15 cigarettes yields 1 mutation - the new world of cancer genetics

I remember when "oncogenes" were on the cover of TIME. They were the key to understanding malignancy, and would change the ballgame. That was about 30 years ago.

Now that handful of oncogenes have become tens of thousands of mutations ...

BBC News - Scientists crack 'entire genetic code' of cancer

... scientists found the DNA code for a skin cancer called melanoma contained more than 30,000 errors almost entirely caused by too much sun exposure.

The lung cancer DNA code had more than 23,000 errors largely triggered by cigarette smoke exposure.

From this, the experts estimate a typical smoker acquires one new mutation for every 15 cigarettes they smoke.

Although many of these mutations will be harmless, some will trigger cancer...

We used to think a cancer involved a few mutations. Maybe two or three. Not 23,000.

It takes an astounding number of mutations to knock off the systems that prevent cancer -- while managing not to kill the cell.

Among other things we now have a good explanation of why cancer risk falls after someone stops smoking. It appears that the mutated cells are replaced by healthy cells. If they're lucky the bad ones die off before one goes rogue.

So it's never too late to stop - or to get out of the sun.

--
My Google Reader Shared items (feed)

The most marvelous world of the virus

A superb essay on the virus ...

A Gazillion Tiny Avatars - Olivia Judson - NYTimes.com

.... whether you count viruses as living or not, there’s an awfully large number of them: a single drop of seawater may contain more than 10 million viral particles. That’s more than 10 billion in a liter (two-and-a-bit pints) of ocean. Some people have estimated that, in the oceans, there’s more carbon stashed away in viruses than there would be in 75 million blue whales.

Moreover, viruses are extremely diverse; there are zillions of different kinds. Some, such as MS2, a virus that attacks bacteria like Escherichia coli, have as few as four genes. Others, such as the gargantuan Mimivirus, have more than 900. (Mimivirus mostly attacks amoebae, although it is also suspected of occasionally causing pneumonia in humans.) And each time we look in a new place, we find more and more viruses that are different from those we have known before.

Fortunately for us, most viruses don’t attack humans; they attack bacteria and other microbes, which they kill on a colossal scale. In the oceans alone, viruses are reckoned to kill about 100 million metric-tons’-worth of microbes every minute.

.... viruses play a fundamental role in regulating the food chain. This is because death-by-virus is different from death-by-predator. When a predator kills a microbe, it consumes it: the microbe’s cell is incorporated into the predator’s body. In contrast, when a virus kills a microbe, the microbe’s cell bursts open, or “lyses,” releasing new viruses and a lot of cellular debris back into the environment. This debris can then be consumed by other microbes. In other words, by lysing their victims, viruses are constantly making food available to other life forms...

So do bacteria have a fundamentally different relationship to viruses than multicellular organisms? Why are they so much more lethal to bacteria than to us? Did the way our DNA propagates facilitate a "truce" with viruses?
--
My Google Reader Shared items (feed)

Never a good feeling – an attack on my Google account

Someone just made 3 attempts to reset my Google Password. The reset notice I received includes this statement …

… If you've received this mail in error, it's likely that another user entered
your email address by mistake while trying to reset a password. If you didn't
initiate the request, you don't need to take any further action and can safely
disregard this email….

A mistake. Suurre it’s a mistake.

I have a robust Google password, but the risk here is that someone has access to a secondary account that receives my Google password reset requests. Those have robust passwords too, but there are always weaknesses.

Just to be on the safe side I’ve reviewed my Google accounts password recovery options and they look good.

Brrr. I hate passwords. I’d have bet good money in 1996 that we’d have robust biometric authentication by now. I’d have lost every penny. A good lesson about predicting the future.

Update 11/18/09: Amit Agarwal was hacked around the same time I was attacked. It's not clear how they hacked in.

The ultimate climate conspiracy …

If I were an alien entity observing the earth, and I wanted to test humanity to the breaking point, I’d come up with a scheme that required China, India, America, Canada, Australia and the rest of the world to come together to solve a huge problem with uncertain consequences that unfolds relatively slowly and requires painful action from everyone on a time scale of years.

