Category: Uncategorized

A friend sent me this editorial cartoon showing the federal budget and proposed $85 billion “sequestration” put into perspective with a pie illustration. I think it a fun data visualization exercise, but when I looked a bit deeper I became skeptical. The cartoon shows federal spending as having grown 40 percent since 2007, from $2.7 trillion to $3.8 trillion. That did not seem credible but it is backed up by

• http://en.wikipedia.org/wiki/2007_United_States_federal_budget
• http://en.wikipedia.org/wiki/2013_United_States_federal_budget

What is it that we’re getting 40 percent more of?

[Related: my own attempt to make federal budget numbers comprehensible by dividing by 10 to the 8th power (April 2011). This analysis caught the public imagination and appeared in a bunch of magazines and Web sites, usually with the original source long forgotten.]

Full post, including comments

The cruelly objective folks at DxOMark have tested the Sony RX1, a camera that has inspired Leica-like devotion (e.g., see the reader reviews at Amazon). It turns out not to be that exciting from an objective measurement point of view. The fancy Zeiss lens tests out with similar numbers to a Canon 35/2 lens and it lacks the Canon’s image stabilizer. The Sony’s Zeiss lens does come up with a better overall score but the notes say that is mostly because the Sony sensor is better than the Canon 5D Mark II sensor that was placed behind the Canon lens for the test. How about the sensor? Really great.. just like the same sensor, more or less, that Sony sells Nikon to put into the D600.

So for the price of a Nikon D600 and a good 35mm lens you get the same image quality and seriously compromised picture-taking capabilities compared to a standard DSLR (see the conclusions in the dpreview.com review).

The Sony does seem like kind of a fun toy for people who love the 35mm lens perspective (I’m not one of them) but the test results don’t show it to be magic compared to a modern Nikon body and well-engineered 35mm prime lens (e.g., from Sigma!).

Full post, including comments

A friend of mine passionately follows New York Times articles about new medical diet studies, e.g., “eat blueberries and you won’t get Disease X.” My response to her admonishments to read and then adhere to these diet fads was always “You can read all of the best medical literature on diet and by the end of the project you won’t know any more than a good physician in Ancient Rome. The Roman doctor would have said ‘Don’t eat too much meat. Eat a lot of vegetables. Maybe some fish. Get exercise.’ What more or different would a modern doctor say?”

Now it turns out that even the New York Times is confirming what I was telling my friend for the last 10 years with “Mediterranean Diet Is Shown to Ward Off Heart Risks”. After 2000 years of progress the 10 lb. heads behind the study advocate that people eat “olive oil, nuts, beans, fish, fruits and vegetables, and even drink wine with meals.”

So that gets back to my question… what do we know today that the Romans didn’t know about diet and lifestyle? The main thing that I can come up with is the dangers of lead exposure.

[Separately, this video is a very thoughtful analysis of the latest diet bestseller… Wheat Belly.]

Full post, including comments

Yesterday’s New York Times carries an article entitled “Fear of U.S. Cuts Grows in States Where Aid Flows.” Google Chrome says that the word “cut” occurs 32 times on the page. It sounds like a natural disaster: “The impact would be widespread as the cuts ripple across the nation over the next year”. Yet buried near the bottom is an interesting paragraph: “Even with the automatic cuts, the analysis found, states are still expected to get more federal aid over all this year than they did last year, because of growth in some of the biggest programs that are exempt from the cuts, including Medicaid.”

So.. federal spending is being cut. Which is why more money will be spent this year than last year.

Full post, including comments

Sheryl Sandberg is getting people excited about her new book, Lean In. It seems as though this would be a good time to add to the “Aid to Evaluating Your Accomplishments” page (hit reload a few times to see the names change). Most of us would be happy either to (1) care for an infant child, (2) run one of the world’s most valuable companies, or (3) be the author of a bestselling book. Sandberg is doing all three simultaneously.

