Monday, December 31, 2007

Physics Feast: Sean Carroll's favorite posts

Sean "Cosmic Varience" Carroll has indexed his personal favorite posts. A true feast for the physics fan.

After the crash of '07: what 2008 may bring

The Economist tells us where we are months after the property crash of '07 (emphases mine):
Economist.com

... On December 18th the European Central Bank lent almost €350 billion ($500 billion) to tide banks over the new year. And yet most fear-meters, including, crucially, the price banks have to pay for funds (see chart), still register chronic anxiety...

... Subprime borrowers will probably default on $200 billion-300 billion of mortgages. That is a lot of money, to be sure, but hardly enough to imperil the world economy. For that, you need the baroque superstructure of mortgage-backed derivatives that enabled investors to bet on the housing market. From a mathematical viewpoint, the combined profits and losses on these derivatives will, by definition, cancel out, so they should not add anything to the total underlying loss. But that is only half the story. Individual investment vehicles may have sustained huge losses, especially if they borrowed heavily: it is the fear that your counterparty might be in that predicament that is gumming up the markets.

... banks now facing up to these contingent liabilities have not had to set aside capital in case of trouble—that gap in the regulations was precisely what made it so attractive to get their investments off the balance sheets in the first place...

... the money-market funds have gone on strike, cutting off the interbank markets' main source of cash ...

... Nobody yet knows whether the extreme borrowing in the credit boom was a sensible result of the powerful new machinery of debt, or the sort of excess still unwinding in Japan...

The markets will not recover until lenders believe the banks have credibly owned up to their losses...

... the frenzy of innovation around debt and securitisation got out of hand. Risk was supposed to be bought by those best able to afford it, but often ended up with those seduced by yields they did not understand. Mathematical brilliance was supposed to model risk with precision, but the models evaporated along with the liquidity that they had failed to quantify. Rating agencies were supposed to serve the market, but their first loyalty seems to have been to the issuers who were paying their fees...

Until that moment, the burden will fall on the central banks. They have tried to help by tinkering with the technical operations that supply liquidity (though they keep overnight interest rates on target by draining money elsewhere)...
The rating agencies failed us in the Enron scandal too, not to mention the last market crash. Maybe we need to oblige rating agencies to make financial bets aligned with their ratings -- so if they rate wrongly they go out of business. Oh, and the CEO's compensation should be aligned with those predictions as well.

The editorial claims that the primary fault with the financial instruments was a failure to model liquidity correctly. If so then increasing liquidity is a case of closing the proverbial gate after the horses have exited. I think DeLong and Krugman have been saying this, but I hadn't understood until now.

A significant part of the screw-up seems to have been that banks were allowed to hide very speculative bets from their balance sheets. Shades of the accounting scandals at Enron and others throughout the 90s! It would be interesting to know who put that loophole in, and how they got paid off for it. Now, to get the system going again, we need banks to reveal their liabilities -- to put this stuff back on the balance sheets.

It will be interesting to watch that fight.

Lastly, the reference to Japan is intended to scare. Japan's real estate bubble collapsed in the '80s -- more than 20 years ago. I don't think it will take 20 years for our real estate to recover, but there is a precedent.

As for recession, I guess we'll find out this year how big the pull is from India and China.

The NYT may be ready to fight

Today's New Years eve editorial suggests they won't be endorsing any of the GOP candidates -- with the possible exception of John McCain ...
Looking at America - New York Times

...There are too many moments these days when we cannot recognize our country. Sunday was one of them, as we read the account in The Times of how men in some of the most trusted posts in the nation plotted to cover up the torture of prisoners by Central Intelligence Agency interrogators by destroying videotapes of their sickening behavior. It was impossible to see the founding principles of the greatest democracy in the contempt these men and their bosses showed for the Constitution, the rule of law and human decency.

It was not the first time in recent years we’ve felt this horror, this sorrowful sense of estrangement, not nearly. This sort of lawless behavior has become standard practice since Sept. 11, 2001.

The country and much of the world was rightly and profoundly frightened by the single-minded hatred and ingenuity displayed by this new enemy. But there is no excuse for how President Bush and his advisers panicked — how they forgot that it is their responsibility to protect American lives and American ideals, that there really is no safety for Americans or their country when those ideals are sacrificed.

Out of panic and ideology, President Bush squandered America’s position of moral and political leadership, swept aside international institutions and treaties, sullied America’s global image, and trampled on the constitutional pillars that have supported our democracy through the most terrifying and challenging times. These policies have fed the world’s anger and alienation and have not made any of us safer...

