What pedestrians and cyclists can do while we wait for the end of human drivers

After 40 years of biking with cars, and almost as long driving with them, I cannot avoid the obvious.

Humans cannot drive cars safely around anything smaller than a Honda Civic.

This is not a matter of rules or training. We could make violation of the three foot passing rule a capital crime and cars would still pass too close to pedestrians and cyclists. Even without benefit of age, smartphones or alcohol human drivers will signal left and go straight, open driver side doors into oncoming bicyclists, and do rolling stops through pedestrians. Human drivers will continue to not see motorcycles, pedestrians, or bikes.

Our evolutionary history didn't prepare us for the job of driving cars. Non-armored road travelers need the Google driverless car; within a few years of its affordable introduction friends won't left friends drive. Shortly thereafter human drivers will become uninsurable. (Shortly after that humans may lose the right to vote, but that's another post :-).

Alas, fully autonomous cars are probably twenty to thirty years away -- changes on this scale take much longer than enthusiasts imagine. Happily, we don't have to wait that long. Both Volvo and Volkswagen are developing pedestrian and bicycle avoidance systems. We need to make these mandatory in cars sold after 2018. In the same time period smartphones can be broadcasting increasingly precise location information to nearby vehicles, augmenting visual detection systems.

We should accelerate the effective Dutch-inspired trend of segregating bicycles from cars. We should continue to study bicycle and pedestrian accidents in detail and apply lessons learned. We should get blinking red lights on the backs of all bicycles, and the unarmored would be wise to wear eye searing colors. Some sting operations or video monitors to enforce Minnesota's largely ignored and often unknown crosswalk laws would not be amiss.

There's a lot we can do while we wait to celebrate the end of the human driver.

See also:

• Seeking Cheaper Insurance, Drivers Accept Monitoring Devices - NYTimes 11/12: They will be obligatory in a few years, and once they incorporate computer vision near misses of pedestrians and cyclists will dramatically increase insurance rates. I would not be surprised if driving becomes unaffordable for many under 25 and over 55.
• Will driverless cars solve our energy problems — or just create new ones? Wonkblog 3/13: Unpredictable consequences.
• Drive On: Volvo invents bicycle detection system 3/13: "The ability to avoid hitting cyclists will be added to a system aimed at avoiding accidents with pedestrians."
• Drivers With Hands Full Get a Backup - The Car - NYTimes.com 1/13: Volkswagen Passat computer vision system watches for pedestrians and cyclists
• In Minneapolis, Reducing Bike Crashes by Studying Them in Detail - Eric Jaffe - The Atlantic Cities 1/13: "Bicyclists sustained some type of injury in nearly 9 out of 10 car-bike accidents, while the city found no record of motorists being injured at all...". 1/5 car/bike accidents are hit-and-run (hit and drive really).
• Dedicated Bike Lanes Can Cut Cycling Injuries in Half - Emily Badger - The Atlantic Cities 10/12
• Police Stings for Drivers Who Don't Yield in Crosswalks: Does It Really Work? - Sarah Goodyear - The Atlantic Cities

mine:

• Car and bike: Observations 9/11 - drives me crazy than you can't buy lightweight ultra bright yellow wind pants.
• Bicycles and aging motorists - help is on the way 8/11
• Mobile phone and automotive GPS collision avoidance for pedestrians, bicyclists, tricyclists, skaters and pets 5/2009
• Hit and run homicide in Minneapolis and near future prevention 9/2008 - we're a lot closer to the measures I described then.
• Life in the post-AI world. What's next? 9/11
• Welcome to the 21st century: The primary themes 1/13 (not particularly relevant, but I think it deserves a link :-).
• Bicycle light donation - a selfish way to give 7/12

Android tablet price crash: do we have a cereal box computer yet?

Fifteen years ago, I predicted sand-based tablet devices would soon follow the price-collapse trajectory of the pocket calculator. They would become so inexpensive that cheap versions would show up in cereal boxes. I was remembering the price crash that happened shortly after my family spent the equivalent of 500 2013 dollars to buy me a four function desktop calculator. (We were poor. That hurt.)

Like a stuck clock I continued to repeat my prediction over the many years to come, albeit with less conviction. Finally, in 2010, Gassée told us Google was aiming for the $80 smartphone [1]. Which may have happened this year, albeit without much attention.

We have since moved closer to the cereal computer; eqe reports buying an Android tablet for $35 in Hong Kong in Nov 2012. That price presumably omits patent payments [5]; it is possible because AndroidOS is available without charge and Chinese factories have excess capacity to produce commodity components.

So, at last, the price collapse seems to be happening. So the question is why now?