Sandberg’s life sounds pretty good but not every female executive can rise to such levels; there simply aren’t very many jobs as good as being Facebook COO. The typical female (or male) executive can expect to end up somewhere in the middle of a pyramid.

The men that I know seem to seek promotions uncritically. They don’t need a much larger salary in order to take on a larger responsibility. They’re happy to go from “Director” to “Managing Director” or “VP” to “Senior VP”. As Napoleon said, in reference to medals, “It is with baubles that men are led”.

Some women friends though, tell a different story. One who lives in San Francisco is typical. She earns a very comfortable salary, more than $200,000 per year, managing a small team. She was offered a promotion recently that would have paid her about $9000 per year additional. She cited federal income taxes, California state income taxes (among the highest in the nation at up to 13.3 percent), and an additional San Francisco city payroll tax in calculating that at most she would be able to spend about half of the additional money, netting perhaps $4500 per year. She decided that it wouldn’t be worth it because the bigger job would involve more hours of work and more travel time. Her hourly after-tax wage would actually have fallen. Other women talk about being bored to death watching PowerPoints in endless meetings. The higher up in the pyramid, the more time spent looking at PowerPoints and the less time doing anything productive and therefore satisfying. Asked about the value of titles, a female MBA said “I care about the hours that I have to work and the salary that I get paid. They could call me ‘secretary’ and I would be just as happy.”

It is far from obvious that it is rational to want to climb a corporate hierarchy. There are a lot of good jobs near the bottom of the pyramid for talented people. One can get paid $250,000 per year without having to manage anyone. If you don’t manage anyone you can work flexible hours and not lose sleep at night over whether or not a subordinate will complete an assigned task. The first step into management brings a radical reduction in quality of life and, typically, only a small raise. Being a COO or CEO is great (and being fired from a CEO job is even better; Robert Nardelli got paid $210 million to stop working at Home Depot) but being a middle manager is not necessarily a great job. As there are thousands of middle management jobs for every CEO or COO job the probability that a career in management will lead to one of those great jobs is tiny.

Sandberg and others posit complex reasons for women failing to claw their way slightly higher in management pyramids. Perhaps part of the answer that women, on average, do a more rational cost/benefit analysis.

Related: http://philip.greenspun.com/careers/women-in-science

[Note that the same can be said about programming jobs, albeit typically at lower wages than the ones above. To be a programmer can be heaven. To manage programmers is often hell. Why go from heaven to hell for a 15 percent raise that, after taxes, will have only a tiny effect on one’s lifestyle? If money is that important, why not do a little consulting on the side instead?]

Full post, including comments

Politicians and newspapers talking about upcoming scheduled cuts in government spending are concentrating on the dire consequences from any cuts to the Federal Aviation Administration (FAA). The implication is that this is an agency with no fat and any cuts will go straight to the bone. (They probably felt that way in Spain when they couldn’t pay air traffic controllers $1.2 million per year anymore; in 2011 they started a process of competitive bidding for control tower operation and the results have been a roughly 50 percent cost saving (source).)

This gives me a chance to recycle my June 2011 posting about how three FAA employees came out to my house on two separate occasions and from as far away as Florida in order to make sure that I was surprising myself with random drug tests… of myself. Also to point out the 2004 Government Accounting Office report that says “Historically, the modernization program has experienced cost overruns, schedule delays, and performance shortfalls of large proportions and has been on our list of high-risk programs since 1995. To date, FAA has spent $41 billion and expects to spend an additional $7.6 billion through fiscal year 2007.” Have things improved since then? This fiscal year 2013 report says “Increasing airspace capacity and reducing flight delays depend on the successful implementation of the En Route Automation Modernization program (ERAM)—a $2.1 billion system to replace hardware and software at FAA’s facilities that manage high-altitude traffic. FAA originally planned to complete ERAM by the end of 2010. However, software problems have impacted the system’s ability to safely manage and separate aircraft and raised questions as to what capabilities ERAM will ultimately deliver. FAA rebaselined the program in 2011, which pushed its expected completion to 2014 and increased cost estimates by $330 million.” Four years and $330 million over budget… but actually “If software problems persist, the program’s cost growth could exceed $500 million, and delays could stretch out to 2016.”

