Category: Uncategorized

Free-range parenting, and by implication the opposite (helicopter parenting), seems to be in the news all the time (example from today’s New York Times that talks about the “narrowing of the child’s world has happened across the developed world”).

At the same time I have been poking around to find a new car seat for our son (will be 16 months old when the seat arrives). I’m discovering that the goal of safety advocates is to keep children rear-facing until they are age 4 or 5. All of the articles talk about how this makes children 50 or 75 percent “safer” but there is no mention of the actual statistical risk. Is the risk of injury in an accident being reduced from 10%/year or from 0.0001%/year? None of the articles include this information. Nor do any say “You could cut your child’s risk to zero by leaving him or her at home, buying a house that is walking distance to school (and where no streets need to be crossed), not signing up for Russian Math or Kumon unless those are offered within walking distance from your house, etc. You could also cut the risk in half by getting a minivan instead of a compact sedan. You could cut the risk by at least another fact of two avoiding driving at night, in the rain, or when you’re tired.”

Personally I try to avoid schlepping children around in cars because it seems like a second-rate environment for a child to learn/develop. But to the extent that they must travel in a car with me I try to use the time to point out stuff that we can see out the window. I’d be interested to hear from readers, e.g., in Scandinavian countries where supposedly rear-facing until older ages is common, how in-car conversations about the scenery work when a child can’t see the same things as the adults in the car.

Parents spend so much time these days trying to make sure that every possible moment is spent on some sort of enrichment activity. Could it be that the rear-facing idea reduces the child’s mental enrichment to the point that the reduction in injury risk is not worth it? (“There is no level of acceptable risk” is not a sensible answer because if a parent truly felt that way the child would almost never be in a car at all (see above).)

[Separately, I recently got a letter from the school superintendent here in our (rich) suburb of Boston. This lists a lot of hazards facing children but car accidents are not among them. Here’s an excerpt:

“If you are a parent, at some point it is likely that your child or someone they know will face one or more of these issues…

• Mental Health
• Substance Use/Abuse
• Domestic Violence
• Learning Differences
• Bullying
• Stress
• Suicide Prevention
• Eating Disorders
• Sexual Health
• Autism Spectrum
• Resiliency

… Don’t wait until your child is facing an issue to become educated. Prepare yourself now. Early awareness and intervention are the best methods of prevention.”]

Full post, including comments

Folks:

Today marks the release of Patchmania for the iPad/iPhone. It is a free download with in-app purchases. So I would be interested to hear from game-loving iOS users what they think of the puzzle game. I love it (on an iPad 3), but I’m wondering if that is because I know the developers.

Thanks in advance for any feedback.

Related:

• post about this game being tested at the California Department of Motor Vehicles

Full post, including comments

“Hospital Discharges Rise at Lucrative Times” (Wall Street Journal, February 17, 2015) has a great chart showing how the American health care system follows economic incentives held out by the federal government. Hospitals can make nearly twice as much if patients are discharged one day over the Medicare threshold rather than one day under. This results in a huge spike in discharges on the magic day. (Click on “3” to see the best chart.)

Related:

• my December 2014 posting about the American Economic Association conference, in which the response of medical doctors to drug company incentives is examined

Full post, including comments

“U.S. taxpayers stuck with the tab as helicopter flight schools exploit GI Bill loophole” is a Los Angeles Times story (thanks Mike Aracic!) about the way the Veterans Administration funds flight schools. The journalist misses the corporate welfare angle that I’ve written about here for a few years. Flight schools affiliated with a standard college can collect 100% funding plus living stipends while independent flight schools can only tap into the VA for about 50 percent of the total cost. The veteran is therefore much better off training at a school affiliated with a university, thus resulting in the taxpayers paying 5-10X as much (add in the living stipend, tuition for the college on top of the flight hours, and much higher costs per flight hour). The story is about the waste of taxpayer dollars but it is missing the fact that it isn’t a waste from the perspective of the capitalists (some at “non-profit” universities) who have successfully lobbied to operate under rules that aren’t available to non-cronies.

Note that the college-affiliated/VA-funded Upper Limit school discussed in the article charges $600/hour for the Robinson R44. At East Coast Aero Club, operating at a much higher cost airport in a state with a much higher cost of living, we charge $329/hour, including fuel, for the same aircraft. Some of the veterans also fly around in circles in a Bell 205, the civilian Huey, with nine empty seats in the back. The companies that employ low-time helicopter graduates operate Robinson R22s and R44s, not Hueys, so there is no practical value to this training. It will be many years before the graduate gets into a helicopter like the Huey and that will be after a thorough operator-run training program (probably for a European-designed-and-built Airbus helicopter, whose rotor system spins in the opposite direction and therefore requires opposite pedal inputs; Bell got so fat from its own government contracts that they didn’t bother investing in new designs and has been steadily losing civilian market share to the Europeans).

