2012-12-31

2012 predictions evaluated

Looking back at my predictions for 2012, how did I do?

Eurozone: Eurozone governments engage in a sequence of progressively more desperate kicking-the-can-down-the-road exercises. A replacement source of funding fails to appear. The tension between the Germans resisting inflation and the rest of the Eurozone demanding economic relief. The ECB is inexorably pushed towards turning on the printing presses. Greece, Ireland and Portugal turn on the screws demanding more help with the threat of default. French and German banks turn out to be shockingly undercapitalised, to the surprise of no-one who was paying any attention.
6/10: there's certainly been nothing in the way of solutions here, and Greece has been leading the screw-turning.
North Korea: Kim Jong Un has an attack of common sense that may or may not result from being hung from a lamp post by a length of rope. North Korea opens the shambles of its nuclear enrichment program to international inspection in exchange for desperately needed aid. The humanitarian crisis turns out to be even worse than expected, with deaths of tens of thousands from cold and famine before the West and South Korea can organise aid shipments. China is less than helpful.
1/10: talk about hopeless optimism. China, at least, has been less than helpful.
UK economy: Growth peters out to practically nothing, perhaps dipping in and out of negative territory. Huhne gets squeezed by popular pressure resulting from ever-rising energy bills as the Conservatives keep him in the firing line. More effort is finally made on new gas plants, probably some more test drills for shale gas, and the planning permission and local challenges for nuclear plant additions grind on. Inflation stays above the 2% target as groceries in general and goods from China in particular rise in price.
9/10: UK economy barely grown over 2 years, inflation stayed above 2%. Grumbling about energy bills but no action yet, Huhne is still around. We are dashing for gas and building new plant.
UK politics: Con-Lib coalition effectively falls apart on several issues (e.g. energy). Labour fails to capitalise on this. Grumbling in the Labour party about Miliband and some early manoeuvering by potential challengers.
5/10: Coalition having problems, but gay marriage appears to be one of the key issues. Not much grumbling about Miliband, perhaps they've forgotten he exists.
Olympics: Substantially poorer showing for the UK than 2008, except in sailing and cycling. Boris makes at least four major gaffes during the Games, making him the only real entertainment. Fewer visitors than expected results in a significant financial loss for the UK.
2/10: Glad to be proven wrong in most of this. We still ate a pretty solid financial loss though.
USA: SOPA passes albeit in a modified and mostly annoying rather than harmful form. Congress and the Senate continue to be bought and sold. Obama starts feeling the pressure from within the Democratic party but just edges the election against a Romney/Bachmann ticket.
6/10: SOPA died, unexpectedly but thankfully. Plenty of buying and selling in politics persisted. Obama had an easier ride than I expected, and Romney's running mate was Ryan rather than Bachman.
China: A slow-motion implosion, rising popular anger at financial losses mostly held in check by increasingly brutal actions from the PLA. China makes an increasing effort to diversify out of US Treasury holdings but is stymied by lack of a reasonable alternative given events in Europe.
4/10: financial problems are clearly bubbling under the lid, but the PLA and Party are keeping the lid on; their continuing actions to tighten Internet access show what they're really worried about. Looks like Africa is one of the areas China is trying to expand into.
Middle East: Iran continues to posture, Iraq's new government breaks apart and reforms a couple of times. Afghanistan is still a mess, Pakistan becomes an even more dangerous snake pit.
7/10: generally nailed, apart from Iraq government breakage. If anything, I understated the problems in Pakistan.
Climate: 2012 weather proves to be a combination of too hot, too cold, too windy, too wet and too dry. Much like 2011.
7/10: drought up to April, unseasonably warm March, then a deluge for the next 8 months.
Random: Britney's engagement doesn't last 2012. It may barely survive 2011.
3/10: Despite persistent rumours about a breakup it looks like Britney and Jason will end 2012 together. Whodathunk?

50/100 overall, slightly better than the UK Met Office. Wonder what 2013 will bring?

Conspiracy courtesy of the GRU

Too entertaining not to share, mostly because it's just within the boundary of what's plausible: why we haven't seen much of US Foreign Secretary Hillary Clinton recently.

Within minutes of leaving Bahrain airspace, this report says, the C-12 Huron carrying Secretary Clinton and her US Navy Seal [sic] protectors, "without notice," deviated from their assigned flight path heading, instead, directly towards Iran's Ahwaz International Airport where, coincidentally, Iranian President Mahmoud Ahmadinejad had previously landed on an "unscheduled" visit.
[...]
Upon the C-12 Huron landing at Ahwaz, however, this report says it encountered "extreme turbulence" causing it to leave the runway where its main landing gear then collapsed causing it to crash.
The article quotes the Kremlin's GRU (Glavnoye Razvedyvatel'noye Upravleniye a.k.a. Foreign Military Intelligence) as the source. Obviously there's no possible motive for them to exaggerate...

At least some of the facts are verifiable: Clinton has indeed been out of the public eye for several weeks, and the cause is quoted as concussion after fainting at home, some time in the week leading up to December 16th. The SEAL commander suicide is quoted as happening on Saturday December 22nd though, which is a rather big time gap. The Army alone has 112 C-12 Huron craft so concealing the loss of one is at least plausible.

The main points that jar, though: why would a SEAL unit commander be on diplomatic protection duty, even for Hillary Clinton? Protection work is a young man's game, and Commander Price was 42. Why indeed would you need a SEAL unit? If you're deliberately landing an aircraft in Iran at a commercial airport and the Iranian military are expecting you, then if things go wrong you are already so deep in the yoghurt that even SEALs aren't going to help you. And what would Clinton hope to achieve with a covert meeting that would not be possible with an overt meeting? It's the job of a Foreign Secretary to go around the world and meet dubious people; no-one would have batted an eyelid if the existence of the meeting was public.

If Clinton was "bleeding profusely" after the crash then transporting her out of Iran and back to the US would potentially be dicey - she's 65 years old, and one's response to trauma at that age is less than elastic. However, if they managed to get whole blood in her and exclude the possibility of significant head trauma then a medevac would just about be plausible. I wouldn't have liked to take the risk of moving her far from that crash site though.

Overall this is a great example of conspiracy theories: just on the edge of plausible, some facts lining up but others forming something of a ragged edge, originating from a source with ample reason to foment trouble, and failing to answer the basic question of why all this would be so secret in the first place.

The innocent have nothing to fear

Eric Falkenstein addresses the current flap over TV presenter David Gregory who brandished an empty high-capacity gun magazine while on the news show "Meet the Press". There is now frenzied debate about whether Gregory should be prosecuted, since he did the brandishing in D.C. where possession of such items is illegal. Gregory and his friends are contesting that, since he is clearly not a criminal and had no criminal intent in possessing the magazine, police action is not warranted. Sadly the law seems to have no such loophole, and "do you know who I am?" is rarely a successful strategy in a legal defence.

Falkenstein points out that this is a classical consequence of the disease of over-regulation:

There are so many laws regulating your average business that at any time one is probably being broken. This puts everyone at the mercy of their regulator's goodwill, because like David Gregory you will probably get off if the right people are on your side. If they are indifferent you are at the mercy of the mob, government, or wealthy antagonist.
Some poor schmuck in D.C. who finds an empty 30-round magazine in the trash, walks home with it to show his mother, gets pulled over by the cops and searched will find himself at a substantial legal disadvantage to Gregory's situation, despite being (if anything) less culpable.

This is why any MP, Congressman or Senator who tries to promulgate an over-reaching law with the assurance "it's OK, we would never use it for inappropriate prosecutions" should be a) ignored and b) hung from the nearest lamp post. Laws are always used inappropriately. Those drafting the law only care that the law is sufficiently acceptable to 51% of voting parliamentary representatives, and that all sponsors have what they want in the law. Once the law is on the books, a thriving business around negotiating the new law will spring up; as Falkenstein notes, the regulator and/or government will normally be the gatekeeper of the law and therefore able to gain power and favours by exercising its discretion around that law.

2012-12-29

Bribery as the oil for the wheels of government

In this case, the US Congress:

Without the persuasive powers of the political 'carrot' [earmarks], congressional leaders and the President no longer have the 'stick' required to move Congress to get anything of significance accomplished.
The moratorium on earmarks went into existence in February 2011. Since that time we have seen some of the greatest legislative fails in the history of the nation, highlighted by the debt ceiling fiasco of 2011, the inability to pass a jobs bill, an ever-increasing vacancy rate in the federal judiciary as one nominee after another is shelved and, of course, the current fiscal cliff clunker that might be the most embarrassing and damaging display of congressional incompetence of all.
Author Rick Ungar, who is by no means a hard-core right-wing minimum-government type, makes a persuasive case that in order to get Congress to do anything, you've got to let them bribe each other with earmarks (spending items attached to major bills which bring federal dollars to particular members' districts). He estimates earmarks costing $16bn annually, which is not chump change to be sure, but from his point of view it's very cheap; a fraction of a percent of the federal budget, allowing the business of government to proceed as normal. To be clear, he regards this as an awful state of affairs, but maybe allowing Congress to bribe itself is the least evil way forward.