It seems safe to assume that the FAA is representative of other federal agencies, neither dramatically worse or better in terms of efficiency.

If the unprecedented did occur and federal spending were to be cut it would be interesting to see what these agencies would do. Companies get more efficient when revenue falls but they have to worry about competitors. A government agency with no competition could simply cut back on services delivered. On the other hand, the agency could do that even without a budget cut if they were truly indifferent to output and productivity. Perhaps it is fear of being privatized or having responsibilities assumed by another agency that drives government agencies to try to deliver services. If so, in the event of a cut they should be motivated to do as much as possible with the available funds.

Full post, including comments

The New York Times published an article yesterday about how people who donated money to senators and election-related groups such as the Democratic Senatorial Campaign Committee earned billions of dollars in additional profits from selling software that typically has few benefits for patients or society (see this January posting).

In that January posting I talked about how a system such as Google Docs would deliver, for free, most of the benefits of the systems for which our society is now spending tens of billions of dollars. One of my spies within Google sent me the following email in response:

In the days immediately after the Haiti earthquake there was a severe problem around collecting basic patient information and records… folks were getting moved between medical camps, doctors were moving around, all sorts of workers from different non-profits were helping out in different areas with no information sharing standards. Frequently patients would go to different doctors and due to the lack of infrastructure – buildings, power, etc., there were basically no medical records or patient histories being collected ever, which caused problems if one doctor prescribed medicine and then the patient went to another doctor, lots of wasted effort collecting the same information every time a patient came in, etc.

Anyway the google.org engineers came up with a total hack that they wanted to try out – create a Google Apps for Business domain, stick every doctor and nurse into it with an account, and then setup an enormous domain-wide shared Google Docs folder where anyone could put patient histories. Each patient history would be a single doc with the patient’s name and everything was appended as text. They thought the cell network could be kept up and running on generators so they made a really simple iPhone app (at the time very few doctors were using Android) that used the Docs APIs to make it a little bit easier to find patient records and append text to them.

They got the whole thing working in a couple of days; I don’t know how widely it was deployed, I was only involved briefly at the beginning when they needed to figure out how to set up the domain and provision users and came to the Enterprise team with questions. It was interesting how the requirements affected their design – because of the totally absurd time-crunch – people were dying and they needed to make efficient use of the doctors – basically all the requirements around privacy of information, structured records, limiting who can see whose data and keeping audit trails, etc – went out the window. And once they were gone it was actually a really easy system to build and deploy.

It is a little bit interesting that the New York Times never loses its enthusiasm for Big Government. They publish articles lauding proposals by politicians to spend billions in taxpayer money on something that is supposed to do a lot of good. Then a year later the newspaper will publish an article about how great it is that the do-gooding is actually happening. Then a year or two later the newspaper will do a follow-up about how much or most of the money turned out to be wasted, funneled into the pockets of cronies, etc. These cycles continue, usually about 50 of them in parallel, without the Times ever running an article on how government spending tends to be wasteful and to result in the enrichment of cronies.

Full post, including comments

A friend of mine is involved with Kids on Computers, a non-profit organization that sets up computer labs in schools in poor/remote regions of the world. This is how my old laptop computers make it down to Oaxaca. I have a simple Windows 7 Lenovo 13″ laptop whose broken screen did not seem worth replacing until I tried Windows 8. As an experiment, and with the ultimate goal of eventual donation to Avni Khatri and Kids on Computers, my friend John and I decided to see how hard it would be to replace the Lenovo’s screen.