[It is only fair to note that the total dollars involved in this program are negligible compared to the money spent by the federal and state governments on things such as health care, employee pensions, etc. Whether the VA flight school program continues in its present form or not won’t make any difference to U.S. prosperity.]

Related:

• February 2014 report from Heli-Expo explaining that a flight school could bill the Feds $10,000/year for instrument training in a Gulfstream if they wanted to
• April 2014 report from Sun n Fun talking about how “Startup flight schools are overtaking 50-year-old schools”

Full post, including comments

“Break-in at Y-12” is a New Yorker story about three pacifists (one an 82-year-old nun) who broke into one of America’s most heavily guarded nuclear storage facilities. One of my favorite parts of the story is how the government calculated its damages from the graffiti scrawled by the miscreants.

Full post, including comments

The NTSB factual report on a November 10, 2013 crash of a Mitsubishi MU-2 says that the pilot had 11.5 hours of dual in the airplane and that was also his total turbine time (though he had a fair amount of multi). He elected to fly solo after 11.5 hours of MU-2 time and crashed just before what would have been his first solo landing. His family is now suing Honeywell, et al., claiming that an engine decided to fail a minute or so before arriving at the runway (story). The Honeywell TPE331 pilot notes say that “a mean time between in flight shut down of more than 63,000 hours has been attained.” Note that this is not the mean time between failure. This includes engines that are shut down because an oil pressure or temperature gauge becomes defective, for example. If we take the last five minutes of flight as critical due to the proximity to the ground, the lawsuit essentially says that it is more likely that a 1 in 756,000 probability event occurred (engine shutdown during a randomly selected five-minute block of operation time) than that a pilot with virtually no turbine experience or experience with the MU-2 crashed the notoriously tricky airplane through pilot error. (The MU-2 is probably easier to fly than a jet from the same era but due to its lack of turbojet engines or weight over 12,500 lbs. it did not require the special type rating that the pilot of a jet needs. This led to a high accident rate, the imposition by the FAA of some special training requirements, and the devaluation of used MU-2s to about half the price of comparable Beechcraft King Air turboprops.)

[Note that it is unconventional for someone with this level of training to be flying solo. Insurance companies typically like to see the new owner of a complicated airplane with a “mentor pilot” for between 10 and 50 hours following the completion of an intensive training course. A 40-year-old MU-2 is a particularly bad plane in which to solo as it presents the pilot with a whole wall of switches and gauges (Google Image Search).]

It will be interesting to see what the jury makes of this case.

Related:

• mu-2aircraft.com

Full post, including comments

Michael Stonebraker came to talk to our three-day lab course on SQL programming. He’s a great example of why the research university should not be shut down precipitously. Here’s what I learned from him about the current state of database management systems.

For transaction processing (“Small Data”), the standard RDBMS variants work pretty well as long as the database fits into RAM (1 TB is a reasonable size) and as long as you’re not trying to do more than about 1,000 updates per second.

For high-volume transaction processing you need a new architecture where everything is in RAM, which means that transactions happen fast enough that the system doesn’t need to try to be doing 100 of them simultaneously, in various stages of completion. How is the Durability part of the ACID test met if everything is in RAM? Stonebraker points out that you need to have real-time failover anyway so why not let the failover system give you the D in ACID? If that’s not good enough, put an uninterruptible power supply behind both servers. It turns out that customers don’t actually trust this so everyone takes a 5 percent performance hit and logs transaction requests to an SSD. (The log is what the DBMS was told to do, not a data log of blocks in their old and new states.) Stonebraker has drawers full of companies that he has started for every possible database management challenge and his personal solution in this area is VoltDB.

For traditional “business intelligence” or “data warehousing” queries, the column-oriented shared-nothing DBMSes such as Vertica (yet another Stonebraker-founded company, sold to HP) end up being 50X faster than row-oriented DBMSes (e.g., Oracle, MySQL). Why? The database is too big to fit into RAM and you’re usually interested only about 1/50th of the columns in any one query. Thus the system needs to fetch and scan only about 1/50th as much stuff from disk as would a row-oriented DBMS.

What customers want and doesn’t exist right now is a DBMS to handle Big Data and Big Analytics. This will be an “Array DBMS” and it will be good for machine learning, clustering, trend detection, etc. The Array DBMS will be good at handling the common “inner loops” of “Big Analytics” such as matrix multiply, QR decomposition, SVD decomposition, and linear regression. Somehow I have a feeling that this might be Stonebraker’s next company!

Full post, including comments

Technology Review writes about the Sorek plant in Israel (link), which produces a week of water for one person for less than 58 cents. This may not be sufficiently cheap for competitive agriculture, but it would seem to contradict the prophets of aquatic doomsday (e.g., see “The Coming Water Wars: The next big wars will be fought over water.” from U.S. News).