Of course, the real hard-core right-wing minimum-government types - the Tea Party, who have 61 congressional representatives in their caucus - might suggest that if Congress can't pass laws without bribing each other, then maybe those laws aren't of sufficient benefit to the nation. Government has paralysed itself.

2012-12-28

Sir Hector Sants, CBE

Dear Hector from the FSA has got his knighthood:

Hector Sants has been an outstanding leader of the Financial Services Authority during the most challenging of times. Following the failure of Northern Rock, he launched a fundamental transformation of the FSA's approach to prudential supervision and enforcement, with a far more robust use of civil and criminal powers to achieve credible deterrence. He is also a former member of the Board of the Nuffield Orthopaedic Centre, Chair of the Board of the Said Business School, University of Oxford and a benefactor of a number of Oxford based charities including the Art Room, a charity providing art therapy to children with behavioural and learning difficulties.
Whatever happened to "no rewards for failure"? "Credible deterrence", don't make me laugh.

2012-12-27

Why I have problems with the death penalty

While I support the theory of the death penalty - that there are certain people who cannot be tolerated in a free society, and certain crimes so heinous that death is an appropriate response - my considerable unease about the practical implementation of the death penalty is perfectly expressed by this case: the jailing of Texan Michael Morton for the murder of his wife where, after 25 years in jail, his legal team finally managed to discover proof that he was innocent:

So it was a sudden reversal of fortune for [prosecutor] Anderson when, eight days after Michael's release, the CCA overturned Michael's conviction on grounds of actual innocence. The ruling meant that Anderson had secured a guilty verdict against an indisputably innocent man. Yet whether he, or anyone else involved in the case, would ever be held accountable for the wrongful conviction remained an open question.
It seems that, in their zeal to prosecute Morton, the prosecution team had failed to disclose key evidence that would have pointed Morton's defence team at a lead which contradicted the claim that Morton was the killer. Now the prosecution's job is to present the state's case in the best possible light, but that doesn't (or at least shouldn't) extend to hiding exculpatory evidence from the defence.

Morton was lucky not to get the death penalty for the crime of which he was falsely accused. If he has (finally!) been exonerated, due to a tireless defence team, how many more people sitting on Death Row (or buried in an grave) have been victims of the same practices but less fortunate?

2012-12-26

BBC repeats myths about the fiscal cliff

Lordy, this kind of shoddy journalism about the fiscal cliff really irritates me:

What is the fiscal cliff?
On 1 January 2013, tax increases and huge spending cuts are due to come into force - the so-called fiscal cliff
"Huge" spending cuts? Let's remember: of the $600bn comprising the fiscal cliff, only $110bn is spending cuts. This is a little under 3% of the $3.8tn 2012 federal budget. These are not "huge" spending cuts. The $490bn of tax rises is a lot more "huge", but for some reason is not described as such in the article.

The BBC journalist who wrote this crap (Zoe Conway is implied, though not definitively) should be beaten with a stick until they show evidence of a) being able to understand relative sizes of big numbers and b) no longer wilfully dissembling to advance their personal prejudices on taxation and spending.

Tax, made real

Dick Puddlecote has the right idea on how to introduce his kids to the concept of taxation:

I'm not sure exactly when the practice started, but we've taken to pinching small portions of the little uns' bits and bobs every now and then. For example, a few chips from a Maccy D's meal, a spoonful of their after dinner dessert, a couple of sour sweets from the pick'n'mix. You know, that sort of thing.
We call it 'tax'.
[...]
They each received a Galaxy selection box, amongst other things, yesterday morning and we - with tongues firmly in cheek - announced that the tax would be a Ripple bar from one and a bag of Minstrels from the other. It was heartening to find out later that the very sweets in question had, without our noticing, been spirited away into hiding places in their rooms.
Is this tax avoidance or tax evasion? Whichever, the parents have been deprived of illegitimately acquired cavities.

Benghazi: the price for failure

The price for failing to prevent the first killing of a US ambassador since 1988 is, apparently, a few weeks of administrative leave:

The highest-ranking official caught up in the scandal, Assistant Secretary of State Eric Boswell, has not "resigned" from government service, as officials said last week. He is just switching desks. And the other three are simply on administrative leave and are expected back.
The only person to have suffered an actual career setback was Ambassador Susan Rice, and even now there is diligent re-writing of her failure to be considered for the role of Secretary of State as a sexist and racist conspiracy:
The fact that she [Rice] understood—as do far too many women and women of color—that we are doomed once the stereotype "code words" start flying, is tragic. It's the oldest "old boy" trick in the game. The "boys," e.g., Senators John McCain and Lindsay Graham, came at her with code speak: "unqualified," "incompetent," "misleading," and "not prepared." And sadly they enlisted the help of other female senators such as Susan Collins and Kelly Ayotte. The deployment of other women to undercut a powerful woman is something we as women have all experienced, and we grimace each time it occurs.
Yes, clearly Republicans have a real downer on the idea of a woman of colour as Secretary of State. No doubt the surname "Rice" was also suspiciously Asian.

It used to be that senior politicians would take the fall for failures in their department, reasoning that their department's failure was their own. Lord Carrington is a classic example, taking the fall for failures of UK foreign policy that led to the Argentine invasion of the Falkland Islands in 1982. Clearly, "shame" is held to be an antiquated concept.

I realise that I'm banging on about Benghazi quite a bit, but I remain appalled that awful failures in the US military and State Department led to the death of an ambassador, information officer and two extremely ballsy embassy security personnel, where ample time was available to the military to intervene and save lives but someone decided that inaction was better than reaction and left US citizens to die. The attempt post-attack to cover up what happened and blame Sam Bacile's "The Mohammed movie" just made it worse.

2012-12-25

Inaction better than reaction

From the Streetwise Professor, why it is that most if not all of the post-Sandy-Hook proposals are in the wrong direction:

We live in a fallen world. Sometimes there’s just nothing you can do about that, and trying to do something is often the worst thing you can do. But that is something that people are loath to concede.
And commentator Charles is almost as good:
The goal for policy makers is not to prevent every mass shooting. That is an impossibility. The real goal, much as was the goal after the 9/11 hijackings, is to appear to be doing something to solve the problem and to engage in the theater of security.

2012-12-23

The value of protecting the supply chain

Malaria kills 600,000+ people in Africa each year, and has attracted attention from some very well-funded and determined organisations aiming to wipe out malaria. But where there's money, there's opportunity for fraud, and this area is no exception.

It seems that the modern effectiveness of anti-malaria medication such as artemisinin derivatives is somewhat undermined by large-scale imports of fake medication into Africa:

That even doctors are unable tell real malaria drugs from fake is testament to just how complex the situation has become in Tanzania and Uganda, which together accounted for 20m of the 94m malaria cases reported globally in 2010.
Estimates vary, but some recent studies suggest that as many as a third of malaria drugs in the two countries are fake or substandard, and most are believed to have originated in China.
Let's review what happens here. Someone operating a pharmaceutical factory in China spends time and effort duplicating packaging and form of existing medications, to the point where even doctors can't be sure which package is fake and which is real. They insert it into the supply chain, presumably mixing it with legitimate shipments to muddy the source, and a doctor in Tanzania or Uganda spends time and effort treating a malarial child with no effect for several weeks until they realise that the medication is no good.

There's no comeback to the counterfeiters, as far as I can see. Tracking down where the fake medications got inserted into the supply chain would be near-impossible; the poor state of documentation and widespread bribery in Africa means that you can't trust any documentation or verbal assurance. The only approach I can see is some form of tamper-proofed boxes from the manufacturers: say, a small electronic lock with a serial number that the doctor can SMS to the manufacturer, then the manufacturer can SMS back an unlock code - and the unlock is one-time. This will obviously cost money, both in development and manufacture but at least it reduces the 30%+ ineffective rate of medicine which is only going to go up. I can't see the Chinese government cracking down on these factories.

The claims are backed up by commentator Chinagirl88:

I live and work in China in the medical profession and despite as some people have pointed out, the lack of hard evidence, the suspicion that the fake drugs originate from China is unfortunately all too believable. Despite nearly every week bringing some story of "crackdowns" on gangs making counterfeit goods of every description (since I've lived in China there has been a crackdown on fake drugs nearly every few months) this seems to provide little deterrent to the people wanting to get rich quick no matter what the cost to other humans.

The next time someone tells you how disgusting capitalism is, remind them of this example of what state-backed cronyism combined with a poor legal infrastructure does to sick and poor people.

What's a contestable monopoly?

"No longer a monopoly", as it turns out. The estimable Tim "Mr. Scandium" Worstall gets to do the end-zone dance in The Register as his 2010 prediction about China losing its rare earths monopoly comes true:

In November 2010 I pointed out (here again) that two mines, Molycorp and Lynas, could alone provide a goodly portion of demand, replacing the Chinese production. We're in December 2012 as I write and as the first quotes show, those prices have been falling fast as the Chinese hold over supply is broken. Even to the point that the largest Chinese mine itself was closed for November of this year in an attempt to maintain prices.
Good call, Mr. W.

So the question then arises: in today's modern technology sector, are there any monopolies which can't (yet) be effectively contested?

