Odd note in a story on a national artificial joint registry

The NYT joins a call for a national joint registry to find problems sooner.

Makes sense, if the FDA had not been gutted by Bush it would probably have done this years ago.

The interesting bit, however, is the article unconsciously suggests an alternative approach:

The Evidence Gap - Experts Seek a Data Safety Net for Joint Replacements - NYTimes.com

.... The use of joint registries has proven beneficial abroad. In Australia, regulators use such data to force manufacturers to justify why poorly performing hips or knees should remain available, and products have been withdrawn as a result. In Sweden several years ago, surgeons alerted by their national registry stopped using a badly flawed hip long before their American counterparts did. A few medical organizations here, like Kaiser Permanente, operate their own registries to good effect and the Hospital for Special Surgery in New York has recently set up a registry....

If I were the governor of MN, I'd use the Swedish and Kaiser registries. How big a registry do we really need?

The NYT story incidentally illustrates how corrupted we physicians are. The physician profiled received lots of money from the hip replacement vendor. He eventually bit the hand that fed him, but we have lots of experimental data to show that paying "consultants" delays or softens any potential "bites".

In the absence of a national registry, or the use of Swedish and Kaiser data, paying off surgeons is a good investment. Not as great an investment as funding a helpful Senator's reelection, but pretty good.

Three Shtetl-Optimized posts for livening the mind

I really need to study, so I'm noting these for future reading.

These are a set of Aaronson posts, found by tracing back from the latests, that taken together are very good reading in physics and philosophy:

• The relativity of originality
• His review of Wolfram (pdf)
• Newcomb's problem
• PHYS771 Free will

All the essays touch, more or less directly, on the question of whether "free will" (as Scott rigorously defines it) is possible. Note this is distinct from the concept of responsibility -- every reasonable thinker over the past few centuries has understood that "responsibility" is a social construction with no intellectual integrity.

By way of background scientists philosophers have gone back and forth on this since science was more or less born with Francis Bacon. Things looked particularly bad in the 20th century; in the absence of quantum mechanics general relativity seemed to predict a universe frozen in space-time, in which every moment occurred at once and invariably. Maybe this is where Vonnegut's Tralfamadorean determinism came from.

QM seemed to restore free will, but it introduced some disturbing predictions of its own. Physicists attempted to restore the concept of an observer-independent reality through the transactional interpretation, but that did in free will

More recently reality (or realism) seems to be out favor, so maybe free will is back.

Now you're prepped to read Aaronson.

PS. See also - Sean Carroll's favorite posts.

The meaningless beta: Apple and Google

The emerging consensus is that iPhone 2.0 is a big, painful beta program. The real release will be iPhone 2.1.

Truthfully, my main pains with iPhone 2.0 have been the optional MobileMe (even if it worked it would be bad) and missing PDA (Palm) functionality. The rest of it has been good enough, and the apps I've installed have worked. I do reboot nightly.

So I'd say iPhone 2.0 is incomplete, but roadworthy if you're ok with GPS and a faster version of iPhone 1.0 (at a far higher cost for AT&T customers mind you).

MobileMe, on the other hand, reminds me of iMovie '08 and 10.5.0 -- an interesting product born 6 months premature.

By contrast, we have Google. Gmail is still beta (look top left at logo), years after it reached a level of reliability Apple rarely achieves.

I think we need to retire the word "beta". It's become meaningless.

McCain embraces Rove. Of course.

The only surprise is that it took him this long.

Editorial - Low-Road Express - Editorial - NYTimes.com

On July 3, news reports said Senator John McCain, worried that he might lose the election before it truly started, opened his doors to disciples of Karl Rove from the 2004 campaign and the Bush White House. Less than a month later, the results are on full display. The candidate who started out talking about high-minded, civil debate has wholeheartedly adopted Mr. Rove’s low-minded and uncivil playbook...

Obama is a Chicago pol, so this can't be unexpected. We know the low road works on Americans.

