Month: October 2009

Folks: In 2004, I posted a query about the best way to manage a party invitation email spam list (and the bottom of the posting links to the perl script that resulted). I’m going to put the question forward again because technology keeps changing and spam filters keep getting stricter. A fair number of my messages were spam-filtered, even though they were from: me, to: recipient. I think this might have been because they were sent from an old server and old email address within MIT and a lot of spammers had appropriated that identity.

Here are the requirements:

• I can build a list of email addresses
• I can spam this list with a common message, e.g., “party on Saturday”
• The recipients do not have to visit a Web site to receive the full message, nor deal with a massive pile of HTML and graphics in the email (I believe this requirement rules out the commercial Evite service)
• Message should be plain text readable on a mobile phone
• Messages are not intercepted by spam filters

It would be nice to have the following:

• ability to tag some people with extra information, e.g., “helicopternerds”, and spam only them or spam the list minus them
• recipients can remove themselves from the list

I would have thought that Gmail or Yahoo mail would have something like this built-in. Yahoo seems to have the ability to create a mailing list, but I’m not sure what the resulting messages look like going out. Gmail doesn’t seem to have anything like this (if you say that you want to email a subgroup of contacts it does the obvious stupid thing of adding them all to a big To: list). Could I create a Google Group and stuff all of my friends in it without their consent? Even then it doesn’t seem ideal for managing a social list because I don’t think there are easy facilities for tagging people as belonging to subgroups.

Doesn’t this seem like a sufficiently common problem that it shouldn’t require Unix shell programming?

Full post, including comments

Wall Street banks have had profitable quarters. JPMorgan Chase reported $3.6 billion in profit (more than $1 billion per month). Goldman Sachs was only slightly behind, at $3.2 billion. These profits supposedly came from “trading.” I asked a friend who has worked in the money business how this was possible. “For someone to make money trading, there has to be someone on the other side of every trade who is losing money. Where does each bank find someone who can lose $1 billion every month?”

He explained that “carry trade” would be a more accurate description of what they’re doing. Because of the Collapse of 2008 financial reforms, the big investment banks are able to borrow money from the U.S. government at 0 percent interest. Then they can turn around and buy short-term bonds that pay 2 or 3 percent annual interest. Now they’re making 2 percent on whatever they borrowed. They can use leverage to increase this number, by pledging some of the bonds that they’ve already bought as collateral on additional bonds.

I asked if they were taking any risk in order to earn this return. “If interest rates went up to 20 percent, even though the bonds are short-term, the price of the bond could fall enough to make the trade a money-loser.” (Though since the banks are too big to fail, they would simply be bailed out with additional taxpayer funds.)

What kind of bonds are they buying? Are they investing the money in American business? “No, they are mostly buying Treasuries.” So the money is just being shuffled from one Federal bank account to another, with each Wall Street bank skimming off $1 billion per month for itself? “Pretty much.”

[A more old-fashioned way of making supranormal returns is insider trading, which was perfectly legal until the Crash of 1929 (history). The New York Times ran a story yesterday on Raj Rajaratnam, a hedge fund manager who invested heavily in inside information. Rolling Stone published “Wall Street’s Naked Swindle” on October 14. The story is much more sensational and entertaining than anything from the Times. It covers a guy who spent $1.7 million on out-of-the-money put options on Bear Stearns on March 11, 2008. The options would become worthless on March 20, just 9 days later, unless Bear Stearns basically went bust. Bear Stearns collapsed the next day and the guy made a $270 million profit. He has never been identified by the SEC.]

Full post, including comments

Andrew Grumet and I have drafted an article on software design review and would be grateful for comments (persistent comments underneath the article itself; comments about what should be changed in the article or typos should be posted here on the blog). Software projects would seem to be extremely amenable to external design review and yet though this practice is common in business agreements (have a lawyer look over a letter agreement) and in construction (builders bring in an architect or an engineer; architect brings in an engineer to review the structural design), it has been almost unheard-of in software development. A team of programmers is selected and then is able to do more or less whatever it wants.

Full post, including comments

… mostly by themselves. This AOPA article reveals that the relaxation of regulations for pilot and aircraft certification did not result in a skycar for every garage. In fact, only 1688 LSA planes have been registered and only 3064 sport pilots certified. There are roughly 200,000 pilots in the U.S. who fly primarily for private or recreational purposes and also roughly 200,000 airworthy planes that an individual could reasonably afford, so Light Sport has not had a significant effect.

Training minimums have been reduced from 40 hours to 20. The price of a new two-seater has been reduced slightly, though it is still substantially higher than that of an older airworthy four-seat certified airplane. Perhaps the sad answer is that Americans are too busy working (to pay for their health insurance!) to take up a hobby that requires a lot of time, effort, and dedication.