...We can only hope that this time, unlike 2004, American voters will have the wisdom to grant the awesome powers of the presidency to someone who has the integrity, principle and decency to use them honorably. Then when we look in the mirror as a nation, we will see, once again, the reflection of the United States of America.
I'm hope I'm not being delusional when I say that I see signs of a new attitude at the NYT. A move away from credulous "neutrality" of "he said, she said" towards speaking truth.

It's not just today's editorial. It's in a number of editorials that say things like:
"Today Giuliani/Bush/RomneyHuckabee/etc said .... None of these things are true."
If the NYT is really going to rouse itself from the trap it fell into ten years ago, then there may be hope for America after all.

An amusing summary of the Bush years: records lost in the Executive Building fire

Daily Kos: White House confirms the following records destroyed in the 3rd floor fire. It's a darkly amusing summary of the Bush years and legacy.

Of course it's not over yet. America is perfectly capable of electing any of the GOP candidates.

The great product problem: consumers

One step in my convoluted cell phone Sprint/RAZR -> AT&T/iPhone strategy is getting a "free with contract" AT&T cell phone before I buy the iPhone.

There are two "features" I want in the "free" phone. I want to be able to charge it from a standard USB cable/charger and I want it to have a standard 3.5 mm headset mini-jack. The former means one less charger when I travel. The latter is obviously important.

Any geek would agree these are very valuable features. Note the iPhone fails both tests; though it will at least charge from an USB source and you can modify a mini-jack plug to fit. The evil RAZR fails the mini-jack test and will not charge from a standard USB power source.

So, try to find out what AT&T phones actually charge from a USB power source. I'll wait until you get back ...

Right. You can't find out. I tried the AT&T chat sales support (comes up if you plink around their site long enough) and the rep didn't know. He suggested I try Phone Finder - search database of cell phone specs & features (Phone Scoop).

Right. Nothing there either. (They do have a 3.5 mm mini-jack criteria though!)

Here's the interesting part. This problem isn't just limited to the benighted cell phone world. I find manifestations of this problem in my day job, in the software I buy, and in most of the products I buy.

The problem is there's a big gap between what's important and what sells. Since we live in a world of finite resources, resources are directed to what sells, not to functionality that delivers ongoing value (like one less charger when traveling).

So where's the gap come from?

Consumers.

Humans, in other words.

In an increasingly complex world the African Plains Ape is increasingly adrift, no longer able to make rational buying choices. So the ape buys based on interchangeable face plates rather than USB charging.

So, short of upgrading the APA, what's the best we can hope for?

This is going to hurt.

Apple.

The company that knows what's best for us, and will ram it down our throats until we agree. I think the "strong brand" and "company with a reputation" is about the best we're going to get.

At least until we can start to "print" our own devices ...

Update: PhoneScoop did respond to a request to add the USB power attribute. I'll be impressed if they do that!

Bush didn't want to know the location of the CIA's torture facilities

As long as the GOP held the Senate, the Bush administration could safely destroy the torture tapes, and, more importantly, they could safely lie about their non-existence.

So the decision to destroy them was a reasonable gamble. Rove really expected to hold the Senate. Heck, I though the GOP would hold the Senate. I'm surprised Lieberman still hasn't switched parties, for example.

They didn't, and so we have an investigation. It will, slowly turn up some interesting details. Like this one (emphasis mine):
Tapes by C.I.A. Lived and Died to Save Image - New York Times:

.... interviews with two dozen current and former officials, most of whom would speak about the classified program only on the condition of anonymity, revealed new details about why the tapes were made and then eliminated. Their accounts show how political and legal considerations competed with intelligence concerns in the handling of the tapes.

The discussion about the tapes took place in Congressional briefings and secret deliberations among top White House lawyers, including a meeting in May 2004 just days after photographs of abuse at Abu Ghraib prison in Iraq had reminded the administration of the power of such images. The debate stretched over the tenure of two C.I.A. chiefs and became entangled in a feud between the agency’s top lawyers and its inspector general. The tapes documented a program so closely guarded that President Bush himself had agreed with the advice of intelligence officials that he not be told the locations of the secret C.I.A. prisons.
Bush, of course, wanted plausible deniability, so he could say there were no prisons in Poland, etc. A lie of course, but an easier lie.

As usual the driving force behind the investigation is the cover-up. People tried to conceal the existence of the tapes, and then lied about their destruction. I think many people are wondering if Rove's retirement will turn out to be related to the cover-up process.

There will be other surprises that turn up as the cover-up is investigated...

Sunday, December 30, 2007

The unclear implications of an iPhone contract

The more I look into the implications of the AT&T and Apple iPhone contracting the smellier the whole thing gets.