I expect the GOP to win. I suppose it's the result of 8 years of Bush -- I really can't imagine a better future.

Doesn't matter, I'll fight for Obama anyway. Miracles happen- at least in politics.

Update: There's a risk to playing this game of course. It legitimizes a strategy of pointing out that McCain may have pre-dementia. This ploy has the advantage of possibly being true.

FailMe: a classy approach to getting even

Apple - FailMe is a spoof of Apple's MobileMess with the perfect combination of humor and angry bitchiness.

We've come a long way from sending snail mail complaints to the CEO.

I love it. They'll have time to build out more content -- it will be months before MobileMe is reliable (assuming Apple can do reliability, which is unclear), and a year before it's a real competitor to Google Apps.

The problem with my 'brain and mind' tag - the mind is not in the brain

Blogger supports tagging of posts, though they're called "labels". One of the tags for this blog is Brain and mind.

There's a problem with that tag. No, not the boring old "mind/body" duality stuff. I'm a pragmatic materialist 2.0 (ie reality + emergence), for me the "mind" more or less runs on the body [1].

The problem came to me as I thought about a post I wrote a week or two ago on current models of chronic pain syndromes ...

Gordon's Notes: The pain is all in your head

... The ideas aren't quite as novel as Gawande suggests. I recall fifteen years ago veteran physicians, with lots of experience with intractable pain and chronic fatigue, had begun to think the problems were 'all in the patient's head'. By which we meant, with intentional irony, that the problem was 'malwiring' of the brain.

The good news is, the brain is plastic. We can't easily alter it directly, but we can slowly reprogram it through the mind. That's how the mirror-box therapies Gawande describes work, and presumably that's how exercise therapy works for chronic fatigue syndrome (albeit both imperfectly)...

Ok, I have to also thank my son, who has an extremely tight connection between psyche and soma. I watched a recent shoulder problem wax and wane in proportion to psychic stress, and I realized what's wrong with both my tag and my prior post.

I have too strong a division in my own head between the central and peripheral nervous system. Yeah, sure, everything is connected to everything else so we do need to draw lines, but I think for the purposes of modeling disorders of sensation and perception, including pain, the line should be drawn around the peripheral nervous system -- not around the central nervous system (and not around the body -- that's too broad to be conceptually useful).

Perception and sensation are core functions of mind, and we physicians may err in ascribing them primarily to the periphery or the core.

Yes, I know this seems self-evident when I put it this way. Maybe it is, but I think there's something here. If we truly believed in this model, I think we would approach all management of sensation, whether arising in a broken leg (peripheral nervous system), or my son's sore shoulder (central and peripheral) or intractable itching (central) with an eye to techniques applied both peripherally (set the leg) and centrally (??).

There's something here ... I just need to think about it a bit more.

[1] More or less, without the "oxygen" of social interaction and coherent sensory input it won't run well for long.

Why Apple can't do the Cloud: secrecy is the enemy

I've been forced to stare deep into the bowels of Apple's MobileMess, Outlook synchronization, and oddball .Mac conversion disasters.

This stuff is extremely hard, but Apple has brains and money. What in their culture caused such a massive screw-up?

The problem, I think, is secrecy -- an obsession with surprises that comes partly from their CEO (though if I read one more article claiming Jobs is Apple my brain will explode), partly from their history, and largely from an insanely successful marketing strategy.

The complexity of changing a densely interconnected system like .Mac to a very different system like MobileMe requires months of public beta before launch. There can't be any surprises, there has to be high reliability.

Google gets this. Microsoft gets this.

Apple doesn't.

Apple won't be able to compete in the (cursed) Cloud if they can't kill their secrecy demon.

