Exactly. Julia will eat R for lunch soon enough, I think. It's an elegant, well designed and efficient language. It's only been around for a couple of years, and has a very vibrant and rapidly growing community.

Check it out for yourself: The Julia Language Homepage. It's got a lot to offer anyone with an interest in mathematics, including statisticians. It's based on the LLVM, and interfaces trivially with C libraries - plus it's a very fast language in it's own right, unlike R or Python.

Liquid fluoride thorium reactors (LFTRs) are better in many regards:

• - LFTRs are provably meltdown-proof.
• - LFTRs run at lower pressure, so are much safer than boiling water designs.
• - LFTRs don't produce explosive hydrogen, which caused the Fukushima explosions.
• - LFTRs produce much smaller volumes of waste than current U/Pu reactors for the same power production.
• - LFTR waste is dangerous for only a few hundred years, instead of over 10,000 years.
• - LFTR installations take much less space than current U/Pu reactor installations.
• - LFTRs don't require water cooling, unlike current U/Pu reactors - they can be sited inland and away from rivers.
• - LFTRs operate at higher temperatures, making power generation more efficient, and provide waste heat for desalination if desired.
• - There's no meaningful weapons proliferation risk with LFTRs, so they can be exported to growing third-world economies.

Education is a good thing. If people actually understood the risks and mortality figures from fossil fuel production, they'd see even conventional nuclear is a no-brainer. We've been operating around 100 nuclear plants in the US for decades with no significant issues. France has also had a great track record with nuclear. On the other hand, hundreds of thousands of people a year die from coal electricity generation alone. The total toll from all nuclear reactor problems ever doesn't even come close to one year's worth of coal power.

There's also mercury pollution, and ocean acidification to consider...

Then there's the "climate change threat"...anyone who's serious about lowering CO2 output will have to accept a role for nuclear, as there's absolutely no meaningful way to lower CO2 production without it.

Regardless of public opinion, I'm confident the grownups will make the right decision sooner or later.

Oh, but it is. It's the only realistic means available today to replace coal-fired electricity generation worldwide with a CO2-free alternative.

Regardless of AGW, replacing coal-fired electric plants with a clean alternative is a good idea.

Whoops, I hit submit before I was ready...

That should read "...LFTR (Liquid Flouride Thorium Reactor)...".

You would think. Clearly, though, this is just a hit piece on thorium, even though it has nothing to do with modern thorium reactor designs.

Thorium is well suited to molten salt reactor designs, and in fact is best used in liquid form. These LFTR (Liquid designs will fission 90%+ of the fuel, instead of the 0.5% fissioned by conventional reactors. This means a lot less waste for the amount of energy produced. Also, the waste from such reactors is dangerous for much less time than that from conventional reactors.

Thorium reactors are being developed by Russia and China. In the US, Flibe Energy is working on LFTR designs. There's lots of interesting information in their site.

Thorium power should most definitely be developed. It's a clean, safe source of baseline power - and doesn't take the vast space required for (inconstant) solar and wind. Plus, eventually it will be great for space applications.

Keys were stolen because of a software bug, free software or not. The fact is that the overall state of software "engineering", is poor. I'm quite sure there are plenty of similar issues with closed-source software, although there is the under appreciated benefit of "security through obscurity".

The point holds, though, that open source software should generally be more bug-free than closed source. What we need are more motivated people (and better tools) to search for vulnerabilities. It's much better when white hats find them than black hats.

True, but it might well fill the same niche and be quite a bit better...

Wrong tense. D "is attempting" to fill that niche, and so far has failed to gain (much) traction.

D is actually picking up some momentum - it's now being used fairly heavily at Twitter for instance. It's still a quite young language, only invented in 2004.

That said, D does have it's challenges, particularly in that only two people decide what actually goes in the language. They're smart, but like anyone they have their blind spots. It also has reneged on some of the early promises, in particular "optional garbage collection". There is some ongoing work to fix that oversight, though.

Oh, don't worry, there are plenty of folks constantly complaining about D (check the forums at dlang.org). However, one thing almost all of them have in common is a desire for something better than C++, and many find that in D despite its various warts.

It's worth keeping an eye on, for sure. Walter Bright (the original inventor) has been writing articles at drdobbs.com, they're worth checking out.

I like Python as well, and pypy is looking pretty good as a performance enhancement. Multithreading is is still a problem though.

A couple of other newish languages worth noting are Scala and Julia.

