenders and radically reduce the cost. I use less than 200kwh per month in the winter and I still get those things.
The super stupid thing about those graphs is they include landlords with vacant apartments and houses that keep the power on for maintenance and easier new tennant move in (they just read the meter and change owner on the bill).
Further, the company in question does A/B test various ways of sending the report, including email-only to find the most cost-effective way to drive down consumption.
Finally, the neighbor selection criteria DOES account for empty homes and other demographics, where that information is available.
Also it's downright stupid to waste resources and postage to mail people who aren't offenders and can do little more to optimize their power use other than turning off the mains.
Just mail worst offenders, people above the averages and save hundreds of thousands of dollars on what becomes junk mail.
Sorry I am too lazy to Google for the paper. The point is: careful with such experiments, the outcome might not always be what is desired.
I could believe that a few people who are in the "below average use" bucket might take these reports as a confirmation that they can relax their energy conservation efforts, but I can't believe that any meaningful number of people would actively increase their usage to "catch up" to their neighbors.
Must resist... :-/
Edit: damn it... Luckily it was the second hit http://www.slate.com/articles/business/the_dismal_science/20...
Why would some energy-conscious Republicans all of a sudden become power hogs?"
The article says that a particular demographic of republicans increased their usage by 1 percent. Then describes them as power-hogs for their 1 percent increase. Thats a bit difficult to take seriously. All the examples of change seem to have been cherry-picked to try and make an interesting article, rather than forming an interesting picture as a whole.
I'm also having trouble parsing the tables in the original paper, but it appears that there are groups with a worse treatment effect (e.g. college educated with above-average use) that the authors did not highlight. Perhaps this is a misreading on my part?
I think it would have been better if they had made that experiment in a more neutral setting (psychology students perhaps), the aspect of the political parties is distracting. The main point is that nudges can go wrong.
Anyway I realized then that what ever program generated those reports, the programmer had been told that there wouldn't be any 'good' houses so that text never got into the code. I stopped paying any attention to their propaganda at that point.
As that was 9 years ago I'm sure they are much better now than they were but still.
Part 1, here are the base lines
Part 2, here is where your power usage puts you
Part 3, here are all the things you can do to reduce your energy usage including, if active, any rebates for various programs.
The funny part was that Part 3 was always created on the assumption that part 2 was significantly higher than part 1. It made me wonder if they were fudging the numbers in part 1 to make that always true.
That is an amazing insight and much more interesting than power bills.
More likely is that Jeff's house is poorly insulated relative to his neighbors. Or that the "average" presented is extremely skewed and not actually representative. (Is Jeff's house actually <1200 sq ft?) It might be interesting to see how many households get reports indicating that they are "above average".
It doesn't matter how low energy your servers are they are still servers than most people don't have.
Jeff's recent server build (which I believe is going to a colo and not sitting in his house anyway), uses 31 watts at idle and 87 watts at full load. Assuming the middle is typical (and this is likely a serious overestimate), you're looking at 59 watts, or 10kWh/week. You could offset this with a more efficient furnace, or better insulation. You could also more than offset this by hanging 3 loads of clothes per week instead of putting them through an electric dryer (~5000 watt).
Changing light bulbs types is like a blind hit.
My brother installed solar panels and could thus monitor consumption. He noticed abnormal consumptions and with his kids search the origin. They found a useless heater left on in some basement room. Problem solved. The nice thing is that it turned out as a game for the small kids and at the the same time they became more careful about devices consuming electricity for nothing.
Of course, if they are successful in reducing overall energy consumption, they will increase the cost of energy to cover their costs and maintain their profit margin.
http://theconversation.edu.au/air-conditioning-is-peaking-ou...
I don't deny that they're efficient, but are they so efficient they're worth spending 3-5x the price of a standard energy efficient bulb?
Don't they last like a billion years longer than even compact florescent blubs?
If an incandescent or CFL goes - it's done for. But a quick check of my LED bulbs shows 22 emitting diodes. When it fails, is it more likely that the controlling circuitry goes or that an individual diode goes and it carries on at slightly reduced output? I'm waiting to see, but they've already outlasted the average lightbulb in my house by a long way - so I don't know how long I'll have to wait to verify.
LEDs use current-regulated power supplies. Lets say your bulb had two chains of eleven LEDs in parallel. When one LED burns out the whole chain is lost and the regulator would increase the voltage across the remaining chain until it was using enough current for two. Naturally this will greatly shorten the life of the remaining eleven LEDs.
This is the worst case scenario and probably won't happen. Two ways around it.
1) Since current is (non-linearly) proportionate to voltage, the regulator need to increase the voltage to compensate. But all regulators have a maximum output voltage. If the regulator is cheap and running near its limits, it can't increase the voltage high enough to damage the remaining chain and the bulb will run a little hot but be fine.
2) LEDs get warm and should not operate above 85C. A good regulator has a thermistor connected to the LEDs and will automatically drop the current to safe levels. The bulb will once again run a little hot but will be mostly fine.
If there are more than two wires connecting the regulator to the LED it might have over-temperature protection. Or if the regulator is physically connected to the heatsink.
The regulators themselves are pretty reliable. I work mostly with low voltage DC lighting (12V-48V, 20W-100W) and there is nothing in the regulator that would wear out easily. Line AC might be a different story, those probably have large filtering capacitors. If the caps are electrolytic they might dry out eventually from the heat.
Also, do you guys use natural gas for heating too?
Although I was slightly surprised that "lighting and other appliances" is as much as 25% of the home energy use.
Usually a home will be built with all gas appliances or all electric, a mixture of the two is not as common.
There are cultural differences too. Americans love to air condition their entire homes with central systems and run them a lot in the summer months. They also dry their clothes with automatic dryers and rarely use clothes lines anymore.
Also, thermal curtains (aka, black-out curtains) can help if you have a lot of window surface area.
Just like petrol/gas, spend your money to buy fuel efficient cars, petrol costs more, no noticeable difference in the bank. Everyone has to get on the wagon or be priced out. It's like an armed robbery, but the gun is an increasing cost of living, with the threat of jail if you can't pay.