Apple begins the MobileMe recovery

It only took the NYT and the squeaking of thousands of customer mice, but Apple has launched @ daily MobileMe status reports:

Apple - MobileMe - Status

... One issue we encountered was a mail outage affecting 1% of our members. Last Friday a serious problem with one of our mail servers blocked those members’ access to their MobileMe mail accounts. As of today a team was able to restore limited web access to those accounts so the affected members can use their browsers to read mail that has arrived since last Friday (though not before) as well as send and receive new mail. The team has already begun rolling out restoration of full access for all the accounts and expect to finish by the end of next week. We particularly regret to report the loss in the affected accounts of approximately 10% of the messages received between July 16 and July 18.

.... fixed over 70 bugs including one that was preventing MobileMe IMAP mail folders from syncing correctly between the web app and Mac OS X Mail or Outlook, plus others correcting display issues in Calendar and in general enhancing the performance of our web apps...

Software-as-service and DRM mean you don't own. You rent. Everything. Another lesson from Yahoo.

This would be the third time I recall that a major vendor has shut down a DRM service and stripped customers of all their products.

AppleInsider | Yahoo! Music's death at age 3 warns of DRM's risk

... Yahoo did its best to stage a rival to Apple Inc.'s iTunes, but after three years of lagging results, the Internet icon is putting its Yahoo! Music service to rest and leaving subscribers with copy-protected music libraries that can't be transferred to new computers...

Due the vagaries of computer life, within a year much of that music will be gone. Yahoo is telling users to burn CDs from the music. Anyone who's ever tried to do this will know what an inane idea that is. It's prohibitively time consuming, and future lossy compression of that music will generally produce awful results.

When Microsoft/MSN (? or was it AOL?) did something similar I think they refunded customer money, though that only works for people with current accounts.

They key lesson is that when you buy a used CD for $3 you have access to that material for an unlimited amount of time. When you buy the same CD new on iTunes for $14 you have use until Apple closes its FairPlay servers, or until it changes your iTunes contract.

We live in an age of transience. I suspect a younger generation will simply accept this as the way things are.

Incidentally, there's a cruel surprise slowly being uncovered. A surprise, that is, to the vast majority of people who don't bother thinking about DRM.

Lots of families are going to have multiple iPhones (great phone, fascinating computer, lousy PDA, Outlook sync broken, don't touch MobileMe before November, wait for 2.1 if you can).

They'll expect they can sync all their iPhones to what they think of as the family music and video library.

Cue evil laughter.

They'll discover then that an iPhone is a personal device, and it must sync to an individual user account. They will also discover that Apple's DRMd music and videos are owned by an Apple username, not a family. Lastly, they'll discover that iTunes libraries are personal libraries, not family libraries.

Slowly they'll realize the jaws are closing around them. They need to buy a copy of each video and song for each member of the family. [1] Eventually, they'll see the shape of a BrainLocked future, where we pay to keep access to our own memories...

[1] There used to be a workaround for non-DRMd iTunes media, but I've not tested it on iTunes 7.7. Sooner or later Apple will close the door on this; my transient DRM optimism has faded. I don't think Americans are going to figure this one out. Maybe the EUs will twig to this, and put some serious laws in place.

Update: Recently Apple terminated its .Mac web page authoring tools. All .Mac web pages are now inaccessible. For a scary moment I thought Google had done the same thing with my old Google Pages. Turns out they're only close to gone. Dang, but I sure as shootin' don't trust that cloud.

Strangely met in night: a blog I’ll never read …

A google search took me to a caustic discussion of a Hofstadter book:

The New Skeptic: I Am a Strange Loop: Gödel's Loop

…Continuing my review of I Am a Strange Loop, today I get to tackle metamathematics. Hofstadter tackles it too, and finds it rich in philosophic insight. Strangely rich, actually.

I suppose I ought to explain who Kurt Gödel is and why he is a hero of many, many nerds today (I am among those ranks). And that tale doesn't start with Gödel, so stay patient while I explain the background…

Caustic and opinionated, but interesting. I started to look at the archives. Should I grab this feed?

Then I saw the “links of note”:  national review, weekly standard, rush… savage … beck …

On closer inspection, the stranger in the night wears a necklace of human noses.