What do readers think? If I can get some Chinese solar cells to charge up my Tesla, drive to the water factory, also powered by those solar cells, and fill up my travel mug, why would I go to war? (And of course, as my friend Rob says when offered bottled water: “Wouldn’t it be great if there were a pipe going to every building. Then we would not to have haul water around by truck.”)

 

Full post, including comments

Part of Ellen Pao’s direct testimony in her lawsuit against Kleiner Perkins included that she was having sex with a co-worker whom she knew to be married (Business Insider). Presumably everything in Pao’s “case in chief” is intended to persuade the jury that they should be sympathetic toward her. Thus we can take this as a barometer of how times have changed in the U.S. If we were to go back 50 years, for example, it seems unlikely that a plaintiff seeking sympathy from a jury would have started out “So I was having a multi-month adulterous affair with a married guy at work…” Perhaps a public statement like that would be made on some sort of television investigative news program and the person confessing his or her part in what was then a crime would been shown with a face obscured in shadow but certainly nobody would have expected to be paid $2.2 million in 1965 (equivalent to the $16 million Pao seeks today, adjusting for inflation) after relating a tale of this sort.

Separately, I wonder what the jury will make of Pao’s complaints that “After she filed the lawsuit, Pao said the work environment at Kleiner Perkins became ‘extremely difficult and very uncomfortable. Partners were uncomfortable talking to me… if I went to their office they seemed nervous.'” (arstechnica) Maybe someone could sue General Motors and expect to keep working on the assembly line. The money sought from litigation would be coming out of the shareholders’ pockets, not the personal checking accounts of co-workers. But Kleiner Perkins is a partnership. Pao was essentially suing these people individually and then being surprised that it was hard to work with them after that. (There is some precedent for Pao’s expectation. Unlike some Scandinavian countries that have an administrative process for divorce (98 percent of Swedish divorces happen in this non-adversarial manner), in the U.S. the government encourages one parent to sue the other and then pays workers, such as judges, to express dismay at the resulting lack of cooperation between the litigants. We just finished analyzing some data from all 243 May 2011 divorce lawsuits in Middlesex County, Massachusetts. Only about 16 percent of the couples had filed a “joint petition” indicating that they had gone to a mediator rather than one running to the courthouse to sue the other. So 84 percent decided that a good way to embark on the next phase of a multi-decade co-parenting relationship was a lawsuit.)

Finally, after reading the news coverage of this case I am wondering if I myself might have a case against Kleiner Perkins. This New York Times article is typical: “Ms. Pao, who came to Kleiner with the dream of helping direct such a fund, graduated from Princeton with a degree in electrical engineering. She got a law degree from Harvard and worked for Cravath Swaine & Moore for two years doing international deals. She returned to Harvard for a business degree and worked for a variety of tech companies, including BEA Systems and Tellme Networks. Her geek cred is pretty unassailable.” In other words, a person who got an undergrad engineering degree and then never engineered anything has “geek cred” and, implicitly therefore, is entitled to be selected as a senior partner at Kleiner Perkins. What is the basis of the Philip Greenspun v. Kleiner Perkins lawsuit then? I have an engineering degree and have actually had jobs doing hands-on software engineering and electrical engineering. So the Greenspun geek cred should be at least as strong as Pao’s. Like Pao I have been inside the Menlo Park offices of Kleiner Perkins (not to have sex with a married employee, but to talk about some of their portfolio companies’ projects and to attempt to hector them into finding a team and building a startup around my “Mobile Phone as Home Computer” idea). As happened to Pao, the Kleiner Perkins partners with whom I spoke failed to recognize my potential to join them as a senior partner. Instead of issuing me $16 million in paychecks they escorted me to the front door and into my rental car.

Related:

• previous posting regarding Ellen Pao v. Kleiner Perkins

Full post, including comments

Apple is now taking $200 of electronics that typically retails for $349 and will be worth $0 in a few years, wrapping it in $640 of gold (source: Forbes), and selling the combination for $12,000 as “Apple Watch Edition.” I’m looking at a 6-year-old Panasonic 1080p plasma TV on the wall. It might be worth about $349 on Craigslist and will be worth $0 as soon as 4K goes mass-market. I am going to remove the black plastic bezel and replace it with a metal bezel containing some gold, fusing the two together so that they can never be separated. The result will be the $12,000 “Apple TV Edition.” The weight of the metal bezel will not generate complaints such as this one from Bloomberg about the Apple Watch.

How many people are out there willing to spend $12,000 on a fancy case inextricably linked to almost-immediately-obsolete electronics? Vertu supposedly found about 350,000 customers for its expensive-on-the-outside phones. I’m guessing that Apple could sell at least 10 times that number over the next three years. That’s $4.2 billion or a little over a week of revenue for the company.

What do readers think of the $12,000 Apple watch? Is it more evidence for the philosophy of David St. Hubbins? (“It’s such a fine line between stupid and clever.”) On which side of the Hubbins Line does the watch fall?

Full post, including comments