It turns out that buying the screen is a little tough. Given the number of laptops that are dropped you’d think that Amazon.com would sell these or that they’d be at the local Best Buy, but in fact http://www.laptopscreen.com/ was the only source that we could find. A $65 investment, including shipping, resulted in the arrival of a new LCD screen in a box. The site has helpful videos showing consumers installing the screens. Using some inappropriately large household tools, John and I were able to get the new screen installed and working in about 10 minutes.

I wonder why this isn’t a more popular home craft project given that it actually takes less time than driving to a store to have someone else do it (and a quick Google search indicates that repair shops charge a lot more than $65 to do this work).

Full post, including comments

Folks:

I came back from Antarctica with a bunch of panoramic photos made with a Sony DSC-RX100, a truly awesome point and shoot camera that, set to “green idiot mode”, produced far better JPEGs than the $3000 Canon EOS 5D Mark III. The little Sony has a nice feature for showing panoramics on its rear LCD. You press a button and the photo is expanded to occupy the full vertical height of the rear LCD and then slowly scrolls from right to left.

I’d like to replicate this interface on a Microsoft Windows 7 or Windows 8 computer, but so far I can’t find any software that does it. All of the software is either entirely (1) unaware of panoramas and has no option to set the zoom so that the height of the image matches the height of the monitor, (2) a little bit aware and premised on the idea that the viewer wants to be active with a mouse to pan and zoom.

Ideally I would like a slide show that displays one panorama at a time, first shrunken to fit and then panned right to left or left to right. I guess this could be done in Javascript but it seems that the software should exist already somewhere.

Ideas?

Here’s an example of one of the photos (click twice to download the original file):

Full post, including comments

Looking at the news today, some friends and I were attracted to photos of the Carnival Triumph being towed back to port and of Hugo Chavez’s daughters visiting their beloved papa in a Cuban hospital. As there presumably won’t be a tremendous near-term demand for leisure cruises on the Triumph it occurred to me that perhaps the engine-room fire on the Triumph represents a great opportunity for the U.S. taxpayer.

Seventeen percent of Floridians are age 65 and older, i.e., nearly 3.5 million people. That’s the age at which an American becomes eligible for Medicare, i.e., a cash cow for the local doctors and hospitals. America is chronically short of doctors while Cuba has a large surplus of physicians. Cuban doctors must be doing a pretty reasonable job since life expectancy in Cuba is higher than in the U.S. Also, it seems doubtful that Hugo Chavez would choose second-rate care for himself when his life is at stake.

The Carnival Triumph can cruise at about 22 knots, which means that it could cross the 90 miles of water that separate Florida from Cuba in about 4 hours. Why not set the ship up as an ambulatory care clinic staffed with Cuban doctors? The ship can sail every day from Florida to Cuba and back. Any Medicare patient who can be treated on board will enjoy the round-trip sail, the waterslide park, and the rest of the amenities on board. Any Medicare patient whom the doctors deem to require more extensive treatment can get off in Cuba and be admitted to a hospital there for a procedure to be performed at a tiny fraction of the cost to the U.S. taxpayer.

The Triumph would leave every morning at around 8:00 am. Medicare clients would enjoy a Cracker Barrel breakfast on board the ship. The ship would arrive in Cuba at 12 noon. Those who were well enough to walk could enjoy a stroll around Havana. The Triumph would pick up patients returning from hospital care in Cuba and anyone who’d been enjoying the sights, then depart around 2:30 pm. An early bird special dinner would be served on board starting at 5 pm, with an arrival back in Key West at 6:30 pm.

As the Triumph holds approximately 3000 people, approximately 1 million patient-days of ambulatory care could be delivered each year via this means, plus however many days of hospital care delivered to those who stayed in Cuban for a few days. If we assume that each procedure performed by a Cuban team rather than a U.S. team saves Medicare an average of $500, operating the ship should conservatively save U.S. taxpayers $500 million per year while relieving Carnival of an embarrassment.

Anyone have a better idea for what to do with the Triumph?

Full post, including comments