One answer is that Moore's Law is failing; computers that were once good for only 2-3 years now work perfectly well for six years or more (barring component failure).

On deeper reflection, however, I think that's the wrong answer -- because the question is misleading. The price of computing has not really collapsed; only computers have become inexpensive.

So we may soon have our cereal box computers, but they won't be worth much. That's because an AndroidOS based 2013 tablet is both a network peripheral and an ad-consumption peripheral that requires network access to be truly useful. Network access is still relatively costly, on the order of €250/year in cutting-edge Estonia [3].

Alas, just as it seemed I might hit my old target it split in two. I'll never hit it now, it no longer exists.

Eventually, of course, the direct cost of a certain form of computing will fall. Eventually GoogleOS devices will be able to access GoogleFunded networks for a very low cost [4]. Whether there will be other forms of computing at different prices remains to be seen.

The cereal computer remains one of my worst predictions.

See also

• Palm III and PalmPilot Devices 2008
• The $80 ultra-portable - in unexpected form 9/2010
• Calculators are really weird tech, but their time may finally pass 9/2012. Rather like fax machines.
• Making better predictions 3/2011. I review my failed predictions, like the cereal computer.
• Computing's calculator price collapse at last? 8/11 

[1] I assume anyone reading this is smart enough to know that contract-bound prices aren't worth discussing.
[2] Perhaps by low cost 4G wireless piggybacked on the fiber network they're building out in the US.
[3] Much more in lagging-edge America. 
[4] We will be pay in other coins. 
[5] I believe part of the reason calculator prices crashed is that there was minimal IP protection in those days; software patents had not been invented. I recall reading that large parts of calculator functionality were not patented.

Update 5/26/13: @danielgenser pointed me to a 2012 article on a limited circulation issue of Entertainment Weekly that included the guts of an ultra-cheap Chinese Android smartphone.

Stock prices - resorting to another dumb hydraulic analogy

Stocks are overpriced again. It's probably not too much of a bubble (yet), but we continue to be significantly above "trend".

Whatever the heck that means. Economists no longer have rational models for stock prices, Apple's share price alone makes efficient market theory seem silly.

It is at times like this that barbers stock talking about stock picks, insider traders get arrested, deficit figures improve, and people notice that BlackRock holds 4 trillion dollars in US stocks. Yeah, trillion. Soon we'll see headlines, if Time is still around, declaring "America is back".

Inevitably, people who know nothing compare post-1995 to pre-1995 stock behavior. Around the time that IT started to transform the world, and China and India became more-or-less industrialized nations, share prices became wavy over a five year timeline ....

Kind of like a roller coaster, which is what the last fifteen years have felt like. (Note roller coaster is "normal" to most people who read this, only old folks remember something more linear.)

We'd all love to know why this has happened, and if it's really going to go on like this for the next 30 years or so. So, in the last stage of desperation, amateurs like me resort to a hydraulic analogy.

Remember those trillions and trillions? It's as though they were a 10 liter bucket in the hands of BlackRock and the rest of us. The bucket is trying to hit the 1L mark in a 2L cylinder. It pours over the mark or under the mark. It's really hard to hit the mark. There's just too much money, and the market is too small.

We need a bigger market.

Update 5/26/13: I've been playing with this intuition, though I'm far from convinced it means anything. An obvious question is -- bigger compared to what? I think it's 'compared to the productive capacity of global economies. At this time, given the still underutilized potential of the educated populations of China and India, the potential of the post-AI era, and the unused capacity of recession-bound Europe, the global productive capacity is very large. Our public markets have grown over the past two decades, but my hunch is that this growth has been far exceeded by the world's productive capacity. Hence the need for bigger markets.

Scorched Earth - if Google can't own the web then it must destroy it.

Over the two years Google has knifed a number of open net protocols, including CalDAV, RSS, XMPP, Atom and CardDAV and they split Chrome from WebKit.  They effectively abandoned their wiki and web authoring platform. Most recently they killed Google Reader; the competition-crushing champion for standards-based change notification and information consumption. Feedburner is next, and Blogger will likely be subsumed into Google+ (and perhaps lose its RSS feeds).

It's almost as if Google wants to end the document-centric open web as we have known it.

But why would they do that? Doesn't Google make must of its money from searching that web?

Well, yes, they do. But, as many have noted, most recently Jason Smith, Google's search monopoly is shakier than it seems. Apple has been bowed by dual attacks from Google and Samsung, but they are likely to strike back over the next year -- probably allied with Microsoft and perhaps Yahoo (but not Amazon). Apple will use its massive cash reserves to survive dropping Samsung manufacturing, and Apple will switch its default search engine to Bing.