So far Scala is only on the JVM, but it's a powerful, clean language and a huge improvement over Java. It includes functional language features along with OO.

Julia has the potential to be general purpose, but thus far is oriented towards "technical computing" (the same niche as Matlab and R). It's LLVM based and offers easy interoperability with C libraries.

Again, clearly the urgency is not there with regard to AGW. Senor Reid is an alarmist Dem, yet he couldn't do the right thing with regard to Yucca Mountain (or most things lately, for that matter).

If gridlock continues on underground storage, it might be time to examine dropping suitably packaged waste at the mid-Pacific subduction zone. At least there's no NIMBY problem there...

Too many people are terrible at assessing risk. Fossil fuel energy production kills hundreds of thousands of people a year...not counting tertiary effects. How many have died from Fukushima radiation again?

The waste issue is easily solvable with thorium power, as there is much less of it and it only needs to be stored for a few hundred years. A Yucca Mountain type site would be fine, or there are alternatives.

Weapon proliferation is not a concern for a US power buildout. I believe it would not be for thorium reactors in most situations.

Siting them underground would mitigate almost all "disaster scenarios".

Simply put, the grownups are going to have to take charge. If AGW is to be taken seriously (and in fact, even if not) there is no realistic alternative to nuclear going forward.

I believe the prudent thing to do is replace as much coal-fired electricity generation with modern nuclear generation as possible. All of it would be fine. Coal is a very dirty source of energy. More base load power generation is needed even if solar and wind are added - those aren't reliable sources of energy. As electric vehicle technology improves, even gasoline usage could be replaced with clean power.

Thorium based MSR technology should be a national priority. If pursued, it should actually lead to cheaper power over time. It has numerous advantages over U/Pu based technology. For starters:

• - Meltdown proof.
• - Radioactive waste produced is much less, and becomes harmless much faster.
• - More abundant, less expensive fuel.
• - No water cooling required.
• - Much smaller installation for the same power production.

TigerDirect, Overstock.com, and Gyft.

Be surprised. :-)

I think BTC will do well in the long run, simply because of low transaction cost/friction. Not to mention the small size and light weight. ;-)

You should take a look at the Mining Dasboard (it's a great resource in general, with the best mining profit calculator I've seen). You'll note that the total compute thrown at mining continues to grow, and just hit 35 petahashes per second (that is, 35,000,000,000,000,000 hashes per second). Clearly a lot of folk think there's value in mining bitcoins, and the increasing rate shows no sign of slowing down. On January 1, the rate was only 11 PH/s, so it's more than tripled in 3.5 months. Why are they doing it when for almost all of them it's not profitable now? Clearly they think the value of BTC will rise - a lot!

You're conflating two different things. A "coin" is one bitcoin. Each coin may be subdivided into 100,000,000 "satoshis" currently. The Bitcoin Foundation could allow further subdivision in the future. So, the supply of BTC is not of any concern in the long run.

True, it's more like having gold destroyed (say in an explosion) than burning paper money. BTC was designed to have a limited amount, to prevent inflation. Given its divisible nature, that's not really a problem, although it will tend to drive up the value of integral BTC over time (thus the allure to investors).

As mentioned above, a "satoshi".

What people? Certainly not investors.

Why is that a problem? Gold is currently worth four times (400%) what is was just fourteen years ago, and that's after taking a big hit a while back. Does that mean nobody wants to buy and hold gold now? When they need to buy something, they exchange some gold for dollars (or bitcoins) and make a transaction. BTC has nice attributes of both commodities and currencies.

There are a couple of factors that affect BTC price. The first, and the main cause of extreme volatility, is its newness and novelty. There have been giant "buyins", as when the Chinese became aware of it. Those things drive prices up. In the longer term, I expect it will stabilize quite a bit...but of course it will be worth more and more dollars over time if for no other reason than dollars are constantly inflating by design.

I expect BTC will do fine in the long run as long as the protocol remains secure.

True, but when the "authority" is a Nobel Prize winning physicist specializing in the exact area being discussed it means something. Further, "appeals to authority" invite the reader to investigate the claim, including detailed arguments made by the authority.

So, while not conclusive, an "appeal to authority" in this case is of interest.

WTF are you talking about exactly? Be specific...

They don't "accept" them now, and banks don't have control over external transactions. Thanks for playing! ;-)

Almost no modern currencies are "backed" by anything. Go buy some gold or silver and have fun! They're not nearly as convenient for Internet transactions as BTC though...