A trans-galactic gambling scheme? An alien art form?

Cue twilight zone music.

The common core of human language – as shown in speech recognition systems

Just one phrase in a wonder filled post on Google’s new Japanese speech recognition system …

…speech recognition systems are surprisingly similar across different languages…

I bet some Google researcher has a multi-axial plot of the speech recognition attributes of the languages they work with. That will be a great graphic one day soon.

The essay is required reading. How the hell does anyone learn to write Japanese? Yes, I know people do it, but, really, how?

Most of all, this essay is a small measure of what Google does, and why I swear allegiance to the House of Google (3 on Gordon’s scale of evil). These are gray days in America, but we will return …

Understanding secure systems: The Chromium extension example

This very brief Google Chromium blog posting gives a lovely view into modern secure system design ...

Chromium Blog: Security in Depth: The Extension System

... To help protect against vulnerabilities in benign-but-buggy extensions, we employ the time-tested principles of least privilege and privilege separation...

The original has wikipedia* links to relevant articles. These principles are broader than computer security. Think of them when you provide access to your Facebook information.

"Least privilege" and "Privilege Separation" should be a part of grade school and high school curriculum.

If you want lots more detail, the authors refer us to their academic treatise on securing browser extensions.

I love blogs.

*Yeah, Knol was a bad idea.
--
My Google Reader Shared items (feed)

Lazy journalism and the both sides fallacy – Ed Lotterman edition

A classic example of the lazy journalism of false equivalency …

Edward Lotterman – Real World Economics – Pioneer Press (TwinCities.com)

… Unlike in most other industrialized nations, U.S. citizens remain divided on whether climate change is really occurring. Indeed, the proportion that is skeptical is growing rather than shrinking…

…This is not a lack of consensus, but rather a fundamental division that is not likely to be solved in the foreseeable future. For significant portions of both camps, it has become a matter of faith rather than reason…

When one camp is aligned with the overwhelming majority of the peer reviewed and respected scientific literature, and the other camp is not, this is not a “matter of faith rather than reason”.

One camp is on the side of reason, the other camp is faith-based.

This is, at best, a lazy invocation of the easy cliché. Most likely, it’s intellectual cowardice.

The historic pricing of an Ella Fitzgerald CD set

This range of prices for Twelve Nights In Hollywood: Ella Fitzgerald feels historic ...

List price CD: $70
Amazon CD: $56
iTunes AAC (256 kpbs, AAC encoded*): $40
Amazon MP3 (256 kpbs, LAME encoded): $34.31 (why the 31 cents?)

The Amazon MP3 is less than half the cost of list price CD.

I'd like a physical CD for Emily's gift, but at this price I'll burn a single representative sample from 50 song collection and put the entire set on her iPhone Christmas eve.

Oh, and the Amazon CD is "temporarily out of stock" anyway.

CD R.I.P. We may miss you yet.

--
My Google Reader Shared items (feed)

Gordon's scale of corporate evil - 1st edition

Top end of the scale is 15. It's a linear scale.

1. Philip Morris: 15
2. Exxon: 13 (see link to #1)
3. Goldman Sachs: 12
4. Facebook: 12
5. For profit health insurance companies: 11
6. AT&T and Verizon (tied): 10
   
7. Microsoft: 10
   
8. Average publicly traded company: 8
9. Google: 6 (revised up after the Google Buzz fiasco, then down when they showed some wisdom)
10. Apple: 5
11. CARE International: 1 (They're not a PTC, so this is merely a non-evil reference point)

What's your ranking?

Update 12/15/09: I added Exxon thanks to a comment and because of the Philip Morris synergy. Exxon's astroturf climate change denialism (see also) campaign puts them in contention for the most evil publicly traded company of the modern era.

Update 1/6/10: Both Google and Facebook moved one notch up the evil scale. Google because of their arrogant, haphazard and uncaring Pages to Sites migration and Facebook because they sold their users out to their often crooked "Apps" vendors. Facebook is now more evil than Microsoft, and Google is tied with Apple.