Backing slowly away …

Thank heavens for the link list – who knows what horrors I might have been exposed to!

Bayes theorem and the anthropic principle

Years ago I used to teach Bayes theorem to informatics grad students.

I was reassigned to other lectures though. Truth is, I had a hard time focusing on the boring stuff we were doing with Bayes. It just seemed like there was a deep weirdness about the Bayesian model of probability, but I couldn't quite put my finger on it. I was sure someone who understood it deeply would justify my queasiness.

Since then there's been a bit of a renaissance in thinking about Bayes. In Our Time even did a recent programme with a good bit of Bayes. Physicists are all over Bayes these days, and the Bayesian vs. Frequentist combat is out in the open (I knew this stuff was weird).

These days, I could assign students this essay to read:

PHYS771 Lecture 17: Fun With the Anthropic Principle

... So if Bayes' Theorem seems unobjectionable, then I want to make you feel queasy about it. That's my goal. The way to do that is to take the theorem very, very seriously as an account of how we should reason about the state of the world...

Of course that assignment might also shrink the class size ...

Apple fails the crisis test as MobileMess blows up

David Pogue is a sympathetic journalist, but even he has lost patience with Apple’s MobileMess flop. I hope Microsoft gets some kicks in, Apple clearly needs more pain.

In addition to seeing Apple’s executive failures in motion, we more education on how disastrous synchronization bugs can be.

Apple’s MobileMess - Pogue’s Posts New York Times Blog

… a strange note showed up on the MobileMe support Web site: “1% of MobileMe members cannot access MobileMe Mail. We apologize for this service interruption and are working hard to resolve the problem.”

Now, even if 1 percent is accurate, Apple has 2 million .Mac/MobileMe customers. So that’s at least 20,000 people…

… For most of them, the e-mail features of MobileMe just don’t work. The online Mail program at Me.com shows up empty; mail you try to send from your e-mail program never goes out; and messages sent to you get bounced.

For a few, it’s a lot worse. “This morning, I woke up and turned on my computer,” wrote one reader. “Happily, it seemed that the MobileMe e-mail service was back up. However, a few seconds later, when my computer synced with .Mac/MobileMe, ALL of my e-mail — every single e-mail I’ve ever sent, received, and filed on .Mac — disappeared. Every e-mail file on my hard drive (in the Mail library) was gone. I immediately went to Me.com to make sure that all my e-mail was still saved to Apple’s server. It wasn’t. All of the mail was gone.”

Apple escalated her case and dedicated top technicians to it, for which she was grateful. In the end, however, they recovered only 43 messages. The rest are gone forever…

… MobileMe tech support, my correspondents tell me, is nearly impossible to reach; the recording says that the support team is “unavailable due to the overwhelming interest in MobileMe.” (Somehow I doubt that “overwhelming interest” is the problem.) When you do reach them, they’re apologetic but can do nothing to help.

…the real problem is how Apple is responding. For a company that’s so brilliant at marketing, it seems to have absolutely no clue about crisis management…

…This is an airplane that’s stuck on the runway for hours with no food or working bathroom. And the pilot doesn’t come on the P.A. system to tell the customers what the problem is, what’s being done to fix it, how much longer they might be stuck, and how he empathizes with their plight. Instead, he comes on once every three hours to repeat the same thing: “We apologize for the inconvenience.”…

I wonder if people who’ve lost all their email have grounds for litigation. A nice class action suit might concentrate Apple’s mind. I also wonder how many of the victims are running an older version of OS X desktop (ex. 10.4).

Unlike Apple’s victims, I have backups, and I haven’t dared active Mail.app email synchronization. My trial account simply forwards any MobileMe email I get.

Apple needs to grovel, and they’re not good at groveling.

Right wing hack hit and run - caught by a bicyclist

Robert Novak, a cruel GOP hack, runs down a pedestrian and takes off.

That's fine of of course -- nobody should be walking. Problem is, he gets caught. Worse problem, the victim is a K street lawyer. Humiliation is, Novak's car is stopped by a bicyclist.