It doesn't help that Apple is routinely locking discussions about contracts, such as this one: Apple - Support - Discussions - End Of Contract.

On the one hand an Apple thread I initiated concluded that if I were to buy a second iPhone, that I could sell my first iPhone - unlocked. On the other hand Apple has never admitted this publicly -- and they prosecute sellers of unlocked iPhones.

Doesn't quite add up.

Eckels on managing software projects

Bruce Eckels (new feed of mine, so I'm working through the archives [1]) distilled a lot of professional experience into a commencement address [2]. I've emphasized one statement that a less kind person than I would suggest be applied to certain persons with a branding iron ...

The Mythical 5%

... some companies have adopted a policy where at the end of some predetermined period each team evaluates everyone and drops the bottom 10% or 20%. In response to this policy, a smart manager who has a good team hires extra people who can be thrown overboard without damaging the team. I think I know someone to whom this happened at Novell. It's not a good policy; in fact it's abusive and eats away at company morale from within. But it's one of the things you probably didn't learn here, and yet the kind of thing you need to know, even if it seems to have nothing directly to do with programming.

Here's another example: People are going to ask you the shortest possible time it takes to accomplish a particular task. You'll do your best to guess what that is, and they'll assume you can actually do it. What you need to tell them for an estimate like this, and for all your estimates, is that there's a 0% probability that you will actually get it done in that period of time, that such a guess is only the beginning of the probability curve. Each guess needs to be accompanied by such a probability curve, so that all the probabilities combined produce a real curve indicating when the project might likely be done. You can learn more about this by reading a small book called Waltzing with Bears...

I admit, I'd not thought about the inevitable unintended consequence of the "bottom 10%" cuts. Once one person figures out the "hire human sacrifices" strategy everyone will soon learn it, just as ingenious hacks percolate in prisons. Now it's hit on the web, so it's known to the metamind. Humans adapt, and a ruthless corporate culture will breed ruthless employees -- which might not work out as intended.

The rest of the essay is all good advice, most of which I've learned the hard way.

[1] Tip: When you find a good blog, explore the archives deliberately. In bloglines I mark a post as 'keep current' as a reminder that I'm still mining the knowledge. If it's a decent blog there will be far more gold in the best of the archives than in the day-to-day (average) posts of even the very best blogs. (Basic stats - sampling curves.)

[2] I remember when things like this were said once, barely heard, then vanished. Now they live on. I don't think we understand how much difference this makes.

Spolsky tips on travel, gear, phones and so on

Spolsky would be darned annoying if he weren't so generous. Entrepreneur, geek, knows everything ...

Fortunately, he's generous. Here he writes about lessons learned from a promotional tour for his software company - FogBugz. Many of them apply to routine travel (except the first class part). Emphases mine, alas I live in Saint Paul so I cannot escape Northwest.
Joel Spolsky's Travel Survival Guide - Business Travel - Software Demo'ing

  1. We waited until September to start the tour. According to the Bureau of Transportation Statistics, September sees 15% fewer air passengers than August. That may seem like a small difference, but thanks to queuing theory, it's actually significant enough to eliminate most of the lines at airports.

  2. We fly first class...

  3. We scheduled most of the travel in the early afternoon, when airports are relatively quiet...

  4. We never booked a flight until we were sure there was at least one later nonstop flight that would get us where we were going..

  5. Oh, and we refused to fly Northwest Airlines (NYSE:NWA), which routinely schedules more flights than they have the ability to operate.

... we didn't have a single delay of more than an hour, and the longest line I waited in was about 10 minutes for security at Seattle airport.

...Here are a few of my tips for good demos:

  • Ban PowerPoint...
  • ...I try to watch video of myself to learn how to be a better presenter...
  • It's OK to tell jokes...

...In my carryon luggage:

  • My laptop is a Lenovo ThinkPad X61s…

  • To get on the Internet wherever I go, I have a Samsung Blackjack phone on AT&T ... supports GSM, so it works abroad, and HSDPA, for high-speed internet access... I wirelessly connect the laptop to the phone using Bluetooth and get high speed Internet access using my phone's unlimited data plan (this is called "tethering").

  • The projector I use is the NEC NP60 (about $1250)...

  • Because the laptop is usually too far away from the projector for a standard VGA cable, I carry these little gizmos called baluns which let me use a standard CAT-5 LAN cable instead of a VGA cable. I've found that a 25 foot LAN cable is plenty, but the baluns let you go much further if you need to.

  • I also carry my own lavaliere microphones with transmitters and receivers (Sennheiser Evolution G2s) ...

As checked luggage:

  • We had Lands' End Business Outfitters make us up a bunch of piquém polo shirts...