Update 2/16/10: Google had dropped to '3' after unblocking China, but then leaps to '8' after the Google Buzz fiasco.

--
My Google Reader Shared items (feed)

It's not AT&T's fault, it's the iPhone?

My gut finds this persuasive ...

Digital Domain - AT&T Takes the Fall for the iPhone’s Glitches - NYTimes.com

... When I set about looking for independent data, however, to confirm the superior performance of Verizon’s network, I was astonished to discover that I had managed to get things exactly wrong. Despite the well-publicized problems in New York and San Francisco, AT&T seems to have the superior network nationwide.

And the iPhone itself may not be so great after all. Its design is contributing to performance problems.

Roger Entner, senior vice president for telecommunications research at Nielsen, said the iPhone’s “air interface,” the electronics in the phone that connect it to the cell towers, had shortcomings that “affect both voice and data.” He said that in the eyes of the consumer, “the iPhone has the nimbus of infallibility, ergo, it’s AT&T’s fault.” AT&T does not publicly defend itself because it will not criticize Apple under any circumstances, he said. AT&T and Apple both declined to comment on Mr. Entner’s assessments.

Neither AT&T nor Verizon was willing to reveal its internal data on performance. But Global Wireless Solutions, one of the third-party services that run network tests for the major carriers, shared some of its current findings. The service dispatches drivers across the country with phones and laptops equipped with data cards. They have covered more than three million miles of roads this year, while running almost two million wireless data sessions and placing more than three million voice calls, said Paul Carter, the president.

The results place AT&T’s data network not just on top, but well ahead of everyone else. “AT&T’s data throughput is 40 to 50 percent higher than the competition, including Verizon,” Mr. Carter said. AT&T is a client and Verizon is not, he added.

Why do I find this persuasive, even though one of the sources gets AT&T money?

1. We only hear my fellow iPhone users screaming about AT&T quality.
2. Remember Apple's rivals saying Apple didn't have the engineering background to make a quality cell phone? I suspect this is what they were talking about. Apple did amazingly well, but perfection is not human.
3. Quality and reliability are not Apple's top priority (most recent example: my 2 day old flickering, stuttering, $2K iMac i5). It's not in their DNA.
   

Mind you, I despise AT&T. I think they'll shaft their customers whenever they can get away with it. Apple is flawed, but they're still better than everybody else. It's just that this time, when it comes to phone service, I suspect Apple is at least as flawed as AT&T.

Update 12/13/09: If the iPhone does have technical limitations that cause connection issues, is this why AT&T has not allowed tethering?

Update 12/14/09: Two rebuttals from Gruber: One, Two. The second points to Pete Mortensen, who shows the form of the question changes the answers.

A really tough cop ...

This is impressive ...

In Holiday Crush, a Fatal Shootout in Times Sq. - NYTimes.com

... The video showed Mr. Martinez turning, the police said, but he moved out of camera range. Police officials, who did not immediately release the video, said it also showed the sergeant reaching for his weapon and raising it.

It also showed Sergeant Newsom, who has been on the force for 17 years, raising his left arm over his chest in hopes of protecting his heart. It is a defensive move rookies are taught in the Police Academy. Police officials were astonished that the sergeant, less than 15 feet from the stubby barrel of a semiautomatic weapon with no hope of taking cover, was cool-headed enough to remember to do so.

Police Commissioner Raymond W. Kelly said Mr. Martinez fired first, getting off two shots. Then his gun jammed.

Mr. Kelly said the sergeant fired four shots. All four hit Mr. Martinez: in the chest, below the neck and in the left arm; he also suffered a graze wound to the right arm...

--
My Google Reader Shared items (feed)

Kurzweil and homeopathy: now crackpot certified

Ray Kurzweil made a bundle on speech recognition and speech related software. He's also done some good work with assistive technologies.

In later life he wandered off into the cultish fringes of the Singularity. He's seemed less anchored to reality in recent years, and now he's nuked the fridge (emphases mine) ...