It's hard to imagine greater GOP humiliation than being caught in the course of crushing the weak by a bicyclist -- who's also an attorney. Emphases mine.

Novak cited after hitting pedestrian - Jonathan Martin and Chris Frates - Politico.com

...The bicyclist was David Bono, a partner at Harkins Cunningham, who was on his usual bike commute to work at 1700 K St. N.W. when he witnessed the accident.

As he traveled east on K Street, crossing 18th, Bono said "a black Corvette convertible with top closed plows into the guy. The guy is sort of splayed into the windshield.”

Bono said that the pedestrian, who was crossing the street on a "Walk" signal and was in the crosswalk, rolled off the windshield and that Novak then made a right into the service lane of K Street. “This car is speeding away. What’s going through my mind is, you just can’t hit a pedestrian and drive away,” Bono said.

He said he chased Novak half a block down K Street, finally caught up with him and then put his bike in front of the car to block it and called 911. Traffic immediately backed up, horns blaring, until commuters behind Novak backed up so he could pull over.

Bono said that throughout, Novak "keeps trying to get away. He keeps trying to go.” He said he vaguely recognized the longtime political reporter and columnist as a news personality but could not precisely place him.

Finally, Bono said, Novak put his head out the window of his car and motioned him over. Bono said he told him that you can't hit a pedestrian and just drive away. He quoted Novak as responding: “I didn’t see him there.”...

Novak is wealthy, but it's bad luck to run over a lawyer on K street and be witnessed by another lawyer. His best defense will be to plead dementia. He's on record as hating pedestrians, I bet he's not to keen on bicyclists either.

Update: Novak has a brain tumor. Which may explain a lot about why he hit a pedestrian, and why he behaved irrationally afterwards.

Data Lock: 661 MORE 3.1 documents keep me in 10.4

My guess is that MobileMe and iPhone 2.0 will never work properly with OS X 10.4.

So I need to upgrade my iMac to 10.5. I'm a little apprehensive because this is a PPC machine and I'm not sure how well behaved 10.5 is on PPC. Still it works well on my MacBook. Anyway that's not the real problem.

The real problem is that my ancient copy of MoRu tells me I've 661 MORE 3.1 documents on this drive. MORE 3.1 needs MacOS 9 Classic, and 10.5 doesn't run classic.

I looked at a few of my old files. There's a lot of knowledge in there I don't want to lose.

Inspiration and OmniOutliner Pro will open these as outlines, but both will lose presentation graphics. Brad Pettit's free XML converter will switch the files to plain text XML, and I think it might be able to process multiple files at once. Otherwise I can open each one and save it to another obsolete file format, or I can use CUPS-PDF to create a PDF output classic can see.

This is going to hurt.

Which is why I use Nisus Writer Pro as my word processor. The file format is RTF, and Word can read it. I'll transition to an Open Document Format in a year or two.

This is why I'm averse to adopting Evernote until they have an export tool.

Try not to avoid these traps!

Our communication and computing costs are comparable to our gasoline costs

A few people have commented on a blip in gas prices over the past few months.

Much is made of how gas prices are becoming a significant portion of many people's budget, reducing money for food, medicine and entertainment.

Less mention is made of rising communication and computing costs.

We recently did a quick back-of-the-envelope comparison. The sum of our internet, mobile, and landline costs is pretty impressive, and the mobile costs are going to take another big jump.

If we add in purchasing a $1,000 to $3,000 of hardware every year (average), plus software, services (MobileMess and probably 5-7 other recurrent service bills) and infrastructure costs, I suspect our communication and computing budget is larger than our gasoline budget (though smaller than our total transportation costs).

My sense is these costs are rising at least as fast as the cost of gasoline. Sure, so are the capabilities, but in general the capabilities don't have direct revenue attached.

It puts the gas price rises in a slightly different light -- there are many new expenses competing for the modern household budget.