Basically that - various sources say that the royalties charged to Spotify per play are on the order of 1p, which very rapidly becomes the dominant cost. If you have a million listeners online, it's about a hundred grand an hour in royalties. That adds up quickly. Other sources (radio, both classic and internet-based) which don't offer on-demand music don't have to pay on a per-listener basis, and so can actually benefit from economies of scale.

That said, based on Spotify's revenues which they have publicly stated along with subscriber numbers, it looks like they made about 50p in advertising per "free" user over the whole of 2009. While some argue they're paying for their experience with ads, 50p per annum for unlimited streaming music cannot look like a good business proposition to the music industry, and it's hard to say they're being unreasonably greedy by refusing to drop their royalty demands down to that level.

"DNA molecules" can have radically varying lengths - a single base pair on a strand typically occupies about 3 Angstroms, but the number of base pairs in a given molecule can range from a handful to many million. Based on the quoted size (about 250 Angstroms, assuming a 100 um human hair), the number of base pairs here is probably a few hundred to a thousand, depending on how ring-like and well packed these structures actually are.

One human cell's worth of DNA is about 3 billion base pairs, which times 3 Angstroms is about 1m, which is probably where the "meter long" value you heard came from.

And decays to technetium-99, so almost all the initial technetium-99m from the fission reaction while the reactor was active has almost certainly become plain old technetium-99 by now.

What is your reasoning for the need to entomb the site some time soon? There's no evidence of ongoing meltdown, so the heat generated by the reactors is constantly dropping, so the challenges of cooling and stabilising the reactors is getting easier, not harder. And while there's tremendous damage to the surrounding buildings, there's no significant damage to the reactors - there were reports of spikes which may have been consistent with radioactive steam venting through small cracks, but given the radiation levels at the site and the fact that they still need to periodically vent gas from the reactors, it suggests they're still relatively close to air-tight.

It's not under control in a "we can all go home and put our feet up" fashion, but there's no evidence that more drastic measures are needed any time soon - what they're doing appears to be working, and if they keep at it the reactors will likely be able to be brought down relatively smoothly to a stable state.

Exactly what the cleanup after that is like does depend on how bad the inside of the reactors themselves are - I admit, my clean-up and refuel analysis is a bit too gung-ho, but I would be surprised if the chosen path wasn't to take it through as much of a normal recovery and disposal process as possible, as it's dramatically easier and cheaper, and may well be safer providing the damage isn't much worse than it appears from the outside. Most of the radiation release thus far is transient (Iodine and radioactive seawater byproducts) and the levels of caesium scattered around the site should hopefully leave the radiation levels below the threshold required for special cleanup operations outside the reactors. Thus from a radioactivity point of view, decomissioning will be (comparatively) straightforward - the massive structural damage is obviously a significant issue, but that's probably as true anywhere else along that coast of Japan at the moment.

Most countries have at least some high background areas, and this is usually down to the composition of rocks in the area, where regions which contain a lot of heavy radio-isotopes (uranium, thorium) have high levels of Radon gas, which seeps up to the surface and leads to increased radiation exposure and correspondingly higher natural doses. If there are other factors which act to enhance the transport of radiation to the surface, it can become even more striking - Ramsar, in Iran, is the most dramatic example, where the local hot springs cycle up large amounts of dissolved radioactive material, leading to background doses of tens of mSv per year or more.

Interestingly, there appears to be little evidence of a significant increase in cancer rates in such regions, and perhaps even the suggestion of a slight adaptive response.

All of these points are, I believe, at least hyperbole, and at worst outright scaremongering.

While it's true several plant workers have been taken to hospital for monitoring after receiving acute doses higher than safety recommendations (>100 mSv), this is many times lower than the typical onset of "radiation sickness". The safety threshold is chosen as the limit of detectability for increased cancer risk over a lifetime, which puts it on the order of 1 or 2 percent increase in lifetime risk of cancer. Given they're doing very valuable work, this is not that dramatic a risk - the risk to other emergency responders in the wake of the tsunami is probably much greater.