Full post, including comments

Two of the great minds that the New York Times has picked for its editorial page debate the future of the U.S. in “The Economy is Still in a Funk”. David Brooks concludes with “The economy has always bounced back. There is an energy to the country so that after a time the animal spirits get moving and before you know people are starting businesses (even if there are slight chances of success) and expanding current ones, and a few of those turn into Apple and ESPN.”

None of the 100 or so comments seem to note Mr. Brooks’s survivorship bias. The U.S. economy has prospered to this point and therefore we believe it will always prosper… because we live in the U.S., a country whose economy has happened to do well for several hundred years. Plenty of folks in Argentina thought that way as well. Each bust was followed by a boom. Until it wasn’t and Argentina slipped from having a comparable level of wealth to that of the U.S. to being a relatively poor country. The U.S. economy has never faced the challenges of a health care system that costs 20 percent of GDP, of staggering pension obligations (mostly for public employees), of effective competition from China and India, or of global warming. It is, of course, quite possible that we will recover from our current recession as we have from earlier recessions. But past recoveries don’t guarantee anything about the future. It is possible that the U.S. recovered in the past because it was one of the world’s best places to do business, not because it was the U.S. In that case, we would have to do whatever was necessary to ensure that we were an attractive place to invest (see my economic recovery plan, for example).

Full post, including comments

If you’ve given up on The New Yorker‘s coverage of business management due to excessive Gladwellization (the latest work by North America’s greatest thinker asks “Is football any better than dogfighting?” and uses 9 full pages to show that in both activities there are likely to be injuries), reading “Not So Fast” may restore your faith in the magazine. This piece by Jill Lepore covers the early days of management consulting and business education. The most interesting part is the second half, which concerns the life of Lillian Gilbreth, mother of 12 children, Ph.D. in Psychology, author of several pioneering books on scientific management, and inspiration for the movie “Cheaper by the Dozen.” She had done a lot of her work under her husband’s name and when he died in 1924 found that businesses would not pay for advice from a woman-run enterprise. Despite her lack of any experience or competence in the kitchen, she remade herself into a home economics expert and died at the age of 93 in Scottsdale, Arizona in 1972 (before Phoenix sprawled out to swallow it!).

More: read the full text of the article

Full post, including comments

A friend from graduate school visited this weekend with his three kids. Before they departed, I began to feel my throat getting sore. Yesterday the parents called to say that one of the children had tested positive for strep throat. I called my primary care physician to find out what I should do. His assistant returned my call three hours later, referring me to a lab affiliated with a local hospital. I visited the lab, handing over a health insurance ID card (required by law now in Massachusetts). I would have expected the hospital’s IT system to be able to grab my address and phone number from the insurance company’s IT system, but instead I was asked to hand-write a one-page form with this information. Meanwhile the clerk searched among a stack of 30 or so FAXes that had come in that day, trying to find one from my doctor with my name on it. After she’d found the test order, she started entering my contact and insurance information into the hospital’s IT system. She made multiple transcription errors, necessitating two reviews by me. I was presented with a full page of fine print in which I signed away various legal rights to privacy. The clerk said that without this they would not be able to disclose the test results to my doctor. Nor do they ever disclose test results to a patient. So a patient who refused to sign the form would end up getting tested and the results would never be useful to anyone. The paperwork took about 30 minutes to complete.

The strep test itself was done fairly quickly after the paperwork was done and I walked out, having been told to call my doctor’s office the next day for the results.

I called the doctor today and the phone menu said “Press 1 if you are having a medical emergency or need to talk to your doctor’s office”. I pressed 1 and got a busy signal, then was disconnected. I called again and repeated the process. I was disconnected again. I called a third time and pressed 0 for the operator and explained the situation. She said that my doctor was out today and I should call again the next day. I explained that I had been told to expect this and that anyone else in the office should be able to retrieve the test results and read them. She said that someone would call me back. About 20 minutes later, a different physician’s assistant called me back to say that the “quick strep” test was negative but that they were still waiting on the throat culture test, which had been started in parallel. I asked how long the quick strep test actually took. She said that she had no idea. I looked it up on the Web and the answer turns out to be less than 15 minutes. So it took 21 clock hours to get the results of a test that takes 15 minutes. Two FAXes were sent and at least three additional pieces of paper were consumed; six voice phone calls were placed.; roughly 60 minutes of patient time was consumed, not including driving to/from the lab.