  • We had our printer make up two big professional vertical banners with the FogBugz logo...

Shipped straight to each hotel via UPS ...

  • 12-page 4-color brochures for each attendee
  • Logo pads and pens for each attendee
  • A couple of 25' extension cords, for the projector and the laptop
  • A roll of duct tape to tape wires down on the carpet
  • A couple of hundred "Hello, my name is…" stickers and sharpies so attendees can socialize before and after the event.
I didn't know you could do that with a Blackjack. Imagine if the iPhone were to support Bluetooth tethering...

There's a lot of advice in one place. I liked the "ship extension cords" -- obvious only in retrospect.

Thanks again Joel!

Saturday, December 29, 2007

Music industry sows the wind

In the immortal words of Bugs Bunny: "You realize, of course, this means war" (emphases mine):
Download Uproar: Record Industry Goes After Personal Use - washingtonpost.com

...in an unusual case in which an Arizona recipient of an RIAA letter has fought back in court rather than write a check to avoid hefty legal fees, the industry is taking its argument against music sharing one step further: In legal documents in its federal case against Jeffrey Howell, a Scottsdale, Ariz., man who kept a collection of about 2,000 music recordings on his personal computer, the industry maintains that it is illegal for someone who has legally purchased a CD to transfer that music into his computer.

The industry's lawyer in the case, Ira Schwartz, argues in a brief filed earlier this month that the MP3 files Howell made on his computer from legally bought CDs are "unauthorized copies" of copyrighted recordings....

...At the Thomas trial in Minnesota, Sony BMG's chief of litigation, Jennifer Pariser, testified that "when an individual makes a copy of a song for himself, I suppose we can say he stole a song." Copying a song you bought is "a nice way of saying 'steals just one copy,' " she said...
Meanwhile, in Canada, the music industry proposes to put a rather substantial tax on memory cards -- with an iPod tariff to come.

Fools.

Until now a substantial number of elder geeks (i.e. people with money) have been on the sidelines of the copyright wars. For us the iPod has meant a renewed interest in music, and a steady stream of CD purchases (since we distrust DRM intensely, we don't buy online). We've been on the establishment-friendly side of Pogue's demographic copyright gulf.

But now ... Now the music industry is trying to change the rules of the game.

That's not fair. The Geek Code of Honor requires us to respond by the ancient rule of sheeps and lambs. Some will say the Code obliges us to support the theft of music and video alike, for the criminals have now become the honest outlaw.

The RIAA really shouldn't have crossed this line.

1/2/08: Turns out WaPo got it wrong, but the blogosphere is correcting. It's easy to see why WaPo jumped the gun, the RIAA's legal argument may partly rely on the fact that it has not been shown in US courts that ripping a CD is "fair use". So the RIAA hasn't pulled the trigger yet, but they've pulled back the hammer...

Security theater: airport liquid dumps

A wonderful example of what Bruce Schneier calls "security theater". I admit, I hadn't realized how ridiculously stupid the liquid disposal bins are:
The Airport Security Follies - New York Times Blog


...At every concourse checkpoint you’ll see a bin or barrel brimming with contraband containers taken from passengers for having exceeded the volume limit. Now, the assumption has to be that the materials in those containers are potentially hazardous. If not, why were they seized in the first place? But if so, why are they dumped unceremoniously into the trash? They are not quarantined or handed over to the bomb squad; they are simply thrown away. The agency seems to be saying that it knows these things are harmless. But it’s going to steal them anyway, and either you accept it or you don’t fly...
My experience is that airport security people act as though the feds "jumped the shark" with the liquid ban. Security staff do their best to ignore the whole thing.

I've never been asked to pull out liquids when I forget and leave them in my toiletry kit. On the other hand there's a 40% interception rate when I forget and leave my tiny swiss army knife in my backpack. In both cases I think the contraband shows up on the scanner, but security staff are smarter than their political bosses. They've decided to spend their resources sensibly.

Incidentally, my question for the next candidate "debate" is: "will you eliminate the meaningless "orange alert" status?". We might as well go to a "normal" and "red" alert status, where red means it's time to evacuate the airport and ground the airplanes.

Friday, December 28, 2007

Sudden death in frontotemporal dementia - a clue to sudden infant death syndrome?

A friend of mine recently lost a friend and neighbor to Frontotemporal dementia (Picks disease), a particularly loathsome form of dementia (but aren't they all?) with a 50% occurrence risk in children of affected persons.

He mentioned his friend died fairly suddenly. Puzzling, I thought. Why suddenly?