Amazon.com: Transcend: Nine Steps to Living Well Forever (9781605299563): Ray Kurzweil Ph.D., Terry Grossman M.D.: Books

According to futurist Kurzweil (The Singularity is Near) and homeopathic medical doctor Grossman (The Baby Boomer's Guide to Living Forever), medicine is transforming into an information technology, which by its nature advances at an exponential rate..

Homeopathy is fundamentalist crackpottery (and "homeopathic medical" is an oxymoron). Kurzweil has really lost it.

--
My Google Reader Shared items (feed)

The state of social networks: Facebook, Twitter and one other

I’ve interacted with three social network systems. Two get a lot of media attention, but the most interesting one is currently invisible. Here’s how they look to me in December 2009.

Facebook is currently useful, but worrisome. It's useful because it’s been a useful way for me to share family news that (very) close friends may enjoy seeing. Facebook is where I announce that my 12 yo scored a goal in a hockey game, and where my readers would understand that a goal is not always just a goal. Facebook is worrisome because their business model is currently based on exploiting the weakest members of a social graph, and then on selling information and marketing access to the entire graph. It’s remarkable how uninteresting Facebook’s ads are. That’s a bad sign, even thought it says something interesting about the limited predictive value of one's friendships.

Twitter would be more interesting if I had an archaic cell phone with unlimited text messaging, or if I had an interest in the domestic disturbances of celebrities. Twitter’s usage is on the same track as Friendster and MySpace. I don’t think it will last very much longer.

The most interesting social network I know of, however, is one that has very few members, no media coverage, no books, little documentation, no clear strategy, and mysterious privacy and revenue models. This is the Google Reader share and comment graph including the (currently) “like-based” discovery model.

Through the Google Reader (GR) “like” link I’ve identified about six English-language writers around the world who share an interest in topics I want to know about. When I find one who is sharing interesting items I don’t see or know about, I add them to my GR graph. They may choose to follow me or not – that’s not relevant to me. The value is that I can follow what they do.

I add one such meta-feed to my knowledge stream every few weeks. The stream volume does not increase much, because I can in turn drop direct reads of streams my experts cover. Given the uber-geekiness of the GR graph membership the quality of shared items is currently very high, but I don’t see why this approach won’t scale even in the event that the GR graph gets market attention.

The primary risk to this model, of course, is that Google will lose interest. I suspect, however, that this experiment will provide Google with interesting ways to explore and classify the world’s information stream – a mission very dear to their revenue model.

Google’s machine translation is improving every month – I’m looking for my first Chinese-language source. That will be interesting.

The GR graph means Google wins and I win. Maybe, if this increases the value of the world’s knowledge stream, we all win.

I like that model.

See also

• Gordon's Notes- Google reader “like” and the shared discovery process
• Gordon's Tech- Google Reader- Experiments with notes, following and sharing
• Gordon's Notes- Google reader micro-blogging and changes to Gordon Notes
• Gordon's Tech- More of me- My Google Reader Shared Item Feed’
• Official Google Reader Blog- Share anything. Anytime. Anywhere

Update: Of course the day I post this is the day it seems to stop working. I am following about 17 people, but nothing happens if I try to add someone new. Google has been doing something funky with sharing permissions; it's possible that when I "follow" someone they have to approve before anything happens. So it's now more of a "request to follow".

Financial Times – the feeds and the Fail

I’ve finally given in to DeLong’s imprecations and added some Financial Times sources to my feeds. The FT sources ought to complement my much appreciated Guardian feeds.

I do get one feed from the Murdoch paper – the WSJ’s Health blog is actually pretty good. Otherwise, unlike Brad who seems unable to stay away, I ignore the WSJ.

Here’s the set I’m starting with, I’ll tweak it up and down over time.

• Markets main
• Africa
• World main
• Global Economy
• Asia-Pacific
• Comment main
• FT Editor’s blog
• Tech blog
• Science blog
• China main
• India main
• Personal Finance main

So what will I drop? Probably some of the Economist’s feeds – that mag seems to be continuing its downward course. I expect Murdoch to buy it any time now.

