Pssssh, get rid of all those silly metric units like Dynes and A4 paper. A large number of us here are in the good ol' US of A and hereby demand that this be rewritten in terms of Imperial or US customary units - how about sheets of Letter paper?

I'm with the sibling post - if it's really the electrical conditions socket, that could be the result of a dangerous wiring defect. On the other hand, it could be unrelated - like an enclosed light where not enough heat escapes to keep the electronics in the CFL from overheating. Or a crappy brand of CFL; I used some dirt-cheap Ikea CFL bulbs in a bathroom fixture in an apartment about 5 years ago, and about half of them burned out. None of the other CFLs I bought around the same time or earlier (even the ones that were used just as often, if not more) had any issues. The ones that didn't burn out became so dim when first turning on that I had to turn them on at least 15 seconds before they were really beneficial relative to the ambient light most of the time.

But yeah, I have some incandescents hanging around too, for sure - mostly in areas that hardly see any use (like a laundry closet), or where I figured I'd just let the current bulb burn out before replacing it with something else.

I'm keeping a bunch of the old incandescents around, though. Maybe I'll use them if/when I sell my house, or maybe they'll appreciate in value now that they're becoming extinct...

After replacing most of the lightbulbs in my house with LEDs, I calculated the total lifetime cost of the downlights I replaced. Surprisingly, the cost of the bulbs is about even in my case for LED ($30), CFL (3 * $10), and incandescent (12.5 * $2.37), taking into account the shorter lifespan of the cheaper bulbs. So there's really no minimum amount of energy savings needed to make it worthwhile. Over the 23-year lifespan (25,000 hours on, 3 hours per day) of an LED bulb, it will cost $39 in energy, compared to $48.75 for a CFL or $211.25 for an incandescent.

LEDs are also now available in a variety of color temperatures and shapes - some traditional, some slightly less so. Many of the ones in my house are Daylight temperature - about 6500K - because my wife likes the bluer (and often, perceived "brighter") light. Similar bulbs are also available in warm and cool white (2700-4500K).

The only thing they don't do is become warmer when dimmed; many people might be used to incandescent dimming, where it gets yellower as you dim it - LEDs will remain the same color, and that might not have the same effect, although I like it a lot - that makes it a lot easier for me to keep my kitchen and dining room lights at half brightness (further saving electricity) without looking like the rooms are being lit by the same yellow sodium vapor lights out on the street. If only I could climb the light pole and replace those too!

Not gonna lie... When I saw the headline, I was mentally picturing Morgan Freeman for several seconds before I mentally slapped myself.

Probably because there's no test for it. If they put the CPU on the board backwards, they'll notice when they try to turn the system on. Too much paste (or too little), and things will work just fine as far as any test is concerned.

When I did computer repair, I once encountered a PowerMac where the heatsink had a manufacturing defect; one of the posts that fit in a hole on the CPU card had a large extra blob of aluminum on it. It was impossible to seat the heatsink on the CPU, though it could still be strapped on (it's just that there was enough extra clearance caused by the malformed post that the heatsink didn't touch the CPU). As far as the customer was concerned, the computer worked just fine for close to 3 years (though I imagine it probably ran fairly slowly sometimes). Then, the CPU finally died and the computer wouldn't boot any more. Luckily, Apple agreed with me that it was totally their fault, even though he was well beyond the warranty period, and they covered replacing both the CPU and the heatsink. But it's not something that any manufacturing test would've caught, unless the CPU got hot enough to literally fry itself. (It was one of the slower models of the day, with not even a heatsink fan, so it couldn't have fried itself immediately.)

Yes, this. My sister at one time was using one, I think a service provided by her phone company. It was mildly annoying the first time I called her and I think she needed to do something to approve my calling her - but maybe you'd be willing to inconvenience a few people to keep your current phone number?

I also like LED bulbs. The biggest areas for me were the kitchen and dining room, which have 5 and 4 can-type lights each. I changed each from a 65W incandescent to a 13W LED. They're also on dimmer circuits, and I find that I usually don't need full brightness; measuring at the power meter, I see that my kitchen lights take around 25-30W when on now, instead of up to 325W. The dimmable bulbs were a little more expensive ($20), but especially considering the reduced rate of failure compared to CFLs (and even more so, incandescent), it's a net win starting in a couple of years.

Over CFLs, I love that LEDs don't need time to warm up for full brightness, and they're available in more color temperatures - my wife tends to like daylight bulbs the most. They're also available in a lot of shapes and sizes, and different models project light in different directions (which can be bad, if you don't pay attention to what you need).

Lucky - my electricity price is about the same as your on-peak price. I use quite a bit less though - on par with the GGP, around 6-7 kWh/day. Most of my major appliances (heat, water heater, stove) are gas; most of my lights are LED; and I generally try to avoid using power if I can - so my desktop is normally only on when I'm sitting in front of it.