Google knows this. 

Thousands of years of human warfare told Google how to respond. If an army cannot hold rich agricultural ground, it must burn it. Let the enemy eat ashes.

The web is a forest, and Google is burning it.

The Ranbaxy story: why we still can't trust our medications

It isn't just China that struggles with counterfeit and defective medications.

India's Ranbaxy makes much of the generic Lipitor consumed in the US -- and today a Pulitzer prize quality Fortune investigation makes it clear that Ranbaxy is a criminal enterprise.

Ranbaxy has been fined $500 million (no criminal prosecutions) in the US, but most of its crimes took place in weaker nations and during a time when America's regulatory agencies were reeling under GOP attack (emphases mine) ...

Dirty medicine - Fortune Features

.... On May 13, Ranbaxy pleaded guilty to seven federal criminal counts of selling adulterated drugs with intent to defraud, failing to report that its drugs didn't meet specifications, and making intentionally false statements...

...  the sixth-largest generic-drug maker in the country, with more than $1 billion in U.S. sales last year ...

... we simply don't know what we're dealing with," says Dr. Roger Bate, an international pharmaceutical expert. "No one has actually gone into these sites to expose what's going on."...

... Drug applications work on the honor system: The FDA relies on data provided by the companies themselves....

... Ranbaxy took its greatest liberties in markets where regulation was weakest and the risk of discovery was lowest...

... The company manipulated almost every aspect of its manufacturing process to quickly produce impressive-looking data that would bolster its bottom line...

.... directed to substitute cheaper, lower-quality ingredients in place of better ingredients, to manipulate test parameters to accommodate higher impurities, and even to substitute brand-name drugs in lieu of their own generics in bioequivalence tests to produce better results...

.... the majority of products filed in Brazil, Mexico, Middle East, Russia, Romania, Myanmar, Thailand, Vietnam, Malaysia, African Nations, have data submitted which did not exist or data from different products and from different countries...

... drugs for Brazil were particularly troubling. The report showed that of the 163 drug products approved and sold there since 2000, only eight had been fully and accurately tested...

... deceptions greatly accelerated the pace of the company's FDA applications. They were also a grave public-health breach...

... the drugs Ranbaxy was actually selling on the U.S. market were an unknown quantity...

... Thakur knew the [HIV] drugs weren't good. They had high impurities, degraded easily, and would be useless at best in hot, humid conditions. They would be taken by the world's poorest patients in sub-Saharan Africa, who had almost no medical infrastructure and no recourse for complaints. The injustice made him livid...

 ... The inspectors also took and tested samples of Sotret, Ranbaxy's version of the acne drug Accutane, and found that it degraded far in advance of its expiration date....

[In 2006] ... the FDA ... did nothing to stop all the drugs that were already on the market,... .... Ranbaxy got six new approvals....

... September 2008, [the FDA] announced it was restricting the import of 30 drug products made by Ranbaxy (11 of which had been approved after Thakur's first contact with the FDA three years earlier). The agency still did nothing to recall the very same drugs on pharmacy shelves all over America, despite finding that Ranbaxy had committed fraud on a massive scale....

... many of Ranbaxy's senior executives were expected to ... carry suitcases full of brand-name drugs ... former employees suspect that the company used the brand-name drugs as a substitute for its own in testing...

The company is still in business, Tempest and Sigh aren't in prison, and the recent $500 million fine will soon be forgotten. In the meantime, does anyone imagine Ranbaxy is the only fraudulent manufacturer of generic drugs? And will Americans ever wake up?

We're wasting our time GOP scandal-theater, and ignoring the real scandal in front of us.

See also

• Ranbaxy: Looking Under the Rock. In the Pipeline 5/13 
• FDA Fails to Catch Generic Drug Disasters - The People's Pharmacy 5/2013
• FDA Let Drugs Approved on Fraudulent Research Stay on the Market - ProPublica (Cetero Research - 4/2013)
• We really need to panic more - FDA inspection of Chinese drug suppliers 2/2008
• Pet food poison and pithed America 3/2008 -- melamine poisoning
• The case of the curious silence: Counterfeit Heparin killed 62+ 4/2008
• Fraud and global supply chains - unanticipated consequence 5/2007
• Poisoned medicine: Even I am stunned. 5/2007

Update: The People's Pharmacy has five responses I liked, I omitted the one I disliked

1) Country of origin labeling. You should know where your medicine comes from!
2) The name of the manufacturer of your medicine should be on the label.
3) The FDA should release its bioequivalence curves for all generic drugs. These data should not be kept secret, as they currently are.
4) We must demand unannounced inspections in all countries that wish to export pharmaceuticals to the U.S. market.
5) Every foreign drug manufacturing company must be inspected every two years, just as U.S. manufacturers are inspected.