PS. Google Reader's "Add to Folder" select menu doesn't scale. At least give it a scroll bar! As an interim measure I've deleted all my "Google Reader" tags. The Reader team really messed up the folder/tag metaphor.

Update 12/5/09: Ok, that was a Fail. The FT allows only a small number of free article views a month and the subscription fees I was shown when registering was about $200 a year - for electronic access alone. That was bad, but I might have considered it -- except a feed link takes me to a view that's incompatible with a mobile client. Delete all.

Singular fun with Fermi

I'm a geek. So I love to puzzle with the Fermi Paradox, and, like many fellow geeks, I'm tempted to connect the Great Silence to the "Singularity" (aka, the rapture of the geeks). That puts me on the pessimistic side of the Singularity cult -- as in "Hello Hal, Goodbye humanity".

Now since we're talking about the extinction of humanity within 40 - 300 years (I go for about 80-100 myself) you might think this would be a bit depressing. Well, it might be, except I've long known I'll be dead before 2070 and probably before 2050. Everyone I care about will be dead within about 110 years. These are the things we secular humanist types know, and yet we can be quite cheerful. Ok, not in my case. Less dour maybe.

The peri-Singular death of humanity is a serious matter for humans, but it's less inevitable than our personal exits, so by comparison the Fermi Paradox/Singularity schtick is more entertaining than grim. That's why I appreciated this comment on a recent post (edits and emphases mine, follow link for full text) ...

Comment by Augustine 11/29/09

... I don't trust predictions that are based on extrapolations from current rates of growth. These predictions are, and will be, correct, but only for limited time frames. Extend them out too far and they become absurd. Moore's Law works fine, and will continue to work fine for a while I’m sure, but basing predictions on ever accelerating computing power is about as useful as imagining accelerating a given mass to the speed of light.

The greater problem, however, with the argument lies in the fact that we are at best imperfect predictors ... You cannot accurately infer a future singularity when you cannot know what will change the game before it happens, if you get my drift...

There's more to the post, but I'll stick with these two questions. The "limits to exponential growth" argument is even stronger than stated here since, in fact, Moore's Law itself has already failed. We have some more doublings to go, but each one is taking longer than the last.

So maybe we'll never have the technology to make a super-human AI. I think we'll make at least a human-class AI, if only because we've made billions of human-level DI (DNA-Intelligences). Even if computers only get five more doublings in, I think we'll figure a way to cobble something together that merits legal protection, a vote, and universal healthcare. (Ok, so the AI will come sooner than universal healthcare.)

So we get our AI, and IT's very smart, but it's comprehensible (Aaronson put this well). So this is certainly disruptive, but it's no singularity. On the other circuit, it does seem odd that today's average human would represent the pinnacle of cognition. Our brains are really crappy. Sure the associative cortices are neat, but the I/O channels are pathetic. A vast torrent of data washes out of our retina -- and turns into hugely compressed lossy throughput along a clogged input channel. We can barely juggle five disparate concepts in working memory. Surely we can improve on that!

So I'm afraid that Newton/Einstein/Feynman class minds do not represent a physical pinnacle of cognition. We'll most likely get something at least 10 times smarter. Something that makes things even smarter and faster, than can continuously improve and extend cognitive abilities until we start to approach physical limits of computation. Before that though, the earth has been turned into "computronium" -- and my atoms are somewhere in orbit.

As to the second objection, that we can't imagine a singularity because we can only reason within the system we know, I think that's actually the point. We can't imagine what comes after the world of the super-human minds because -- well, we don't have the words for that world. We can reason within the system we know until sometime close to when these critters come online, then we can't.

That doesn't mean humanity necessarily kicks off. Lots of geeks imagine we'll upload our minds into unoccupied (!) processing environments, or that the AIs will be sentimental. Not everyone is as cheerily pessimistic as me. It's not called a "Singularity" because it's the "end", it's because we can't make predications about it. Super-AI is death to prediction.

--
My Google Reader Shared items (feed)