With regards to the "high" doses 40 km away, these need to again be put in perspective - it is "high" compared to the local background (although often only 50 to 100% more than usual, barring localised spikes), but there are places in the world where natural radiation is almost 100 times greater than the typically quoted "background dose", and people live there just fine. Combined with the fact that most of this radiation is short-lived Iodine isotopes, a ballpark estimate suggests that people living outside the plant would only see a dose of 1 mSv or less by the time the iodine had decayed away, even if they ignored all the simple safety precautions which can be taken to reduce that further. These doses are well known not to cause any significant increase in cancer risk - long term or not.

And your suggestion of a Chernobyl-style sarcophagus being required is still rather unlikely. Since it appears none of the reactors have actually melted down or suffered a substantial failure in containment in the immediate vicinity of the rods themselves, it's quite likely that they'll be able to take them through a more or less normal shutdown and decommissioning once proper cooling is restored, and the storage implications will be no more serious than if they reached their natural end-of-life. Indeed, if they weren't already near or past their expected end-of-life, they could probably be fairly readily repaired, refuelled, and set running again within a relatively short timeframe. (Indeed, there's talk that this is being considered for Reactors 4 through 6, although that may turn out to not be politically viable).

I'm not denying it's a serious issue - but in the perspective of tens of thousands dead, and many times more homeless and short on food and other supplies, it really shouldn't be dominating headlines in this way.

Not at all. One of the issues with this change is it is largely driven by the English outlook - while it's understandable in terms of population distribution, it means that regions which are further north and west suffer quite seriously in terms of daylight hours. For example, sunrise in N. Ireland is typically about 40 to 45 minutes later than London in winter. If they go through with the most extreme change being discussed (that is, moving to match CET year-round) parts of N. Ireland and Scotland won't see sunrise until well after 9am in winter, which is going to be very unpleasant for a lot of people.

There is no doubt that such lasers are possible, but they are not efficient or easy to build - the mirror system in the example referenced in Wikipedia is 40 m high, not a trivial engineering feat.

And, bearing that size in mind, a quick back of the envelope suggests that 1 MW is the input power of the light, not the delivered power of the laser. A quick search doesn't turn up any papers or detailed articles relating to this solar tower specifically, but other examples of such solar-pumped NdYAG lasers suggest a conversion efficiency of about 10 W laser power/m^2 of mirror, or about 1% of the incident radiation [1].

So, assuming that lasing efficiency for this system, this is not a 1 MW CW laser, but a 10 kW CW laser pumped with over a megawatt of input power, which necessitates significant cooling to keep the thing from melting. Compared to traditional laser designs, this is still not that impressive, especially given the effort involved in its manufacture.

[1] A solar-pumped Nd:YAG laser in the high collection efficiency regime

I can provide the following insight: Lasers do not work that way.

More specifically, there are two issues with your suggestion. Firstly, lasers are not power-limited by input light, but rather by the design of the lasing cavity and how efficiently it stimulates further emission. Many types do need a decent kick to get them going, but beyond that a bright source offers little or no benefit.

Secondly, even if more input light was useful, this mirror doesn't actually provide that much power. It's just the use of the parabolic reflector to concentrate the energy into a small energy that makes it look impressive. Looking at the dish, it's a few square metres in area, at most. That's only a few kW of light in total, of which only a tiny portion is at any one wavelength which would be useful for pumping a laser. An appropriate pump laser or even a decent flashlamp would be vastly better than this for stimulating laser emission.

Also, LASER. Light Amplification by Stimulated Emission of Radiation.

Copyright in a journal never extends to preventing others using the work - only to using the particular presentation & formatting used in the journal. You can read it, go "huh, that's interesting" and use the technique to your heart's content, on any of the licenses NPG offer, even their classic, non-CC versions. Any suggestion otherwise is pretty clearly FUD. Of course, your rights to reuse techniques are potentially covered by patents and the like, if the authors have sought them before publishing the paper, and this license does not guarantee that that won't be the case, but nor do the other, "more open" CC options.