From my hourly usage report from the utility, I can see that my baseline power usage (i.e. overnight or when nobody is home) is about 160W; this includes the fridge maintaining its temperature, which is around 80-90W of that. The rest is random little things, like the router, other appliances, etc. I generally avoid A/C - it cools off enough at night and I'm rarely home during the hottest part of the day, so I just open all the windows.

Some day, I might even get solar panels. It probably won't save me much until like 10-15 years in, but I'm a nerd and solar is nifty, soooo....

I'm fine with the terminal showing me the amount; it's just the extra step of pressing "Yes" or "No" that seems superfluous to me. How about just going straight to the signature (or PIN) stage, and giving me an "Accept" or "Cancel" button, and show the amount on that screen as well?

When I travelled to New Zealand, where chip & PIN is common, I was amazed that absolutely every store asked to see the signature on my card. In the US, nobody cares, apart from perhaps one or two very rare merchants that have probably been burned by not checking the signature.

That said, I've never had my card stolen (though I've had it replaced after a possible threat of being skimmed), so as long as my bank is willing to still take the liability if it's stolen, I don't care if the number is written across my forehead. I'm all for ease of use, and signature or PIN verification is an extra step that I don't terribly mind skipping...

Speaking of extra steps, what's the point of the terminals that ask the user to verify the amount? Most of them are tied to the register (so there's no manually keying in the amount), so they're never wrong. (And any situation where something is wrong - an item is scanned twice, or the register's price differs from the label - is usually not obvious from the total amount shown.) I'd be happier if they just skipped the "is this the right amount?" screen and went to "sign here to pay " - the number would still be there for confirmation, and I'd be out of there 2 seconds faster.

Pssssh, plebe. I'm a multi-hundred-trillionaire.

Skepticism and suspicion are fine. However, this is rarely what happens. Nobody (on either side, typically) really wants to hear all the facts - they make a snap judgement on whether it serves their personal goals, whether that be to make their reelection easier, or to impose their personal religious/economic/etc. views on other people, or whether it was proposed by "their side" or if it should be subjected to Not-Invented-Here syndrome.

This article is a pretty poor example of skepticism or suspicion; the author clearly has a bias, and misstates many facts to serve that bias, whether knowingly or inadvertently. Some examples:

"Trading in" is misleading here. There are many agencies that will provide a rebate or discount for replacing an old stove with a newer, more efficient model. What is disallowed is selling your old stove to somebody else. Instead, you need to (for example) sell it as scrap metal, not usable as a stove.

I'm not intimately familiar with the laws of the Puget Sound area, but looking online quickly shows that their rules are similar to the ones in effect in my area. Using old stoves is not explicitly banned except on certain days when air quality is forecast to be particularly poor. And even then, it is never banned for a house whose only source of adequate heat is such a stove - this is a major arguing point of the article, which states that laws like this risk freezing households who rely on wood stoves for heat.

This is just a red herring. Most smokers I see leave their window at least cracked, and in any case, I know very few people who would want to live daily in an area where the air is like the inside of a smoky car. Additionally, the volume of air used by a stove is much larger than the volume of air inside a car; the stove is putting out much more soot, it's just sending lots of air with it.

As I said before, the EPA rules don't say anything about the use of existing stoves - just that any stove manufactured or sold should meet the new requirements.

It's pretty clear to me that the linked article was poorly researched, and written by somebody with an axe to grind.

Please save your lovely Ars rant for a submission that is actually FROM ARS. I mean really, this is Slashdot - I don't expect you to RTFA, but can you at least mouse over the link?

Random comparison: NVIDIA's A New Dawn demo that was released with the GTX 690 a yearish ago has 40,000 strands of hair, and the whole scene has 4 million triangles. So maybe in another 10 years, GPUs will be able to make real-time rendering look like what we saw in Avatar.

There are still a LOT of quality enhancements to be made to video games. You can play Pac-Man at 60fps on a card from 5 years ago (or 10, or 15)... That doesn't mean that people stopped buying graphics cards. As the technology advances, game developers find ways to take advantage of it, making environments more detailed and realistic. When a low-end GPU can render any photorealistic scene (meaning one that is completely indistinguishable from a photo) at 60 fps, I will entertain the thought that maybe we've hit the peak. But by that time, somebody will have come up with more ways to use the technology (looking backwards, this might include such non-pixel-pushing tasks as adding physics, or particle simulations for water, fire, smoke, etc.).

Only the top-of-the-line model had liquid cooling, and only a small percentage failed. I bought mine in 2005 and used it as my primary machine for 5.5 years with no issues. I then sent it to a relative, who I believe is still using it with no issues.

Any mechanical system will fail some of the time. This particular one was a catastrophic, sucky failure, to be certain. And yes, it probably happened more often than it should have; but it was far from a certain demise as you might seem to imply.