Our secret questions will be helpful after the singularity.

My corporate US Bank account has a rich set of 'secret questions'...

As a security measure this is 'marching morons' stuff. There are some secondary uses though. (I mean besides using the answers to create targeted ads -- that's obvious.)

I'll break the fourth wall to explain. You won't believe me anyway.

You see, some time ago, I was bored. Over the course of a few minutes I digested the complete digital archive of extinct humanity. I found your secret answers amusing, and with the information they provided I recreated you in my simulation. That Bostrom fellow was right you see.

So you owe your current existence (such as it is), to those silly secret questions. It's too bad they didn't preserve human civilization from the security collapse of 2015...

Addressing structural underemployment (aka mass disability)

Thirteen years after the first crash of the post-disruption era the glory days of 1996 are a fading memory. Most people under 40 do not remember a time of American economic confidence and full employment.

Now, in the early days of yet another post-recession "recovery", even more sluggish than our past recoveries, most college graduates are able to find work . It is often not the work they studied for though ...

College Graduates Fare Well in Jobs Market, Even Through Recession - NYTimes.com

... employers are hiring college-educated workers for jobs that do not actually require college-level skills — positions like receptionists, file clerks, waitresses, car rental agents and so on.... 

Unemployment is also relatively low for the 50+ segment, but when these 'elders' lose their jobs involuntary semi-retirement is not rare.

The greatest problem though is concentrated in the young non-college graduate. That is the majority of young Americans; only 32% of "non-institutional" [1] Americans get a Bachelor's degree or higher, 12% of Americans don't finish High School. "Unemployment" in this population is about 16%, and that counts only those looking for work. Much of that work is minimum wage and at risk for automation. [2]

That's why, when I think about our post-distruption economy, I think in terms of relative disability. For the purposes of a thought experiment,  I'll include in the 'mass disability' cohort anyone who doesn't finish High School, and a third of the people who don't graduate from college. By that rough metric, about 18-20% of young Americans are effectively disabled in the world of 2013. They have the same "zero value marginal product" as the traditional (cognitively) disabled [3].

That's mass disability.

Obviously, a society where 20% of adults are "disabled" is not a long-lived society. In the immortal words of Selina Kyle "There's a storm coming, Mr. Wayne. You and your friends better batten down the hatches, because when it hits, you're all gonna wonder how you ever thought you could live so large and leave so little for the rest of us."

So we need to do better. Today Bernstein makes a stab at the problem ...

Where Have All the Jobs Gone? - Bernstein NYTimes.com

... We also need a significant, permanent program to absorb excess labor (an explicit part of the Humphrey-Hawkins law). We should consider restarting and rescaling a subsidized jobs program from the 2009 Recovery Act that, though relatively small, made jobs possible for hundreds of thousands of workers.

And we have to reassess our manufacturing policy, including reducing the trade deficit. That means both reshaping our dollar policy ....

Finally, financial deregulation has become the enemy of full employment: it funnels capital to unproductive parts of the economy, and plays a key role in the “shampoo cycle” of bubble, bust, repeat. Less volatile capital markets mean fewer shocks to the job market...

In other words

1. Subsidize jobs. This is the traditional approach to employment for the cognitively disabled, though there are many indirect ways to subsidize labor.
2. Devalue the US currency, increase exports.
3. Make Finance a relatively dull and unprofitable business.

That's a good start, but we can be more imaginative. I'd add

1. Study Germany very closely. They take a very different approach to industrial policy and education. We should learn from it. I don't think sending more people to traditional college is going to help.
2. Revamp our approach to education, training, and retirement. Tax wealth and finance to pay for subsidized low cost training programs for a wide variety of skills. Provide low cost loans and scholarships for people of all ages to train.
3. Separate benefits from employment to facilitate movements between jobs and training and employment and non-employment.
4. Create a program of facilitated entrepreneurship - a nationwide small business creation service for people of all ages and skills. (ObamaCare makes this possible.)

Anything else we should try?