Apple's iPhones? Nah, Samsung's shown that can be contested. Facebook's Ts&Cs? If they really get bad enough, Google+ is only a couple of clicks away. Google pleasuring company-owned results in the listings? Microsoft's "Scroogled" campaign is certainly trying to challenge that. Ebook pricing? I can't see that anyone would be able to exploit a monopoly there for any length of time: maybe on the back catalogues perhaps, but not on the new material.
In fact, in the current state of flux I'm not sure that I can see any non-contestable positions in the general computing market. There's enough VC money out there that anyone at all truly trying to exploit a monopoly position is going to face someone contesting it.
This seems like a reasonable challenge: can we identify a likely tech monopoly which can be preserved for (say) at least the next 5 years? Let's talk about Microsoft Office.

There have certainly been no shortage of challenges to Microsoft Office in the past, but it has settled down to being the de facto solution for text documents, spreadsheets and presentations in the enterprise. Its hold on email is weaker: corporate support for web-based email and calendaring like Gmail has managed to provide effective competition for Exchange and Outlook, helped by the need for any email sent externally to conform to the major content standards. Databases have to speak SQL and so you can in theory replace Access with any number of free, supported or propietary (Oracle) alternatives. But Excel is the spreadsheet format in financial industries, and Word is the de facto standard for exchanging text documents - every bloody time a recruiter has spoken to me, they've asked for my CV in Word format.

In theory it is not a particularly difficult task to write a spreadsheet program or word processing program that does nearly everything important that Excel or Word can do, and make it perform better and be easier to use than its MS Office counterparts. The problem is, though, you have to work with a massive installed base of Excel and Word users who will demand that you be able to receive, view, edit, and return their documents in their original format. Despite the availability of the binary file format information for these documents, this task still turns out to be stupendously hard. Joel Spolsky wrote a fascinating and detailed piece in 2008 on why the Microsoft Office file formats were so complicated:

It's very helpful of Microsoft to release the file formats for Microsoft and Office, but it's not really going to make it any easier to import or save to the Office file formats. These are insanely complex and rich applications, and you can't just implement the most popular 20% and expect 80% of the people to be happy. The binary file specification is, at most, going to save you a few minutes reverse engineering a remarkably complex system.
[...]
Anyway, unless you're literally trying to create a competitor to Office that can read and write all Office files perfectly, in which case, you've got thousands of years of work cut out for you, chances are that reading or writing the Office binary formats is the most labor intensive way to solve whatever problem it is that you're trying to solve.
Note how this is different from the examples of email and calendars which have been successfully challenged - email has always been open, and replacing it with a proprietary format is a non-starter. Calendars were implemented on so many non-MS devices that open formats for calendar information interchange such as CalDAV were well-established; Microsoft Outlook and Exchange had to support them in order for users' mobile phones to be able to sync calendar information. Spreadsheets and any text document with a more complex structure than RTF have never required such exchange.

Anyone trying to contest the MS document monopoly therefore has to think of better ways around it. One approach is Google Apps for Business which takes the view that, since any modern computer has a decent web browser and most businesses have good internet connectivity, why not manage not just email and calendar information but also spreadsheets, text documents and presentations on the web. Documents are stored in Google's data centers; anyone wanting to view or edit a document just needs the URL of that document (and have the author grant their user ID granted permission to access it) and can then view or edit it live online. This has several benefits, notably the saving in a company's IT costs - no need to have much document storage space, a version control system or backup pipeline, Google does all the work. It also means it's easy for several people to view and edit the same document at once. You can see how Microsoft is concerned about this selling point by its launch of Office 365 which is essentially Office in the cloud.

The two top problems which Google is going to have to tackle if it's to eat away at the MS Office monopoly are offline access and file format compatibility. If all your docs are online, and your company's Net access goes down for an hour due to a ISP problem or failed switch, all your employees are blocked from getting anything done; this is a very visible potential pain point. How to fix it? Google seem to be tackling this with some support for offline doc editing although presumably it's only good for working on docs that you were viewing before the Net connection went down. The second point is going to be the tricky one - how do you receive a Word doc via Gmail, store it somehow in a format that Google Apps can edit, edit it and then export it in the same Word format without encountering the Chinese whispers problem? It's going to be entertaining to watch how the format wars play out. Expect some manoeuvering from Microsoft if any competitor looks to be close to a fix for this.

There's a third problem regarding how to make available nearly all the functionality of Excel, Word etc. in a combination of JavaScript client code and server-side code without making the interface painfully slow to load and run, but I suspect a combination of improving client and network speeds and smart design should nail that eventually.

The hardest non-technical problem is going to be persuading the IT admins and management of companies currently running MS Office to shift to an alternative solution. Since they have likely spent most of their lives working with Office (oh, the horror!), it would be a poor career move to throw away all their accumulated expertise and adopt an office docs system where a spotty graduate knew as much as (or indeed more than) they did; any CTO worth his salt would shortly be asking pointed questions about what, exactly, these high-paid MS experts were spending their time on. I suspect that moving from MS Office is going to have to be pushed from the top down; in the short term, I expect to see Google Apps waved in front of Microsoft by many of their corporate customers as a tactic to reduce their annual fees.

Given the above, I claim that MS Office is, at the moment, an effective technical monopoly in small-, medium- and large- sized businesses. I expect it to be challenged and de-monopolised in small businesses in the next year or two, but it should hold on and be near-unassailable for at least the next five years, unless Steve Ballmer does something even crazier than normal.

Update: (2012-12-26) Mr. Worstall responds:

Whether Office's near monopoly is contestable in the economic sense is an unknown as yet. We'll only find out if it is if someone manages to replace it. However, it is most certainly being contested by Google. And the very fact that it is being contested is going to put a brake on the ability of Microsoft to exploit that near monopoly position.
It's a fair point. It'll be interesting to see the position in 1-2 years. I think you could get a reasonable metric of the degradation of MS's monopoly by comparing the per-capita cost of MS Office licenses being paid by medium and large enterprises in 2010 thru 2014, discounted by RPI. If you see prices being forced down significantly across the board, you can assume that MS is being beaten over the head by credible threats to move to Google Apps or similar alternatives, and the monopoly is cracking.

2012-12-22

Actual facts in the gun control vs shooting debate

Larry Correia, who is the bestselling author of the "Monster Hunter International" book series and (perhaps more relevantly) was a firearms and concealed-carry instructor, owned a gun store and worked with a lot of police and military, has penned a long and detailed dissection of the gun laws and post-Sandy Hook gun control proposals. Go read the whole thing, but here's a flavour of it.

Larry's credentials:

We [the gun store] were a Title 7 SOT, which means we worked with legal machineguns, suppresors, and pretty much everything except for explosives. We did law enforcement sales and worked with equipment that is unavailable from most dealers, but that means lots and lots of government inspections and compliance paperwork. This means that I had to be exceedingly familiar with federal gun laws, and there are a lot of them. I worked with many companies in the gun industry and still have many friends and contacts at various manufacturers. When I hear people tell me the gun industry is unregulated, I have to resist the urge to laugh in their face. I was also a Utah Concealed Weapons instructor, and was one of the busiest instructors in the state.
His main points, most if not all of which are backed up with actual facts and statistics:
  • Arming teachers is not crazy, as long as you don't actually mandate it; let the interested and concerned teachers train and qualify for CCW permits and you'll be just fine. Utah has CCW teachers and the world has not ended.
  • Mass shooters shoot far more people if they are eventually stopped by cops rather than civilians, because cops take 5-15 minutes to arrive. He notes that nowadays when cops arrive a the scene of a shooting they go straight in and engage the shooter rather than forming a perimeter and waiting for SWAT, because the additional wait time costs too many lives.
  • Mass shooters choose gun-free zones for shootings because they don't want opposition. When people fire back at them, they tend to shoot themselves because their rampage has come to an end.
  • The media makes mass shooters infamous, which is likely encouraging others to become shooters and gain infamy.
  • Banning various types of weapon, ammunition, magazine size or accessory makes very little difference. The term "assault weapon" in law has no relation to actual killing or wounding ability. Nearly any effective handgun or long gun is "semi automatic".
  • Banning all guns deprives people of the ability to defend themselves; guns are used much more in self-defence than in offence, but the former occasions often do not result in a shot being fired. Countries that ban guns see violent (if not fatal violent) crime soar as predators know there is no effective defence against them.
  • Confiscating guns from American citizens would result in a legal nightmare and any number of shootouts as you confirm in gunholders' minds that the government wants to take away all their rights and their ability to protect themselves.
Food for thought. You don't have to agree with Correia's politics and perspective to realise that he has valid points about the ineffectiveness of the gun control proposals being bandied around.

Hotness is a valid reason for firing

Good news for those of us who aren't exactly oil paintings, and feel bad about our appearance compared to attractive co-workers; Iowa's High Court says that very attractiveness can be grounds for termination:

The court ruled 7-0 that bosses can fire employees they see as an "irresistible attraction," even if the employees have not engaged in flirtatious behavior or otherwise done anything wrong. Such firings may be unfair, but they are not unlawful discrimination under the Iowa Civil Rights Act because they are motivated by feelings and emotions, not gender
The author notes that the Iowa High Court is all-male, which perhaps tells us something about the judges' assessment of their own self-control.

