As a one-time member of The Green Grid Technical Committee, let me summarize and correct a few points:

• The EPA has said that data centers use around 4% of US power. The federal government uses going on half of that, IIRC.
• Historically PUEs of 2.4 were common. It makes sense. In a closed building, it takes 1 unit of cooling energy to cool the heat produced by one unit of computation etc. The cooling systems were say 80% efficient so that makes 1.25 units of cooling energy. Add in power conditioning and UPS losses, and you easily get 1.4 to 1.6 units of non-computational work for every unit of computation, for a PUE of 2.4 to 2.6.
• A PUE of less than 1.0 is by definition impossible. Nevertheless, a few deluded individuals have reported them.
• In anything less than ideal conditions, getting the PUE below 1.2 is very difficult. A common mistake is to allow equipment fans to do some of the work of the heat exhaust fans. This effectively transfers facility load to the equipment and results in artificially low numbers.
• Low PUEs are harder to achieve in high-resiliency conditions. The large server farms used by MS and Google do not require the level of availability that enterprise data centers commonly do. These companies have the luxury of trading off equipment failure rates with power costs.
• Uninformed data center managers sometimes think they can use unconditioned outdoor air as a coolant. This is ill advised. High humidity levels, high sulfur and other contaminant levels, and particulates can cause premature equipment failure, not to mention voiding warranties. I visited a data center in a hot and humid location once that had had 40% disk drive failure in a year. It's best to consult a professional data center HVAC specialist with experience in low-PUE installations.
• Power costs do sometimes exceed CAPEX over the life of the equipment. It depends on location and up-front equipment cost.
• Power gotten from public hydro is not "green". Power in the public grid is a zero-sum game. Only renewable energy that you produce yourself is more green than the average greenness of the public utility system.
• The data centers located near big dams don't get cheap power because they're being green. They get it cheap because they avoid grid distribution costs. The same thing can be accomplished by colocation with a big coal plant. It just doesn't sound as cool.

Renewable power bought from a utility company is a zero-sum game--only one party gets to use it, and everyone else gets stuck with whatever's left. So until they are actually generating all that power themselves, the claim is just chest thumping. No real benefit to the environment.

This was my first question also. The OP does not say that they develop any sort of WiFi equipment. Assuming they don't, the troll might as well be demanding royalties based on the use of a navigation system in a Buick.

I had a similar thought. Granted that we share the same molecular machinery, but it seems to me that we probably evolved from the hosts the bacteria fed on, not from the bacteria themselves. Unless they ate rock, in which case we evolved from the hosts their mutations ("descendants") fed on. For me, the jump from asexual to sexual reproduction is the really interesting inflection point. Do we really believe that an ordinary mutation caused that? (IANAE).

The winter/spring cross-quarter day is the halfway point between the winter soltice and spring equinox. Apart from any astrological or other significance it might have, it is a good marker for when the days start getting longer FAST.

A couple examples from India, and every 35+ programmer on the planet is doomed?

We don't need to throw out developers with age and experience, we need to throw out clueless journalists.

------

There are two kinds of innovation. One blazes the trail. The other paves the road.

A couple of my academic friends believe that MOOCs like Khan Academy will "invert" the teaching process. Instead of attending lecturers for content, and assimilating the material later while doing homework, students will view lectures offline, and assimilate the material in class, in a more lab-like environment.

Sounds good to me.

> Read below for the rest of what CowboyNeal has to say

No thanks.

I appear to be the only ./er who cares enough to mention that it should be "hear, hear", not "here here".

That said, I agree with the sentiment, if not the spelling. My former employer, who shall go unnamed, is a dynamic Silicon Valley hi-tech company, once one of the best places in America (pre-Google) for bright geeks to work. It is now a sea of H1Bs. A small percentage of them are stars; the rest are merely inexpensive. And the worst part is that H1B managers rarely hire Americans; they prefer to manage other H1Bs.

Unbelievably bad, that is. The light is poor and barren. I have yet to see a "100w equivalent" that was even close to being as bright as a 100w incandescent. Some of them have a power factor of 0.5, which means they're actually half as "energy efficient" as the label says. And "long-lasting"? Not in my experience. But hey, at least they're expensive.

The lighting industry has got to be gleefully rubbing its hands over these regulatory moves.

The building inspector made me replace 160 watts of very nice halogens in my new kitchen with 160 watts of fluorescents because the code says half of the lighting in a kitchen has to be "energy efficient". The overall lighting level went down considerably with this change, in part because the halogens give directed light and decent looking fluorescents don't, and also because halogen light is a lot nicer. Of course the change was reversed the same day the inspector signed off. The $120 fluorescent fixture I was forced to buy now illuminates an area of my home that I don't spend much time in--the laundry room.

I favor a simple progressive tax on capital gains. Tax ordinary capital gains same as now. Tax gains on trades under a day at 50%, under an hour at 75%, under a minute at 85% and under a second at 95%. Preferably on a nice curve instead of by quanta.

Wow, I take it back. Some of this music is pretty awful. A good MIDI performance expert could do better. Oh god, listen to the french horn in the 2nd promenade in Pictures at an Exhibition. Oh, and the violins at the end of that. Shiver. Ouch. Damn.

Possibly the project's initial goals were completed, but that's hardly what springs to mind when one hears the phrase in the context of the classical repertoire.

That said, I'm listening to Eroica, and it actually ain't bad.

WRT the print edition quality, most world-class musicians prefer autograph scores. Heavily edited scores are more suited for amateur performers. An exception is Sussmayer's version of Mozart's Requiem, which has a lot of rough spots, and is usually performed from later fixed up versions.

Received mine (in CS) at 45, 15 years ago. I'm honestly unsure that the PhD was worth it, though it was fun. The Masters was definitely worth it.

The PhD training really only teaches you how to think. Since entering the hi-tech workforce, I've had to learn 3 or 4 new skill sets in order to do a job and have a domain to think about.