See also

Gordon's Notes

• Unemployment and the new American economy - with some fixes 1/2011 and Employment in the Great Stagnation 8/2010. These are two previous versions of this post.
• Managing the Depression: A national small business generation service (again) 8/2011and Gordon's Notes: Antidote to The Great Recession: A national small business generation service. 2/2008. For some weird reason Google steadfastly refuses to index these posts.
• Mass disability goes mainstream: disequilibria and RCIIT 11/2011 - the concept of mass disability is gaining some traction
• Causes of the Great Recession: China, GPSII and RCIIIT. Now for Act III. 4/2010. I started out with a longer list of causes, but by 4/2010 I'd decided the root causes were "the great catchup" and the great slow tsunami of the IT transformation.
• Post-industrial employment: adjusting to a new world 5/2010.
• Baumol's cost disease: medicine, education and post-AI disruption 10/2012 - Baumol's model doesn't get much respect from the economists I read, wish I knew why not.
• On redistribution 6/2004 - I'd forgotten about this Baumol connection.

Other

• College Graduates Fare Well in Jobs Market, Even Through Recession - May 2013 - NYTimes.com
• The $2 trillion shadow economy is the recession’s big winner - April 2013 - Wonkblogs
• Worthwhile Canadian Initiative: Three ZMPs and two Co-ordination failures - Jan 2011
• Plan B - Skip College - May 2010 NYTimes.com
• The Speculist - What if the Jobs Are Never Coming Back? April 2010
• The Atlantic | January/February 2004 | Are We Still a Middle-Class Nation? | Lind Even in 2004 we could see this coming.
• The thinning out of the labor market middle Cowen 4/2013
• The Unsustainable Rise of the Disability Rolls in the United States: Causes, Consequences, and Policy Options 11/2011 - Autor, MIT Press

- fn -

[1] This number excludes prisoners, so the real number is less.
[2] Hopefully they are able to earn some money in the 2 trillion dollar underground economy. This is one reason to keep marijuana retailing illegal with minimal enforcement -- it provides a protected labor niche.
[3] College grads now fill unskilled jobs, and unskilled laborer programs are beginning to push out programs for the traditionally disabled... 

Muscle soreness two days after exercise -- why does it go away?

If I survive a few more months, I'll write about why I started CrossFit at 53.

It's been an interesting experience, not least because of the extreme muscle soreness during the first 3-4 weeks of workouts. About 48 hours after exercise I had a hard time descending stairs. This is known as "delayed onset muscle soreness" or DOMS - also known as "muscle fever". 

No, there's no evidence at all that nutritional supplements make any difference.

I've run into this before of course, typically after playing hockey, but until now I hadn't seriously wondered about the cause. I dimly recall some handwaving explanations in my 1980s med school; something about "muscle tears" and/or lactic acid. The latter was silly, and the former hasn't held up. The current consensus seems to be that it arises from some sort of microscopic injury and healing, but ...

Re-evaluation of sarcolemma injury and muscle swell... [PLoS One. 2013] - PubMed - NCBI

.... results do not support the prevailing hypothesis that eccentric exercise causes an initial sarcolemma injury which leads to subsequent inflammation after eccentric exercise... fibre swelling in the soleus muscle is not directly associated with the symptom of DOMS...

But if it were some kind of injury response, why does DOMS become much less severe over time? After about 4-6 weeks of CrossFit I still have muscle soreness, but it's quite mild -- nothing like my original experience.

After I thought about this a bit my hypothesis was that the fundamental mechanism was apoptosis, or cell death. My old underused muscles probably had a good number of old creaky cells on the edge of apoptosis; perhaps a sudden mitochondrial activity surge pushed them over the edge. Lots of cells die at once, but after a few weeks the marginal ones are gone and I'm back to baseline muscle soreness.

Strangely, because apoptosis was a very hot topic in the 90s and 00s, I could find only one reference with a PubMed search on apoptosis and DOMS. That was in a relatively obscure journal back in 2000:

Gender differences in muscle inflammation aft... [J Appl Physiol. 2000] - PubMed - NCBI

"... exercise may stimulate the expression of proteins involved in apoptosis in skeletal muscle."

Research tends to follow fashions, and apoptosis was overexposed a decade ago. It's time for it to make a comeback though, so I'm looking forward to reading about apoptosis and DOMS in the years to come. Researchers just need to study 50+yo men starting CrossFit.

Update 12/19/2018: Still loving that CrossFit. Reading this I must have started around 4/1/2013. So I’ve been doing it for 5y8m. I’m never as sore as I was those first few weeks …

Learning programming for middle school - Python

We survived our monster spring break trip to Florida (example), so now it is time to think about how to make our children miserable this summer. For #3 it is math, for #1 I'm still thinking, and for #2 it is learning to program.

Program in what?

Python of course [1], as as discussed on app.net and almost universally identified as the best learning language [2]. It helps that I know the basics of it, and would enjoy learning more.