Just to balance out the risks, however, it appears that excessive flatulence can also put your job at risk.

2012-12-19

Networks as a bottleneck

Something I've been watching with interest recently is the failure in cross-continent network speeds to keep pace with the rapid growth in storage. And why, you might ask, does this keep anyone up at night? It turns out that if you have been storing lots of videos of dancing cats with hosting firm A, and then firm A starts to exhibit financial wobbles or the country in which firm A is based decides to pass new laws about access to data, and you decide you want to store them with hosting firm B instead, you are going to find that network transmission is a very real bottleneck.

Some baseline numbers first of all:

  • A regular 2-layer Blu-Ray disc contains about 50GB (gigabytes: 10^9 bytes) of data, which could store just over 3 hours of 1080p (HD) video, or 15GB/hour.
  • Leasing 10Gbps (giga-bits-per-second, 1 byte == 8 bits) of bandwidth between New York and London will cost you around $10,000 per month

Let's suppose that we have about 100,000 hours of video, which is 1,500,000 GB i.e. 1.5 PB ("petabytes": 10^15 bytes). If we wanted to transmit that over a 10Gbps link, that's 1.1 GB per second. It would take (1,500,000 / 1.1) seconds - 378 hours, or a litte over 15 days) to transmit that raw data, assuming perfect transmission; let's add a reasonable 20% overhead, and you're looking at 18 days non-stop to send your data over the Atlantic.

Ironically, cost isn't the main issue: leasing the line for that time (and assuming you can lease the exact period at the pro-rata cost) will cost you $5400, and compared to the $90,000 that the 750 x 2TB disks to hold your data will cost, that's just noise. But you are going to have tremendous problems shifting your data significant distances.

That's OK, though; as Andrew Tanenbaum famously said:

Never underestimate the bandwidth of a station wagon full of tapes hurtling down the highway.
You can easily get 1.5TB per backup tape, so 1000 tapes (a cube of 10x10x10 tapes, about 50cm x 100cm x 220cm, fairly easily managed in one person's checked luggage on a reputable airline) should be fine for this data. For redundancy we take two copies of the data and send two gofers on separate LDN-to-NYC flights; say, $600/person. Within 10 hours we can get the data trans-Atlantic with a tiny data loss; say, 5% of tapes fail so only 2-3 tapes of data fail for both gofers, and we can get that data sent via the network for a tiny cost in time (about 20 minutes per tape). Problem solved?

Well, you've still got to get the data onto the tape. Tape writing speeds are about 900GB/hour at the top end, so with one tape drive it will take 1666+ hours to write all the data to one set of tapes, or 3333 hours total. With 100 tape drives, you could get this down to 33 hours of writing - but don't forget the reading at the other end, which means you need another 100 tape drives. At $1000+ per tape drive for a high-speed device you're looking at $200K just for your reading/writing devices, more than double the cost of the drives themselves.

So if you have money to burn, you can get 98%+ of the data across the Atlantic in 33 + 10 + 33 hours - 6 days, one third of the time required for direct network transfer. However it's going to cost you a lot. A cheaper option would be ripping the hard drives out of firm A, duplicating them into two other sets of drives (using very cheap hardware with multiple SATA interfaces and minimum-wage goons to swap drives in and out. SATA 2.0 transfer speeds of 300MB/s == 1TB/hour) are much better than tape; the limiting factor is the hard drive write speed which is still about 100MB/s == 300GB/hour, and the marginal cost of an extra motherboard for the transfer is order of $200 instead of a $1000 tape drive, so you still end up about 40% faster this way for the same cost as tape. You can then send your two gofers with the hard drives instead of tapes, although they're going to be a lot heavier - around 300Kg of luggage per person - so they'll need to pay quite a bit extra; I'd estimate an additional $400 per 100Kg (essentially, one person plus luggage). As a bonus, I'd expect the disk failure rate to be lower.

High-speed Internet is definitely here for the masses, but it hasn't yet really impacted long-distance network connections. Tanenbaum's maxim can be considered safely debunked; tape read/write speeds have not kept pace with tape capacities, and even cheap disks need expensive interfaces.

2012-12-17

A telling firearms death map from the BBC

Their article includes the Top states for gun deaths rate including major cities

  1. DC (12.5) - Washington (50% black)
  2. Louisiana (8.8) - New Orleans (67% black), Baton Rouge (45% black)
  3. South Carolina (4.8) - Columbia (42% black), South Charleston (7% black)
  4. Maryland (4.7) - Baltimore (63% black)
  5. Missouri (4.6) - Kansas City (30% black), Saint Louis (35%)
  6. Mississippi (4.6) - Jackson (79% black)
  7. Michigan (4.6) - Detroit (82% black), Grand Rapids (20%)
  8. Georgia (3.8) - Atlanta (54% black), Augusta (55% black)
  9. Tennessee (3.8) - Memphis (62% black), Nashville (28% black)
  10. Arkansas (3.7) - Little Rock (42% black)
This tells you that the gun death rate is highest in states with major cities with a substantial black population - note that out of the top 10 US cities by population : NYC NY - 25% black, LA CA - 10% black, Chicago IL - 32% black, Houston TX - 25% black, Philadelphia PA - 44% black, Phoenix TX - 7% black, San Antonio TX - 7% black, San Diego CA - 7% black, Dallas TX - 26%, San Jose CA - 3%; none appear.

As a note, the overall % of the USA population that identifies as black is 12%.

2012-12-16

More USA death and injury stats

I was reading through an American article on pentrating abdominal trauma this afternoon (who says I don't know how to have a good time) and it started with this little nugget:

According to data published by the National Vital Statistics Reports, 11,406 homicide deaths occurred from firearm injuries in 2009 and 18,689 deaths from self-inflicted GSWs.[2] Forty percent of homicides and 14% of suicides by firearm involved injuries to the torso.
The 11.4K and 18.7K stats are slightly higher than the stats I saw recently, but let's work with that. The injury split is more or less what we'd expect: you would expect suicide by firearm to be almost universally targeted at the head, which means that 14% of firearm suicides either have very poor aim or something else is going on. Note that these are fatalities, not injuries; the remaining 60% of homicides would be mostly to the head, but there will be some fatal limb injuries (wrecking the femoral or brachial arteries and bleeding out), plus some multiple-site shootings where the fatal shot can't be nailed down.

Anyway, this led to a Medline article on race and insurance status as risk factors for trauma mortality (free registration required). This looked at the race of the victim rather than the offender, and from a reference group of white insured patients across a base of 400K USA victims. Now "insurance status" is a pretty good, if not 100%, signal for poverty; poor people don't tend to have medical insurance, they rely on Medicaid if anything. What this study determined was that poverty (lack of insurance) and African-American / Hispanic race status are independent predictors of poor trauma outcome, but poverty has a stronger association. So the number of trauma (vehicle accident, shooting, stabbing, suicide) deaths in the USA is at least partly because the people who live in areas where shootings occur most (poor, Hispanic + African-American communities) don't have good insurance and therefore don't get trauma care that's good enough to save them.

The CDC says that 180,000 Americans die of violence and injuries every year, and it's the dominant cause of death in the age range 1-44 years. They provide a data source called WISQARS which lets you query injury data including fatal injuries. Let's have a look at the stats for 2010.

  • Accidents (unintentional injury) is the leading cause of death from ages 1 thru 44, about 120K deaths overall
  • Suicide peaks in the 25-34 age group at number 2 and then tails off; about 38K deaths, 30% of accidents
  • Homicide peaks in the 15-24 age group and number 2 and is number 3 in age group 1-4; about 12K deaths, 10% of accidents
  • Flu kills 50K people per year, mostly in the 65+ age group
  • The ranking of death causes is about the same for men, but they are disproportionately likely to be victims of accidents (75K of 120K deaths) and suicide (30K of 38K deaths)
  • For black males, homicide leaps up to become the number 1 cause of death in ages 15-34, and nearly the same rate as accidents overall (8K accidental deaths vs 6.7K homicide deaths) but suicides are much lower-ranked overall (1.5K deaths)

So black men seem less likely to kill themselves, but more likely to be killed by others. And gun homicides are only 25% of the number of suicides overall. Yes, 50% of suicides are by gun - but then, given a 40% gun ownership rate, why wouldn't they be?

Perhaps not as many people would kill themselves if they didn't have a gun handy. Let's look ar the death stats in the UK where guns are very rare (though not non-existent). Data source is 2011 death registrations summary (final) which is 1 year off the 2010 USA figures, but I assert that no significant difference should be due to that time difference. UK population was ~65 million vs USA population of 311 million, so you can multiply UK figures by 5 to get roughly equivalent USA numbers

  • Accidents killed 6.5K men and 4.9K women (11.4K people corresponding to projected 55K and actual 120K in the USA - the USA is clearly a more dangerous place)
  • Transport accidents killed 1.4K men and 400 women - men are clearly much worse drivers
  • Falls killed about 3.9K men and women, equally divided.
  • Poisoning / exposure to poison killed 1.9K men, 900 women
  • "Accidental exposure to unspecified factor" killed 900 men and 1.5K women - WTF? Anyone any idea what this represents?
  • Intentional self-harm (basically, suicide) with deliberate intent killed 2900 men and 800 women for a total of 3.7K, and with undetermined intent killed 3.7K men and 1.2K women for a total of 8.6K (projected 43K and actual 38K people in the USA).
  • Assault and yet-to-be-determined-intent killed 480 men, 210 women for a total of 700 homicide-like instances (3500 projected vs 12K actual deaths).
So the USA has a suicide plague at similar rates to the UK, though the UK seems to have 1000 more suicides than it ought to if the USA is used as a baseline. The USA has about 3.5X the homicide rate of the UK. I note in passing and with reference to my post about US vs Canadian deaths and homicides by ethnicity that the UK has a black population fraction of less than 2% compared to the USA's 12%.