Turns out that's common in FTD:
[Initial symptoms, survival and causes of death in...[Fortschr Neurol Psychiatr. 2007] - PubMed Result: "Causes of death were varied, but pneumonia and sudden unexplained deaths were particularly frequent."
Hmm. Sudden death. Brain disease. Could there be a clue in FTD to the causes of Sudden Infant Death Syndrome?

Well, there are lots of Google hits on the paired terms, though the associations seem coincidental. On the other hand, a Pubmed search combining both terms had no hits (today). So maybe this research is yet to be published ...

Changing world: the Robosoft example

Earlier today I mentioned the impact of globalization on, among other things, American software engineers.

Which brings me to Robosoft Technologies - Management Team. This is a company based in New Udupi, Karnataka, India. Karnataka is the home of Bangalore.

I came across Robosoft because I was curious about who made the Lego Star Wars II game I bought the kids this Christmas. Lego Star Wars II is one of the very few "universal binary" games available for OS X. I figured this had to come from some US based small Mac specialist company.

Wrong.

The game logo mentions "Feral Entertainment" and "Lucas Arts", but the credits say the development work was done by Robosoft.

It turns out that the CEO, Rohith Bhat, did Mac development for some consulting firms. When he went solo, he carved out a niche in Mac software development - based in India. The firm appears to be doing pretty well. Interestingly three of the executives are Bhats and they rather resemble one another.

New world.

A real problem with CO2 controls

This occurred to me some time ago, but now it's made it into the popular press (emphases mine):
FT.com | Clive Crook's blog: Trade and climate

...Suppose the US adopts a cap-and-trade regime for carbon, as promised by Hillary Clinton, or as envisaged by the Lieberman-Warner Climate Security Act (yes, make this a security issue, why not) currently before Congress. Also suppose that China does nothing to curb its carbon emissions. Then Chinese imports, it will be argued, will have an unfair cost advantage in US markets...
Shame on Crook for not correcting the "unfair cost advantage" error. Trade theory tells us there's no such thing as unfair cost advantage in the naive mercantilist sense. On the other hand, if we were to create a set of carbon tariffs to offset this "cost advantage" then we'd amplify the negative economic effects of necessary carbon emission controls.

The legitimate issue, however, is one of relative socio-economic power. If a President Edwards were to ask Al Gore to lead a US-European world saving CO2 emissions initiative, the US and Europe would sacrifice a certain measure of economic productivity compared to non-compliant nations. That productivity hit translates to a power shift, with all the usual implications.

I think the rich nations will probably have to take the hit and live with it, but we need to recognize it's going to have an effect.

Krugman on globalization: how to manage the losers

Comparative advantage (see also) has been famously described as the one social science proposition that is both true and non-trivial. It's the reason that free trade strongly tends, on average, to be a resilient win-win proposition. I've summarized the thesis as:
... It's not that you can't do the work better than someone else, it's rather that you have better things to do...
The operative constraint here is "on average". A win for America is not necessarily a win for programmers, GPs, or the guys in the parts department. Increasingly the vast majority of the winnings from globalization appear to be going to the top 1%

Krugman last discussed this topic in May of 2007. Now he returns with new data ...
Trouble With Trade - New York Times

... contrary to what people sometimes assert, economic theory says that free trade normally makes a country richer, but it doesn’t say that it’s normally good for everyone. Still, when the effects of third-world exports on U.S. wages first became an issue in the 1990s, a number of economists — myself included — looked at the data and concluded that any negative effects on U.S. wages were modest.

The trouble now is that these effects may no longer be as modest as they were, because imports of manufactured goods from the third world have grown dramatically — from just 2.5 percent of G.D.P. in 1990 to 6 percent in 2006...

... Those who think that globalization is always and everywhere a bad thing are wrong. On the contrary, keeping world markets relatively open is crucial to the hopes of billions of people...

... It’s often claimed that limits on trade benefit only a small number of Americans, while hurting the vast majority. That’s still true of things like the import quota on sugar. But when it comes to manufactured goods, it’s at least arguable that the reverse is true. The highly educated workers who clearly benefit from growing trade with third-world economies are a minority, greatly outnumbered by those who probably lose....

...For the sake of the world as a whole, I hope that we respond to the trouble with trade not by shutting trade down, but by doing things like strengthening the social safety net. But those who are worried about trade have a point, and deserve some respect.

Krugman has been more specific in the past about strengthening the social safety net. Obvious ideas include:
Krugman is a gem. One of the reasons I gave up reading Greg Mankiw is that his ideological blinders prevent him from considering the impact of trade on less competitive workers. Paul Krugman and Brad DeLong don't have that problem.

Update 12/28: Today's Cringely column is sadly relevant.