Hell, the only reason costs go down is because we keep moving more and more production to third-world countries. Silicon used to be made in the US, now it's mostly made in Taiwan. Hardware that used to be manufactured in Taiwan, Singapore, and Eastern Europe is now manufactured in PRC, including things like hard drives. DRAM is extremely cheap because the Korean government invested trillions in the DRAM industry and it has extreme overcapacity. The hardware isn't becoming any easier to make, we are just paying the workers less. Obviously, this has very little to do with Moore's law.

ARM Processors consume ALOT less power than X86. With ARM you are talking milliwatts of power used to run the laptop, not watts.

Haha, no way. There are some low-power ARM chips, but they are rather low-performance compared to a modern Intel chip like the Atom (although the OLPC uses a very old and very shitty AMD processor). And the difference in power is like 10%, not two orders of magnitude. If you want reasonable performance, you really can't beat Intel. Their processes are about a year ahead of everyone else, which more than makes up for a slightly suboptimal processor architecture (which isn't even that bad, especially with the 64-bit extensions).

They would have become more successful if they didn't try to do their own software stack. The countries they were selling them to never wanted third-world solutions; they wanted normal computers that could run normal software. A lot of these efforts fail because people think that third-world countries want to be beta testers. In reality, nobody wants to spend millions of dollars on an unproven, half-baked product with no track record, and that's exactly what the OLPC was. They had a lot of ideas, but no ability to execute, no deployment strategy, no quality control, no product support strategy, and no experience developing hardware.

It was never intended as just a replacement for textbooks. There is no point in replacing textbooks. Unless you insist on 4-color printing on thick, glossy paper, textbooks cost pennies to print. Textbooks are expensive because publishers charge a lot of money for the content, and an electronic device does nothing to fix that. For the price of one OLPC computer, you could probably buy 100 textbooks if you didn't have to pay royalties.

Polaroid only tanked because it was managed by incompetents, not because of failures of their technology.

Well, apart from the fact that their market shrunk from hundreds of millions of units to tens of thousands. It's very easy for a company to grow, but it's almost impossible for it to shrink. A factory that is efficient at producing millions of something might be extremely inefficient and cost-prohibitive if you only need to produce a few thousand. I'm sure the entrepreneurs behind this venture will learn about this soon enough.

Yes, but the laptop in question has had its serial numbers removed, which only supports the original premise that it was stolen.

The guy SPECIFICALLY said the serial number and all identifying marks were removed by the previous seller. Coupled with the fact that it was purchased on ebay, I would be 99.9% sure it's stolen. Why else would the seller remove the serial number?

Is that why my local utility is spending tons of money subsidizing and promoting CFLs? Utilities are the biggest proponents of energy efficiency. New power plants and transmission lines cost tons of money, reduced consumption doesn't.

No, they aren't benefiting from it. In fact, you can directly argue that every top 100 download on TPB is a lost sale. What the statistics are basically saying is that the major labels' marketing is working very well, but instead of creating more sales, it's creating more downloads. The labels don't care about how popular their artists are, they care about how many records they sell. I don't think you can honestly argue that their record sales are going to increase as the result of piracy. In fact, I think that their business model is going to be completely gone in another 10 years. Maybe they can reinvent themselves as something else (say, making money by licensing music for commercial use), but it will get harder and harder to sell records to consumers as digital piracy increases.

Probably. But I think the fundamental reason small labels and independent artists are struggling is because they are not publishing music that appeals to a broad range of consumers. The big labels are pretty good about picking out stuff that sells, and artists tend to gravitate towards larger labels. As a result, the smaller independent labels mainly get music that was not accepted by any of the big labels. This is a very narrow niche market that appeals to a very small number of people. All the statistics are saying is that the big labels are doing an extremely good job of picking and promoting music with broad appeal. Of course, that renders such music rather bland, but that's the price of having broad appeal.

I'm not sure how pirates figure into this. If anything, piracy hurts big labels much more than small ones. Small artists typically have more devout fans that would probably be much more likely to support the artists by buying their records. They also don't have a pre-existing business model that's based on selling a small number of hits in extremely large volumes.

Um, those are distribution transformers. They step down from 5 kV distribution voltage to 220V going to individual houses. Generally, power factor is not a significant issue with residential users. It's much more of an issue when you have a factory with tens of megawatts consumed by induction motors (which have a power factor of something like 0.3).

Even if you throw the bulbs in the trash, the environmental impact is mostly positive. The amount of mercury they contain is absolutely trivial, as has been pointed out in other comments.

I don't know why you are paying $15 for CFLs. They cost about a dollar these days. Maybe 3 or 4 for specialty lamps.

If your CFLs are lasting 3 months and giving you poor white balance, you are either buying very cheap bulbs or are using them incorrectly. Good CFLs are practically indistinguishable from incandescent lamps. One reason they might be dying so quickly is dimmers. If you have any type of electronic switch or dimmer on a lamp, you should not use regular CFLs in it. You will damage either the switch or the lamp. I've certainly seen defective lamps, but the average lifetime that I've been getting is on the order of 3-6 years.

It's critical to remember that reactive power + real power = total output of the facility. When reactive power production goes up, real power production decreases. So the idea that these lightbulbs are eating more than their share of reactive power has significant economic implications.

That's a silly argument. If everyone replaces incandescents with CFLs, you will need 2x less apparent power and 4x less real power. And "generating" reactive power doesn't require burning fossil fuels or running generators, it just requires additional equipment (capacitor/inductor banks or active PFC systems). So even in the worst case, the utility still frees up half the capacity used for lighting.

LED lights have exactly the same problem. Any switching power supply has that problem -- computers, cellphone chargers, wall warts, and even dimmer switches. Large fluorescent installations use more expensive ballasts that have power factor correction, so they don't have the problem. The problem is that making a PFC circuit costs money, and CFLs are supposed to be cheap.

Besides, it's really not much of an issue. If utilities didn't see the benefit of CFLs, they would not be encouraging everyone to install them.

The initial power surge to kick off the light will eat into your usage and savings -- very minutely, but still not the benefit you think you're getting.

There is no initial power surge to turn on a light. Where the hell are people getting this retarded idea? It does not take more power to turn on a fluorescent lamp. Any type of fluorescent lamp. The real issue is that some types of old magnetic-ballast fluorescent lamps would wear out after a relatively short number of cycles. This is not an issue with modern fluorescent fixtures.

The only problem with using a CFL for this application is that most take 30 seconds or so to reach full brightness. However, there are several types of CFLs now that reach full brightness almost instantly. Check Consumer Reports, they have a good table.