Overall, if anyone in the US political spectrum is serious at reducing unnecessary deaths of young people, they should be looking to reduce the rates of suicide and fatal (non-gun-related) accidents. Since they are not doing this, instead focusing on e.g. implementing a federal assault weapons ban similar to that which was so effective in Connecticut, I conclude that they are only interested in cheap political posturing and are hence below contempt. Perhaps "the war on poor mental health" and "the war on accidents" are not as vote-grabbing as "the war on guns", "the war on drugs".

2012-12-14

US and Canadian guns

Following today's awful school shooting in Connecticut I wondered what gun ownership / crime was like in the USA versus Canada. So let's take a look.

So in the USA, despite a gun ownership rate that is not massively different from Canada, you are much more likely to be the victim of gun homicide (5x as likely) or gun suicide (3x as likely), and a bit more likely to be the victim of a gun accident (2x as likely, which is about what you'd expect from the ownership rates). Why then do Americans kill each other with guns about twice the rate you'd expect if they were Canadian?

It turns out that the dominant factors for firearm homicide are race, sex, and age group:

The victimization rates for blacks were 6 times higher than those for whites.
The offending rates for blacks were more than 7 times higher the rates for whites.
Males represent 77% of homicide victims and nearly 90% of offenders. The victimization rates for males were 3 times higher than the rates for females. Approximately one-third of murder victims and almost half the offenders are under the age of 25. For both victims and offenders, the rate per 100,000 peaks in the 18-24 year-old age group.
It looks as if the dominant reason for the firearm homicide rate in the USA is the proportion of young black men. 700K Canadians identify as black (2%) compared to 12% in the USA.

That's not the entire explanation - you don't get school shootings in Canada, and the major school shooting perps in the USA were white (or Korean). But it's a disturbing set of statistics.

First Benghazi sacrifice - Susan Rice

Sure enough, a decent interval after the US election, Ambassador Susan Rice takes the fall for Benghazi. Nominated for the position of Secretary of State, she has withdrawn her nomination after Republican politicians have repeatedly asked pointed questions about her comments on the Benghazi attack:

The move followed weeks of controversy on Capitol Hill over the possibility of her nomination, with Republicans threatening to block Rice from the post over concerns about her September comments on the Libya terror attack. Some lawmakers continue to charge that Rice misled the American people when she said on Sept. 16 that the attack was the result of a "spontaneous" demonstration spun out of control.
President Obama, in a written statement Thursday, called those claims "unfair and misleading" but said he accepts her decision.
There have been the expected polemics about the Republicans being unable to tolerate a woman of colour as Secretary of State, but I think the Obama administration must be quite relieved that Susan Rice has been effective as ablative protection of the administration. No harm done, they can point to Rice's departure as a retort that "action was taken" about Benghazi.

I'd like to think that the US and UK media will press for more information about exactly who failed to send support to the besieged Benghazi consulate and why support was refused. However I expect they will use Rice's withdrawal as an excuse to avoid any further investigation. If you don't want to rock a boat, you don't shift from your seat.

2012-12-12

Schmidt's balls must clang as he walks

In noted contrast to Starbucks, Google's chairman Eric Schmidt is up-front about the company's approach to tax:

"We pay lots of taxes; we pay them in the legally prescribed ways," he told Bloomberg. "I am very proud of the structure that we set up. We did it based on the incentives that the governments offered us to operate."
"It's called capitalism," he said. "We are proudly capitalistic. I'm not confused about this."
You may or may not agree with Schmidt's approach to international taxation, but you can't deny that the man has style.

Of course, this didn't go down too well in the usual quarters:

In Britain Vince Cable was unimpressed by Mr Schmidt's views. The Business Secretary told The Daily Telegraph: "It may well be [capitalism] but it's certainly not the job of governments to accommodate it."
Well, Vince, since the practice is apparently legal, it appears that governments have been accommodating it. So are you not doing your job? or would "doing your job" result in companies leaving the UK en masse and depriving the UK of jobs, VAT, NI and income taxes? Let's see.

2012-12-11

27% of UK citizens have a death wish

Reading this Daily Mail article claiming that 64% of people walk on past a gang of teen troublemakers, I can only wonder - what's wrong with the remaining 36%? The poll claims that 27% overall would step in to stop a gang drinking and verbally abusing passers-by, and I'm sure that this overstates the numbers who would actually intervene - still, any rational evaluation of self-preservation would force the average passer-by to avert their eyes and keep walking. I do not, of course, include Chuck Norris or Jack Bauer in that category, but I understand they seldom visit South London or Birmingham

If drunken teens are abusing bystanders, they're not going to stop when someone they don't know from Adam tells them to stop. Why would they? They've been brought up in an environment where there are no effective sanctions on bad behaviour. Even if they've been in the criminal justice system due to thefts, assault etc. the presumption is to hand out suspended sentences to "encourage" them to go straight. Someone mouthing off at them is very unlikely to change their behaviour unless it's their mother - we assume that their father either isn't around, or is a waste of skin.

If it escalates into physical violence, the problems are worse. If you initiate the violence, even if defensively in fear of your safety, you will certainly be arrested and likely charged. The CPS might well not be able to make the charge stick, at least if you're sensible enough to keep your mouth shut and demand legal representation, but that's still several months of your life with the end of your professional career hanging over you. If they attack you without provocation, it's a toss-up as to whether a bystander will back your side of the story when the police turn up; they will claim you attacked them, and there's no downside for them to do this. Perjury prosecution? Don't make me laugh.

Practically, unless you get in fights on a regular basis, you are very unlikely to come out well from a 1 vs 2 engagement with teens; if they're drunk, their reflexes are slower but they feel less pain. Even 1 vs 1, you're gambling that they don't get a lucky hit or kick in and disable you. Once you're on the ground, they're going to kick you in the chest and head, even if they're a woman. A kick in the head in the right place can be permanently disabling, if not fatal.

What's the upside? You won't change anything by intervening. Drunk teens mouth off at bystanders every single day. Even if you scare them off today, they'll be back tomorrow. You can't apply any violent sanction, because you'll either be jailed (if disproportionate), arrested and charged (if proportionate defence), or in hospital or the morgue (if outnumbered or unlucky). Congratulations to the lawmakers and social scientists who have put this incentive scheme in place. May you be verbally abused and threatened daily.

2012-12-10

Short term gain for long term pain (SEP)

This insanity from California expresses perfectly why any system allowing politicians to spend money today borrowed from tomorrow presages the fiscal end of the world:

"We'd be foolish not to take advantage of getting $25 million" when the district had to spend just $2.5 million to get it, [president of the school board] Ramsey says. "The only way we could do it was with a [capital appreciation bond]."
[...]
In the West Contra Costa Schools' case, that $2.5 million bond will cost the district a whopping $34 million to repay.
But why should Charles Ramsey worry? He's not going to be around in 5-10 years when the CAB payments start to kick in. He's not personally liable for the school's debts. What rational risk is there to him? He gets to build a new elementary school, accumulate the resulting kudos (and, who knows, maybe move on to a non-executive director position with the construction firms benefiting from the new build). Some other poor schmuck is going to be in the hot seat when the school has to start paying back the $34 million at $200K per month over 15 years, or whatever. It's Somebody Else's Problem.

I do note, however, that if a regular Joe borrows £50,000 in order to get a 25% deposit and preferential interest rates on a £200,000 mortgage then a charge of mortgage fraud is likely to find him in short order.

Any resemblance to today's public sector pensions crisis and deficit spending behaviour is purely intentional.

[Hat tip: ZeroHedge]

2012-12-09

Affirmative action for college - a game of halves

Gary Younge in The Guardian sings the praises of affirmative action for college:

Meanwhile, the two examples that the policy's opponents use most often to restrict access to good higher education for non-white people – for that will be the outcome – actually, in quite different ways, prove the opposite.
The first is Barack Obama. His success, the argument goes, shows that such assistance is unnecessary. [...] However, Obama, like [Condolezza] Rice herself, says he probably was a beneficiary of affirmative action.
OK, affirmative action for college admissions should continue so that we benefit people like Barack Obama. News flash, Gary: Obama is only 50% black. His mother was about as white as they come. So, are you saying that affirmative action which "take[s] into account that our society has been doing something special against the Negro for hundreds of years" should also benefit students who are 50% black, 50% white? What about those who are 25% black, 75% white? Where do you draw the line, Gary? As the races blend, how are you going to determine if someone is "black enough" to benefit? The Fauxcahontas scandal around Elizabeth Warren related to her claim of being 1/32 Cherokee which was viewed as sufficiently "Indian" to give her school some affirmative action credit.