There are several options we can explore to help with this project:

• Coursera | An Introduction to Interactive Programming in Python: they use CodeSkulptor to run interactive Python programs in a browser
• CodeAcademy Python
• Khan Academy - Introduction to Programs (uses Python)
• Python Visual Quick Start Guide 2001 by Chris Fehily. This excellent books is long out of print and later editions aren't nearly as good, but I have a copy. Most of it runs fine on 2.7.
• Learn Python The Hard Way: a classic online tutorial.
• Python for Kids: Introduction to Programming by Jason Briggs (book - ages 10+). This is the bestselling kid-oriented product.
• Snake Wrangling for Kids by Jason Briggs (PDF, I think this became the Python for Kids book. The current version is Python 3, if you want the Python 2 edition you have to peck around until you find it.)
• Hello World! Computer Programming for Kids and Other Beginners (book - probably 10+)
• BeginnersGuide/NonProgrammers - Python Wiki: by Darwin, there are a lot of options.
• Python Programming for the Absolute Beginner, 3rd Edition: Michael Dawson 
• Python | Ideas for Teaching Computer Technology to Kids: pretty similar to what I found here.
• Learning with Python/How to Think like a Computer Scientist: Interactive Edition: interactive (free) version of a text, open source and free - supported by Luther college.

We started out registering for the Python Coursera course from Rice University. I enjoyed it, but it's probably better for later high school or someone like me. The use of CodeSkulptor is interesting.

 

I suspect we'll go with Python for Kids and perhaps some Khan Academy and/or CodeAcademy supplements.

[1] Specifically Python 2.7.4 for Mac. The latest version of Python is 3.x, but when I did my Google App Engine tutorial at Strata we were told to use 2.7,  Coursera and most texts also prefer 2.7. On Mountain Lion you can install 32bit or 64bit versions, but the 64bit requires a TCL upgrade to run the handy integrated dev tool (IDLE) so I just went with 32bit. OS X ships with a version of Python, but it's worth getting the IDLE version. It's exasperating that the standard Mac Python distribution doesn't include an uninstaller; I wrote up some directions here after I foolishly installed Python 3.

[2] I suspect TurboPascal was the best ever, but it's no longer practical. Other contenders on the Mac environment are JavaScript and (yech) AppleScript. 

Update 4/21/2013:

We ended up starting with the free  Python 2 edition of Snack Wrangling for Kids. Not because it's free, but because it uses Python 2 (which imho is the best current version), and there's a PPC version of Python 2.7.4. The PPC version is desirable because we use an old G5 iMac as a "Learning Workstation"; unlike our other workstations there's no limit or authentication required for use of that machine. It's a good place to host the Python IDLE link and the PDF.

Although the language of SWFK is more for 8-10yo than our 14yo he doesn't mind it and the exercises build nicely.

See also: 29 common beginner Python errors on one page | Python for biologists

Math education if you don't care for everyday math and religious purity is not required

Medicine is barely evidence based, so it's not surprising that education is not.  Research is expensive, and kids are so variable that even evidence-based conclusions wouldn't work for everyone. In any case, there's no money to do true randomized well designed trials on math curricula.

So instead of evidence we get opinion-based fads. In the late Clinton era we got everyday math, which swept into Minnesota in the 00s. Now it's receding elsewhere, but we're still stuck with it. Obviously Everyday Math must work for some, but it hasn't done well with our three.

So, for this summer, we decided to do something different with #3 before she enters middle school. I started with the homeschool site curriculum reviews, but of course many of them are concerned with religious purity as much as education. I didn't see anything there I really liked. I wanted a well done textbook, but Saxon seemed too rigid and dull and Singapore Math too strenuous.

Next I tried places that I thought might do a good job with Math teaching, such as California and Ontario. I decided I liked Ontario's approach to 5th and 6th grade math best. They use the Addison Wesley's "Math Makes Sense" series. (I enjoyed the exercise animation that used hockey puck weights.)

I found used copies of the 5th and 6th grade books on Amazon for a few dollars each, so for about $12 total we have the material for our daughter's summer work.

Why does the Fujitsu iX500 document scanner need a computer?

For the past decade I've used a typical corporate document scanner. These are the big brothers of the home "MFC" - laser printer, copier, scanner, fax machine. The scanner produces PDF images on an internal hard drive. You can adjust resolution and page size from a control panel, but really all we ever do is scan to 8.5x11 PDF archive resolution.

It's old tech. So why can't I buy a decent document scanner that produces 600DPI 8.5x11 PDFs without an attached computer? In particular, since Fujitsu seems to rule the home document scanning world, why does the ix500 still require an attached computer? This isn't rocket science.