I once heard an observation about the ME series that makes a lot of sense: ME1 is following the template of a movie; ME2 is following the template of a TV series with a large ensemble cast, like Star Trek.

After our hero is introduced and the scene set, it's then broken into "episodes" which are heavily focused on one member of the "cast", who the rest of the time just stay in the background and throw in the occasional quip. Every now and again throw in a plot advancing episode to keep things ticking over, and finish with beating on a Big Bad. But be sure to wrap up with a bit of a cliffhanger to ensure people are hyped for the next sesaon.

The actual plot of any given episode, most of the time, is immaterial - any events which happen in a character episode are expected to be contained within that episode, and exist only to frame character development or provide obstacles for them to overcome. Since most games follow the movie template, it does feel very different to play, but not necessarily worse - the focus on characterisation did pay off, I feel. Still not perfect, but then nor is the characterisation in most good TV series either.

Sadly, having said all that, I do agree that it wouldn't work as well as a movie, which does make me concerned about the quality of any adapation, since it's going to have to stray pretty far from the plot to fit it into a movie-shaped box.

The test actually reports uncertainties on runtimes - when I ran it, it was on the order of ~0.5%, so 28.6 seconds is all the accuracy you can meaningfully specify. Those last 3 digits are effectively meaningless - accuracy is a function of the test, not how many digits you can make it spit out at the end.

With regards to this data being relevant to cancers from "yesteryear" - to be blunt, they don't care. The purpose of these studies is to determine whether current mobile phone usage poses a risk to the population. If someone developed a cancer from using a phone which was made to poorer standards a decade ago, that's a shame, but there is no particular reason you should expect this research to be relevant to them, and moreover what good would it do? Unless they were anxious to try and throw around lawsuits, there's no benefit to working out the risk factor they were exposed to.

And as for why this study has taken so long to do - you don't launch a study costing many millions of pounds and spanning decades as a first step in research (particularly in a field with relatively sketchy underlying hypotheses). You start with smaller, retrospective, studies which allow for large effects to be readily detected, at a fraction of the cost. The problem with mobile phones is that there is no evidence for the type of large-scale, acute effect which can be readily quantified by such small projects, so a larger project (like this one) is required to look for smaller-scale effects (which may still be significant on the level of the population).

And the problem with a big project is actually managing to get enough stats for sufficient predictive power - in the early days of mobile phone usage there simply weren't enough people regularly using mobile phones to make meaningful predictions about the effects on the level of a population. Indeed, it notes that even five years ago a study of this kind had to be halted because of a lack of participation.

Berating scientists for wanting to perform good-quality studies is not very productive. The demand for scientists to produce dramatic information very quickly tends to lead to lead to misleading results being presented, and statements of that kind (see: foods which cure/cause cancer every other week) is one of the reasons many people are losing faith in science.

It contains many internet-unfriendly provisions. The two biggies though:

A "Three-strikes"-equivalent law for filesharers is still in (without trial), and the text of the bill can be ready to imply that it will be ISPs responsible for ensuring that their networks aren't used to infringe copyright - effectively mandating monitoring of all internet traffic at a much greater level than is currently done.

There is also a provision which allows the Secretary of State (with the approval of a court, to allow a tiny bit of balance) to require the blocking of any websites which are involved in the infringement of copyright. Or, indeed, may be used in infringement of copyright. When asked if this would include sites such as Wikileaks, ministers said yes, it would, as the material they leak is copyrighted by its original owners. They were quick to point out that they wouldn't use it to infringe on freedom of speech though.

Yep, I feel real safe with that reassurance.

Other aspects of the bill are actually reasonable, there's just a handful of provisions that are really quite shockingly draconian.

This article seems to have popped up because Adobe have indeed released an updated version of the flash player on the 11th of this month. Still alpha, but slightly newer. Pleasingly, it seems to have fixed the only persistent bug I had with the player (which caused Firefox to report a crash every time it was closed - no actual errant behaviour, however).

Why exactly the submitter picked at year-and-a-bit old article as a reference for this news is still a mystery, however.