If affirmative action is to redress wrongs against Jim Crow laws, should it be extended to those of Hispanic ancestry? Chinese people have suffered racism, pogroms, mass famines - and their students out-perform white students, and indeed suffer as a result of affirmative action towards other ethnic groups. How, exactly, does this work? Asian-American groups filed an amicus brief in Fisher v Texas, an affirmative action case that went to the Supreme Court, arguing that affirmative action hurt Asian-American students. Is this fair?

The modern day practical purpose of affirmative action is to have an affirmative action program. It's a great non-job factory, creating lots of federal and state jobs, and is relatively simple in application since the judgement of its success is solely based on the changing fractions of ethnicities in students. It does not, however, fix the academic problems of African-American students. If you want to fix these, you have to fix the academic environment in their elementary and high schools. By the time they are applying to university, the damage is done.

2012-12-07

Starbucks' philanthropic problem

It appears that Starbucks' voluntary donation of £20 million in corporation tax has made the waters murkier, not clearer:

Starbucks said it will make the corporation tax payment even if it makes no profits in the UK – however, there is no legal basis for the company to make such a payment.
Oh dear. The rule of law comes back to bite the UKUncut folks in their bottoms! This is the problem when you insist that companies and people adhere to the "moral" position - this is sometimes a way from the "legal" position - and if companies don't follow the law, where are they?

At least, if it can't voluntarily pay tax, Starbucks can make a charitable donation to the UK. Problem solved!

Suggestions have been made that bosses could make a charitable payment instead. However, the company could claim this back against tax.
Chris Morgan, head of tax policy at KPMG, said: "A charity payment has to be claimed against tax."
Oops. So it would shift tax payments from (say) the USA to the UK. The company would not necessarily be paying more tax. (Say, $1 million annual profit, taxed at 25%; $250K. UK charitable donation of $100K, roughly profit attributable to UK. That comes off taxable profit, so only $900K taxed at 25%; they pay $100K to the UK but save $25K in tax they don't pay in the USA, so they pay the USA $225K and the UK $100K. I wonder how the US tax authorities view this? More to the point, I wonder how the shareholders are going to view this?

2012-12-06

Give the taxpayers what they ask for

Karl at the US political site Hot Air has a proposal so simple that it's genius. Cutting through the Democrat-Republican cat fight about what mix of tax raising and spending cuts should address the deficit, he proposes that noted Republican "shock jock" Rush Limbaugh should back raising taxes to fund current and planned spending with no cutbacks:

If the Democrats want to increase taxes and leave entitlements unreformed, why not propose that the federal government raise the taxes necessary to fund these purportedly essential programs?
It sounds like it might just work - but what levels of tax raising would be required?

The four year plan he proposes would be popular for at least the first couple of years: households with over $250K annual income get nailed in the first year, and those with over $100K in the second year. However, by year 3 all the "rich" have been squeezed as hard as possible, and the spending deficit is still 3% of GDP. So it's time for every other taxpayer to pay up:

In 2019, increase all tax rates on ordinary income 5 additional percentage points, phased in over 10 years. Increase both tax rates on capital gains 10 percentage points (to 20% and 33.8%), phased in over 5 years.
But that still doesn't keep spending under control, so the only thing left is year 4:
Impose a 10% national value-added tax, phased in over 5 years.
This will actually give a balanced budget, finally. Obviously the US still has $20tn or so of debt, but at least it's not adding to it. Everyone who pays taxes is then bearing the spending burden of the state. And in 2016 it's election time once more...

This, I think, expresses the essential dishonesty of people clamouring "tax the rich" as a solution to the current deficit crisis - and this is true in the UK quite possibly as much as in the USA. You can't fix the deficit and keep spending at the present level without hitting everyone who pays taxes, hard. Now perhaps that's what people want, but it would be interesting to make the decision very real for them. "Do I as an average taxpayer want a 60% hike in what I pay in taxes, or should I start demanding that the government stop pissing away so much money?"

2012-12-04

Good news about the change to the UK laws of succession to the throne

Despite the Daily Mail publishing possibly the most stupid article in the world about the royal pregnancy and possibility of twins, the most relevant story about the royal succession is that gender is no longer a determining factor. This resolves a long-standing ambiguity and potential for constitutional crisis - what if the royal baby were a hermaphrodite? Under the old succession laws, the order of succession would have been ill-defined. Now that gender is irrelevant, a hermaphrodite royal can succeed in the same order as a male or female. I expect that's a weight off everyone's mind.

Probability and the TSA

The motto "Be careful what you ask for, you might just get it" is one that TSA IT Strategy Branch Chief Russell Wooten might usefully consider. His request for "ideas on how the Transportation Security Administration (TSA) can improve its capabilities in utilizing its current security technology, upgrading its security technology, or improving its security processes?" landed in the inbox of Sommer Gentry who is a maths professor at the US Naval Academy. Her advice for Mr. Wooten bears careful study. For instance, the explosive screening swab test:

Using Bayes' Rule, we can calculate that even if explosives detection technologies were nearly perfect: catching every actual explosive and only falsely alarming on one in every ten thousand passengers, then only one out of five million positive test results actually indicates presence of an explosive device.
At that level of false positives, the screeners will rapidly lapse into a belief that every alert is a false positive, and let the real bad guys through. Oops! Still, surely backscatter body scanners (AIT) make some contribution to safety, despite slowing passenger progress through security checkpoints?
A recent RAND study of airport vulnerabilities at LAX concluded that "small, portable explosives have been the most likely and most lethal means of attacks at airports" and that "The greatest risks for casualties for most types of attacks are in the high-density areas passengers encounter before reaching the security checkpoint, particularly lines for ticketing and for passing the security checkpoint." Thus, AIT is not only ineffective, it is actually dangerous because it leaves passengers waiting in long lines vulnerable.
Oh. Maybe not.

One is left wondering whether the security theatre of the TSA has actually succeeded in deluding the TSA itself into thinking that it is defending air passenger safety, when all it appears to do is inconvenience passengers and call that "deterrence from attack". It certainly doesn't catch any actual bad guys - the Underwear Bomber avoided TSA scrutiny by the simple expedient of starting from Amsterdam and aiming to detonate his bomb over USA airspace. Mind you, I wouldn't have been surprised if he had managed to go through backscatter scanning without raising any alarms.

Syria's chemical weapons strategy

I'm relatively sanguine about the signs that Syria may be preparing to deploy chemical weapons:

Syrian forces began combining chemicals that would be used to make deadly sarin gas for use in weapons to attack rebel and civilian populations, a U.S. official tells CNN.
This, if true, is certainly a necessary precursor to deploying sarin. It also places a deadline for the use of the sarin: after a few weeks it is likely to have degraded significantly. So what might they be planning to do in the next few weeks?

Why would the Syrian military initiate (or, at least, appear to initiate) their sarin chemical munitions? If they actually intend to use them, what are the targets? Chemical munitions are effective against wide areas; either to deny the attacker the ability to move across them - which seems irrelevant in the context of the guerilla war being conducted by the Syrian rebellion - or to hit a concentration of forces in a town or city. The latter seems more probable, and just the threat of it may force the rebels (and civilians) to flee major towns and cities in order to present a less focused target. The weather in Damascus is currently showing a high of 17C and low of 8C, with potential light rain - assuming similar conditions across Syria, these are good conditions for deploying chemical munitions as the lack of bright sunshine and heat improves the persistence of the chemicals.

The use of chemical weapons against civilians in recent years has been very limited. The most infamous is Iraq's March 1988 attack against the Iraqi Kurdish village of Halabja. This involved a mix of chemical munitions: mustard gas and the nerve agents sarin, tabun and VX, and killed around 5000 civilians more or less immediately. International condemnation afterwards was widespread, though international action against Iraq was notable by its absence. Perhaps this has emboldened Syria, believing that the use (or the threat of use) of chemical munitions is essentially risk-free. Despite Turkey's concerns, I don't see Syria being stupid enough to launch a chemical attack on Turkish territory, causing NATO to invoke Article 5 and giving the USA, UK and Turkey free rein to bomb the bejasus out of Syrian command+control installations; gassing its own citizens, however, may still be viewed as safe.

Perhaps surprisingly, US Secretary of State Hillary Clinton has made the right call:

"I'm not going to telegraph in any specifics what we would do in the event of credible evidence that the Assad regime has resorted to using chemical weapons against their own people. But suffice it to say, we are certainly planning to take action if that eventuality were to occur," Clinton told reporters.
Assuming no-one from D.C. leaks the actual plans (and that seems to be a big assumption, but we live in hope) this is near-ideal. It warns the Syrians that the US has publicly committed to action in the event of an attack, but gives them no real idea what scale, duration or focus that action might take. Uncertainties are the last thing that military planners want - especially where the most powerful military on Earth is involved.