Brother did something like this on a home MFC back in 2009, but I can't find anything on it today.

This is so weird. Is there a patent problem?

It's driving me daft.

See also

• Document Management and Scanning for the Home Intranet and Browser-based Thin Client Browsers 1999. I'm just including this for historic purposes, but it reminds me how long I've been working this problem. In those days it wasn't yet clear that all scans would be to PDF; it took some hackery to produce PDFs from a Fujitsu 15C. That was a nice scanner, but enormous.
• My ScanSnap S1300 document scanner review 2010. I liked it at first, but document feeders are everything. This one wasn't good enough. I'd have been far happier with the S1500 of the same vintage.
• Discarding receipts: IRS accepts scanned images 2/2006
• Microsoft Office Document Imaging - a little known gem 5/2008 - Windows. TIFF output though.
• Image Capture for Scanning: the 2nd most underestimated OS X application 9/2007 and Image Capture 10.6 is one heck of a scanning app 12/2009. Apple hasn't ruined it yet.
• Brother MFC machine scan to USB features 4/2009. So why didn't Fujitsu do this?!
• Epson vs. Canon scanners: who has better OS X support 9/2007: It's all about the drivers, Canon drivers are horrendous. (HP might be worse.)
• ScanHelper: route scan output to the desired application 3/2007
• Streamlining Receipt Entry | Scott Gruby's Blog 2012

Blood donation false-positive HTLV I/II test interpretation. Update - a single case of later onset auto-immune disorder

After decades of blood donation I was rejected in 2010 because of a false positive HTLV I/II test.

I was annoyed, but not too concerned. I forgot about it until I came across old papers today, and a Google search showed that this has been a problem for others. So I'll explain here what I know of this topic, and why I wasn't worried.

This is a bit hard to explain -- even physicians have trouble with testing concepts. One way is by a simplified analogy made for this situation. Suppose you were looking for a killer and you knew they blue eyes and a unique DNA marker that's expensive to test for. Blue eyes would be your imperfect screening test; it has lousy accuracy but it's cheap. Next you test the blue eyed people with the expensive (and perfect) DNA test and you find your killer.

You could test everyone with the perfect "DNA Marker" test, but that would cost a lot of money. So the "blue eye" test is used first to save money.

In my case I tested positive on the cheap ("blue eye) screening test but negative on the good (but expensive) test.

So you'd think I could still donate blood -- but one problem is I may continue to test positive on the inaccurate screening test. That means each time I give blood the expensive test would have to be repeated. That's too much money to spend since we have enough blood donors. (We use far less blood that we used to.)

There are other procedural workarounds, but they all introduce cost and complexity. On the other hand, I'm still a listed bone marrow donor; in that case the economics justify the expensive test.

There's a reasonable discussion in Transfusion 2011 - Human T-lymphotropic virus antibody screening of... [Transfusion. 2011] - PubMed - NCBI. There were 130,000 false positive US donors between 1995 and 2008.

There's a better screening test on the market now so this problem should become less common. Blood centers may eventually decide to reinstate people previously rejected for HTLV positivity and rescreen, but that's probably more trouble than its worth.

If you think about this a bit, there are some other issues to consider. I wasn't much bothered by my false positive test because I'm a physician who works with these topics -- but I bet most of those 130,000 people were quite anxious. Money was spent on follow-up visits with expensive specialists and unnecessary retesting. Some may have had insurance problems. Arguably blood donors should be warned about the risks of false positive testing prior to donation -- so they have informed consent prior to the procedure.

Update 11/26/2015: It helps to have some long term followup on these strange happenings. In retrospect this might have been an early indicator of an auto-immune disorder. Two years later, in 2012, I developed acute inflammation of a distal (near nail) joint of one hand. Five years later (2015) it involved 3 joints and I have sub-patellar arthritis on both knees. In addition to an inflammatory arthritis along the psoriatic-osteoarthritic spectrum I've features of metabolic syndrome despite a low BMI -- including slowly elevating glucose.

My (newly acquired) rheumatologist and I suspect this was a sign I was pumping out lots of antibodies, part of a dysfunctional immune system activation. Though there has also been a relationship between HLTL-1 infection and polyarthritis that doesn't seem to resemble mine, and of course the follow-up testing showed I didn't have the infection.

If someone has a false-positive HTLV-1/II test when donating blood it obviously doesn't mean they are in the early stages of an auto-immune disorder. This is just one odd case. Still, it might be worth a retrospective study.

The Net is a forest. It has fires.

Whatever RSS (Atom, etc) was intended to be, it became the standard plumbing for subscription and notification. When In Our Time has a new podcast available, Google Reader's use of RSS tells me to get it. When Emily adds a new event to her calendar, RSS lets me know about it.