The reaction of members of the UN Security Council to a high-casualty use of chemical weapons against civilians would be fascinating. The US and UK would condemn it and try to organise a vote approving use of military force against Syria to retaliate. France - who knows. Russia and China would no doubt veto it and instead propose some mealy-mouthed admonishment. The other members of the Security Council are Azerbaijan, Colombia, Germany, Guatemala, India, Morocco, Pakistan, Portugal, South Africa and Togo; I suspect most of them would sanction military action on the pragmatic grounds of staying on the good side of the USA, while not actually getting involved themselves. But with the Russo-Chinese veto, it's not going to be sanctioned anyway, so approving military action is consequence-free.

Would the USA and UK try a non-UN sanctioned strike against Syrian command+control? I don't see why not. No-one actually likes Syria, and while Russia, China, North Korea, Iran and Venezuela might protest the attack (wary of the precedent of allowing other nations to have a say in the right of a country to slaughter its own civilians) I doubt there will be much political downside for President Obama and PM Cameron. Sending a few B2s over some randomly selected military bunkers to unload 5000lb GBU-28 bunker-busters would make quite the mess. Alternatively, picking a large Syrian armoured formation in the open and dispensing CBU-97 Sensor-Fuzed Weapon containers would make it suddenly extremely immobile and ineffective.

Assuming the Syrian military is smart enough to see through the politics and doesn't actually want to be pounded, my best guess is that they're worried about concentrations of rebel forces and want to provoke rebels and civilians into leaving towns and cities near the front line; this would give them breathing space to mass forces and either counter-attack or reinforce defences. If so, they have about a month to make the best of it before it becomes apparent that the sarin-mixing was all a big bluff...

2012-12-01

The US Fiscal Cliff

I am entertained currently by the news from our American cousins about the forthcoming "fiscal cliff" which takes about $600bn a year out of the US economy and therefore is viewed as the end as we know it. Amid all the posturing, with the Democrats wanting to massively increase taxes and keep spending while the Republicans want to freeze taxes and slash spending, it's worth taking a look at what the fiscal cliff entails and whether the parties will let it happen.

The spending cuts portion of the cliff is actually (relatively) tiny - $110bn, of which half comes from defence and half from non-defence. Since the 2012 US federal budget had $3.8tn of expenditures this is a little less than 3% of cuts. I do not therefore feel that this is going to be the end of the world. The main Republican objection is that the military cuts will be disproportionate: $55bn out of $688bn is about 8% of cuts. Having said that, I suspect that the Democrats are going to be able to make most of that stick.

The cliff is heavily weighted towards tax increases, which is why I suspect the Democrats won't feel overly pressured to stop it. There are a lot of different areas affected. If you look at the 47% of citizens who don't pay federal tax (income or capital gains), it looks like they'll mostly feel the additional 2% Social Security (on up to $110K of income), the loss of earned income tax credit for some of them, and (depending on income) the increased scope of the Alternative Minimum Tax. If you have a single-earner household getting $40K, with 2 kids, filing jointly, I reckon you could easily be looking at an extra thousand or so in additional tax. Richer taxpayers will be more heavily hit: for $100K earners, the additional 2% Social security raises $2000 by itself, plus a 3%-5% raise in most income tax bands and reduction in some tax credits, means you could be looking at a $4000K+ additional tax bill. The overall impact of the tax section of the fiscal cliff is $500bn, which given 180 million workers is around $2700 per worker of impact. (Actually it's not quite that bad since the capital gains rises will bring quite a lot of money from the invested moderately well off elderly, and the estate tax rise is substantial as well - still, looks like $2000 per worker is the right ballpark.)

Because so many earners will be hit by these tax increases, I suspect the Democrats will need to push to change the balance of where they hit; it is no coincidence that the "tax the rich" mantra has been increasingly strident this year. Unfortunately, as many people note, there just aren't enough of the "rich". If you tax the top 1.8% of workers (1 million people) an additional $20K each, you're still only getting $20bn out of the $500bn you actually need. Even if you tax them $100K each, that's still only 20% of what you need. If you don't want to tax the poor, most of the taxes are going to have to come from the middle class - there aren't enough rich people to make a difference.

I suspect that mortgage interest tax relief is going to be on the table as a moderately bipartisan measure. It loses about $100bn of tax take annually, and even rich homeowners can benefit substantially:

As The Huffington Post previously reported, Barack and Michelle Obama claimed a $47,564 home mortgage interest deduction on their house in Chicago, which they bought in 2005 for $1.65 million. That equates to a savings of about $13,000 on their federal tax bill.
The Democrats will like it as an attack on "the rich", and the Republicans may well not mind it because many of the very expensive houses tend to be in solidly Democratic states like New York and California. I would guess that they'll drop the upper limit for house value from $1mm to around $300K, maybe staging the drops over a few years, and keep the option open to remove it completely.

I don't see the Democrats and Republicans coming to a credible agreement before January, however. I suspect there will be some more kicking the can down the road, maybe some token tax rises on the rich, but the USA will still end up spending far more than it raises in 2013. Let's not forget that the 2012 federal deficit was $1.3tn - even the fiscal cliff would only raise half of that gap.

2012-11-29

We're all going to drown

The BBC reports that the global sea level is rising and we're all going to die:

The study's headline conclusion is that the polar ice sheets have overall contributed 11.1mm to sea level rise but with a "give or take" uncertainty of 3.8mm - meaning the contribution could be as little as 7.3mm or as much as 14.9mm.
14mm a year! That's over 5.1 meters per century! Vast areas of low-lying land will be inundated! We're all doomed!

Oh, wait:

Another author, Dr Hamish Pritchard of the British Antarctic Survey, said: "The next big challenge - now that we've got quite a good understanding of what's happened over the last 20 years - is to predict what will happen over the next century."
So it's 14mm (worst case) over 20 years. 70 mm (7 cm) in a century. Whoop-de-doo. Can we please put the myth of catastrophic sea rise to bed?

I particularly liked the comment:

The findings are in line with the broad range of forecasts in the 2007 assessment by the UN's Intergovernmental Panel on Climate Change.
In 2007 they had 15 of the 20 years of data already, and they were trying to forecast the final 5 years. Colour me unimpressed. Forecast 20 years in the future with reasonable accuracy, and maybe we'll listen. What does "broad range" mean, anyway? Presumably not "claimed range"?

2012-11-28

Break open the popcorn!

This should be interesting: a gay French-Algerian man is planning to open the world's first gas mosque in Paris:

Mr Zahed says the mosque, situated in a Buddhist chapel in Paris, will also break another Islamic taboo by refusing to segregate women and men.
You have to admire the guy's minerals, even if you don't personally agree with his life choices. I can imagine that the reaction from Paris's resident Muslim community is going to be, erm, "spirited". Hope the congregation at the new mosque can afford good security, because I think they're going to need it.

So this is going to be an interesting choice for the more conservative-minded French citizens. Do they condemn him as a gay marriage proponent, or do they applaud him for confronting militant Islam and sexual segregation? It should be fascinating to watch. Good luck, Mr. Zahed, and I strongly suggest you ensure you have good home security and a fully paid up life insurance policy (before the insurance company finds out what you're doing)

2012-11-26

It's not prejudice if it's accurate

An expat friend of mine, acquainted with my political prejudices, sent me a copy of a magazine that his child brought home from school. His child is in the US public school system, and anyone who considers the US public education system to be a bunch of lefty commie pinkos would find their hearts warmed by the magazine's content. The magazine in question is "Scholastic News" (November 19, 2012, Edition 3), published in Missouri by Scholastic, Inc. The lead article is on Obama's re-election:

President Barack Obama will lead the U.S. for four more years. On November 6 he was elected to a second term as President of the United States. Obama is a Democrat. He defeated Mitt Romney, the Republican candidate. [So far, so good]
...
Obama was first elected President in 2008. He made history as our country's first African-American President. [Well, technically, African-American-Caucasian since his mother Ann Durham was mostly of white English ancestry.]
When Obama took office, many Americans were out of work. [And when he was re-elected, about the same percent were out of work: 7.8% in Jan 2009 vs 7.9% in Oct 2012].
He put together a plan to help save peoples' jobs and create new ones. [And what a triumph that was, see above.] He also signed a new health-care bill into law [which is making firms move full time workers to part time in order to avoid huge additional employment costs] and ended America's war in Iraq [which was drawing to a close anyway; in the meantime, anyone want to mention Afghanistan?]
...
"Whether you want to start a business, be a great doctor, or build something that's never been done before, I want to make sure that our young people have the chance to do it," Obama told Amiri. [And then he'll tell them they didn't build that.]
Good grief. Scholastic wouldn't know political neutrality if it was shoved up Scholastic's corporate posterior. Even The Guardian would be ashamed to publish a piece like this as news.

Scholastic is a company with $2bn annual revenue, 9500 employees worldwide, quoted on the NASDAQ (SCHL). Interestingly it looks like they took a real bath last week since investors are worried about the effect of federal spending cuts on them. I wonder whether their attempts to promote Obama are related to this?

2012-11-25

Let's give control of the Net to China and Russia!

Following on from the European Parliament voting to keep the Internet out of the hands of the ITU, we have the Chinese "People's Daily Online" "editorial team" arguing for the opposite - placing ICANN firmly under the thumb of the ITU. One can only imagine about the divergence between the opinions of the People's Daily editors and the Communist Party leadership; I suspect you'd find it hard to slip a sheet of rice paper between them.

It's fairly clear what the Chinese want:

More and more countries are beginning to question the U.S. control over the world's Internet as the international resource should be managed and supervised by all countries together.
Note that, as per the Europarl vote "more and more countries" does not seem to include many of the countries which actually respect freedom of speech. And what will this change consist of?
As a big country on the Internet, China opposes the U.S. unreasonable and unilateral management of the Internet, and seeks to work with the international community to build a new international Internet governance system.
"Governance". Well, we've seen the Chinese and Russian approaches to governance in their countries and the great contributions that they have made to human rights and freedom of speech. This sounds like just the approach that the Internet could benefit from, don't you think?

I submitted a comment to the People's Daily article, since they have a comments box. Oddly, no comments have yet made it through moderation. How strange.

2012-11-24

Napolitano chutzpah

Who is the USA's greatest ally? Britain? Canada? Poland?

Apparently, the answer is France:

According to the l'Express report, Homeland Security Secretary Janet Napolitano reportedly did not deny the allegations [about hacking French government emails] when asked point-blank about them.
"We have no greater partner than France; we have no greater ally than France," Napolitano reportedly answered, at the opening of an interview with l'Express.
Speechless. A little hyperbole to extract oneself from a sticky situation is acceptable, but this is comparable to the President of Poland reminiscing about the congenial border relations with Germany, or the Chinese Premier inviting the Japanese Premier for a cordial tour of Nanking to celebrate their nations' shared past.

For the record, I don't think that even the most ludicrously optimistic and Europhilic UK government minister would have made this claim. Quite why Napolitano thought she'd get away with it is beyond me.

2012-11-23

Europarl in "doing the right thing" shocker

It pains me physically to do this, but I have to applaud a recent Europarl vote: specifically, where the European Parliament is pushing back on the ITU's attempted Internet power grab:

[The European Parliament] believes that the ITU, or any other single, centralised international institution, is not the appropriate body to assert regulatory authority over either internet governance or internet traffic flows," the resolution reads. It was passed by a large majority of EP officials.
For those of you not following this event in detail, the UN-based International Telecommunications Union (which wouldn't know "innovation" if it sidled up and bit it on the arse) has been pushing to be more involved in Internet regulation. This has notably been led by such bastions of freedom and democracy as Russia and China.

Where global Internet regulation really matters is the Domain Name System (DNS), the process whereby a name such as "www.facebook.com" gets translated into a series of numbers such as 69.171.24.56 (the "IP address" of the site). There is a hierarchy of servers providing this look-up, but they end up at the 13 Internet root name servers. If a user in (say) Russia wants to look up information about feminism, she will conduct a Google / Bing / Baidu search for the relevant terms. Baidu will probably not have very many hits, but Google or Bing should point her to relevant sites. With luck, those sites will support the https protocol so anyone eavesdropping on her connection can see that she has connected to an IP address that matches the campusprogress.org site, but not what she is reading about.

The only way that she can normally connect to the campusprogress.org servers is to use DNS to convert the campusprogress.org name to the corresponding IP address. If Russia (via the ITU) can gain control over DNS globally, it can prevent campusprogress.org from obtaining an IP address - and hence shut it out of public awareness. Those in the know can certainly connect to it by entering the IP address directly, but that address form is a lot harder to remember and propagate than the name. The advent of IPv6 will make this even more of an issue. How I suspect this would actually play out is that the rules around purchasing a domain name, and who can serve on one, would suddenly mushroom. There would be various approval processes, codes of conduct, the ability of the ITU to yank or reassign domain names in an essentially arbitrary process... exactly what you expect to happen when totalitarian and bureaucratic organisations get their hands on a global system.

What really bugs me about the ITU power-grab is that it violates the engineering maxim "if it ain't broke, don't fix it." The only way the Internet is "broken" from the ITU's perspective is that a) America retains an "unhealthy" amount of theoretical power over the infrastructure of the Net, and b) "unaccountable" organisations like the IETF define the standards by which the Net operates. What it can't articulate is how it would "improve" matters in any tangible way other than by giving governments more control over the Net infrastructure - and I would be fascinated to see an elaboration on this point that goes beyond waving around the phrase "democratic accountability" like a dead cat.

2012-11-20

Why is helium now so expensive?

With one of the last remaining Zeppelin operators closing its doors due to rising helium costs, you might wonder why helium is so expensive in the first place when we've been happily filling party balloons with it for the past 15 years at rock-bottom prices.

As you might guess, government is involved:

Though new private helium production plants are set to come online in the coming years —including a Wyoming plant expected to open later this year — private industry hasn't been as interested in producing helium as Congress hoped. Until more companies begin producing helium on their own, consumers are left with spiking prices and tightening supplies.
Under the federal system, those prices are unstable partly because they have less to do with supply and demand than they do about the need for the government, under the U.S. Bureau of Land Management, to pay off the cost of creating the Federal Helium Reserve
Because most of the world's helium is produced in the USA (principally Texas) the US Government gets to set helium prices, and for the past 15 years those prices have been low in an attempt to get rid of the US Federal Helium Reserve - currently around 10bn cubic feet of helium. Now that the US Government has to pay off the multi-$bn cost of creating the reserve in a fairly short time, prices have to spike. Nice one.

I knew helium was produced by uranium and thorium decay, but didn't realise that many natural gas fields contain reclaimable helium. Neither did I realise that escaped helium doesn't just go up to the atmosphere - it is so light that it escapes Earth entirely. So reclaiming it from the atmosphere is out.

For those arguing for greater Government involvement in industrial planning, consider what a mess has been made of an effective near-monopoly position in helium which has no effective substitute in many of its uses.

2012-11-17

Proper engineering: Iron Dome

Ignoring the rights and wrongs of the current Israel-Hamas contretemps, what has been interesting is the performance of Iron Dome, the Israeli counter-rocket missile system.

Iron Dome tackles an odd problem which occurs due to the economic asymmetry between Israel and its guerrilla opponents (here, Hamas). It is relatively cheap for Hamas to acquire rockets, smuggle them into suitable locations and fire them at Israel. The rockets have a substantial terror value; you never know when they will be fired, where they will hit, and there is only 15 seconds or so of warning. If they do hit housing areas then casualties are nearly guaranteed

Iron Dome's two key features are therefore 1) providing a relatively cheap guided missile capable of intercepting and destroying an unguided rocket with a high probability of success, and 2) only aiming at rockets which are likely to endanger civilian areas. 2) is reasonably straight forward - it's just an evolution of counter-battery radar which plots the initial trajectory of a mortar or rocket to locate the firing point. Instead of going backwards, however, it plots the trajectory forwards to identify the probable area of impact, and is programmed with a map of vulnerable versus deserted areas so it knows which areas to defend.

1) is trickier than you might think; it's no good destroying the body of a rocket if the warhead is still mostly intact, since it will continue in its ballistic path and do damage when it lands. There's also a variety of potential rocket targets:

  • Qassam: man-portable, up to 100Kg weight, 20Kg warhead, up to 20km range;
  • Grad: truck-mounted, banks of up to 40 tubes, rockets up to 70Kg weight each with 20Kg warhead and range up to 20km;
  • Fajr-5: truck-mounted, 900Kg rocket with 175Kg warhead, range up to 75Km.
Iron Dome "Tamir" missiles therefore have to be able to disable the largest possible target while still being cheap enough to be a reasonable trade-off for the cheapest missiles (around $800 for a Qassam, presumably plus $200-$500 cost of smuggling it to the right place in Gaza). Tamirs cost around $50K each.

To date in this phase of the conflict, around 800 rockets have been fired by Hamas with a single fatal hit (3 civilians). Iron Dome has been intercepting around 30% of these; presumably 2/3 of all rockets land in unoccupied areas, which doesn't speak well for either their accuracy or skill at aiming.

The economic calculation from the stats we have so far is therefore that firing about 800 Qassams will cost around $800,000. Israel will need to expend 240 Tamirs at $12 million cost, and can expect one hit with a civilian cost of, say $3 million. So the cost ratio is about 16:1. This is not great, but it's certainly far better than previously where no missiles would have been intercepted, 240 missiles could have landed in civilian areas and say 100 of them would cause damage at $50,000 and 10 would have killed 2 civilians so $2 million each - $5 million + $20 million = $25 million. Israel has therefore halved the economic damage of rocket attacks, not to mention nearly eliminating the terror value.

What Hamas has done in this attack is given the Israelis priceless data on real-life volley attacks. Rafael Armaments can continue to tune their algorithm for which rockets to intercept, and improve the performance of Tamirs in effectively wrecking missiles that they intercept. The other issue is that Iron Dome functions as an extremely effective counter-battery locator; the Hamas rocket launching teams will be under rapid attack from artillery and air platforms (drones, Apache or Cobra helicopters) and will have to be lucky to manage a series of attacks without being pounded.

This isn't the end of rocket attacks on Israel. It isn't even the beginning of the end. But maybe it is the end of the beginning.