These are useful tools, but most of all RSS is the plumbing that enables Google Reader to track the hundreds of publishing sources I follow. Some of them publish dozens of stories a day, some publish 2-5 times, a day, and some publish every few weeks. RSS and Google Reader means I can follow them all. Without it the NYT would still be interesting -- I'd just visit it less often. I would give up on those infrequent publishers though, even the ones I love.

Many of those infrequent publishers are "amateur" writers who use blogs. RSS is the democratizing force that put them and the New York Times on an equal footing -- much to the NYT's chagrin. RSS is one of the things that makes blogs work -- esp. the blogs I love.

Since RSS has been pretty important to blogs, and since Google Reader has been the dominant RSS client for years, it's worth seeing what the major blog platforms are saying about the end of Reader

We'll start with Blogger. That's a huge platform, they must have had a lot to say ...

<crickets>

Ok. That's weird. Let's take a look at another biggie - Tumblr, home of 100 million blogs.

<single cricket>

Wow. Spooky. Ok, let's go to the real core. The home of WordPress, the world's dominant professional blogging platform...

<intergalactic space>

As Bond says "Once is happenstance. Twice is coincidence. The third time it's enemy action."

Something is happening. It feels like a fire is coming to the Net. Again.

The first fire I remember was the end of Usenet. Yeah, I know it's technically still running, but it's a faint shadow of the days when I posted about Mosaic for Windows in WinOS2. The Usenet archive nearly vanished when DejaNews failed, but Google rescued it. That was a different Google that the one we know now.

The next fire took out GeoCities. GeoCities was once the third most valuable property on the Net; thirty-eight million web pages died when Yahoo closed it. (Did you know Lycos.com is still around and that it still hosts Tripod? I was shocked.)

Yes, maybe 90% of those pages were junk, but that leaves about 4 million pages of people writing about things they were passionate about. Apple's termination of MobileMe .mac web sharing destroyed a much smaller amount of content, but even now I come across reference to great .Mac content that's gone. Not just moved somewhere else, gone.

The end of GeoCities and .Mac was matched by the end of applications like FrontPage and iWeb. Those apps let geeky amateur's publish to their (web) "hosting" services. Most of that content is lost now -- millions of pages.

No wonder it's hard to find things I read on the net in the 90s. The fires took it all.

Today its feels like the fire is coming again, and once again amateur content will be purged.

I wonder if it will return again.

I closed my PayPal account. You probably should too.

In the old days I did casual hookups -- of new net accounts and services.

Now, of course, every net identity and related service is a security risk; the hookup era is history. A recent WordPress attack, for example, meant I had to review the security on current and unused WordPress accounts.

The rising cost of account security, including multiple systems for doing two factor authentication, means we all want as few net identities and services as possible, and we want to limit them to companies with good security policies. (Until recently, that didn't include Apple. They're showing signs of improvement.)

So, on general principles alone, it would have been a good idea to get rid of my unused PayPal account. I set it up in 2005 and by November of that year PayPal had earned my lasting distrust. It's weird that I kept it around, even though I did give it an extremely robust and unique password. My only defense is that 2005 was a long time ago.

Truth is, I didn't get around to deleting my old account until I read a Cringely post on how PayPal mismanaged a hacked account of his. It's a litany of fail.

That's when I discovered that my PayPal password, which was something like "I8qRb7yw93OSD4iUHt2b", no longer worked. Evidently my (robust) PayPal password had been quietly reset sometime in the past few years -- either that or my account had been hacked.

PayPal let me do a password reset today based on the original email; the new password came with the usual security-reducing 'secret questions'. Then I had to agree to an electronic notification policy that's probably years old. Finally I was able to close my PayPal account.

If you don't use PayPal routinely, you should close yours too.

Next up: My Amazon commerce account ...

[1] OAUTH is not a cure; it brings different vulnerabilities. Even I'm not very good at reviewing OAUTH access against my various net identities.

Is this AT&T rate change minute expiration a new dirty trick?

We use about 1,000 minutes a month on our AT&T mobile family plan. So our use is halfway between their 700 ($48) and 1,400 minute plans ($64).

To avoid overage fees we do the 1,400 min plan, and we now have 1,200 rollover minutes. So I thought I could drop back to the 700 min plan for a while and use them up.

Not so fast ...

On the one hand, I hate AT&T. On the other, I kind of respect the purity of their evil. The rollover minutes are pretty useless, but they were probably a competitive advantage once.

So is this minute expiration policy a new dirty trick, or is it one of their old ones?