I would say that this incident demonstrates one point from my health care reform plan: “As the health care industry has never been competitive, nor had any incentive to control costs, we have no idea how much American health care could or should cost.”

Full post, including comments

Today New York Times carries an article on the failure of America’s public schools to teach math to kids. In schools that are among the most expensive in the world (for taxpayers), only 34 percent of 8th graders are “proficient” in math, and 39 percent of 4th graders. One major theme of the article is to beat the dead horse of the Bush Administration, by pointing out that No Child Left Behind does not seem to be working. The taxpayers are doing their part, paying up to $200,000 per year for each teacher (including pension obligations incurred). The students are doing their part, presumably, by showing up to school every day for 6 hours. If things aren’t working, it can only be due to incompetence on the part of the government at this fairly straightforward task.

Let’s contrast teaching K-8 math with managing health care. Instead of compliant 4th graders who show up to school every day, you have clever providers who will figure out where the gaps are in thousands of pages of federal rules and regulations and use those gaps to extract tens or hundreds of billions of dollars in extra profit. Instead of a fixed subject that has not changed substantially since the death of Brahmagupta in 668 A.D., health care presents a moving target of new procedures, drugs, tests, and fees. In the school system, the interests of the students and taxpayers are aligned. Both groups are better off if math is learned. In the health care system, there is a substantial moral hazard. If improved diabetes and heart disease therapies become available, people may indulge more in super sized meals.

You would think that the evidence of failure of trillions of dollars of tax money spent on math education in the period covered by the article (1996-present) would be a sobering reminder of the limits of government power, but none of the 50+ people commenting in the New York Times made that connection.

More: On my non-profit ideas page, I propose teaching math in the context of doing an engineering project, such as designing and building a bicycle.

Full post, including comments

I met with my accountant last week. I learned the following:

• advertised marginal tax rates should be ignored by upper middle class taxpayers; the alternative minimum tax (AMT) is the real tax rate and it is substantially higher. Even in a state with an average income tax rate, such as Massachusetts (5.3% on most income), taking a deduction for state tax paid will trigger the alternative minimum tax
• contributions to charity have very little effect on the total tax paid by someone subject to alternative minimum tax; a charitable deduction lowers one’s standard tax, but does not lower one’s AMT. The IRS forces you to pay the AMT if it is higher. So… give to charity if you feel as virtuous as Elvis Presley, who never took a deduction because it “took away from the spirit of the gift”, but don’t expect it to lower your tax bill.
• we do indeed live in a nearly perfect welfare state for health insurers. A self-employed person who pays $5000 for health insurance can deduct part of that. A self-employed person who pays $5000 directly to doctors and hospitals cannot deduct any of that (except the part that exceeds 7.5 percent of income).
• the costs of complying with the Massachusetts requirement to purchase health insurance are substantial; the state had no idea how many people were uninsured so they are using the tax system to figure it out. The accountant calls customers to obtain their proof of insurance certificates. The customers call their insurance companies to obtain these documents in hardcopy. Then they have to be re-mailed. Then someone at the accounting firm has to open the mail, put the document in the correct file and inform the accountant to stop nagging the client. These costs imposed on taxpayers are in addition to the billions of dollars in direct costs for additional policies purchased and for state subsidies to insurance companies for customers who aren’t wealthy enough to afford what are now the nation’s highest cost policies. (more: see my health care reform plan)

The tax code is becoming ever more complex. If you buy business equipment, for example, you have to calculate depreciation both for regular tax and for AMT. This calculation is done every years for 5 to 10 years. If you rent out an apartment, you pay tax on any income. If you lose money, though, you can’t deduct the loss. You are supposed to accumulate any loss years and subtract the total from whatever you get when you sell the apartment, so the operating loss turns into a reduction in capital gain. (Unless you’re a Congressman in charge of the committee that makes tax laws for commoners; in that case you don’t pay tax on your rental income and you don’t pay market rent on the four apartments that you occupy in Manhattan (see Charles Rangel)).

My taxes are pretty simple because I don’t trade individual stocks. All of my investments are in mutual funds. I do have some self-employment income and I rent out an apartment that I own. The number of pages of filled-out forms and schedules, for both state and federal taxes, that the accountant sent me for review is 124.

Full post, including comments

Barack Obama has won the Nobel Peace Prize because of his sincere expression of a desire for world peace. I’m excited first because I’m hoping that he will use part of the $1.4 million proceeds to take his Aunt Zeituni off the hands of the taxpayers of Massachusetts (more). Second, I’m predicting that the next prize will go to Sandra Bullock due to her expressed wish for world peace in the movie Miss Congeniality (preceded awkwardly, as cinema fans may recall, by a desire for “harsher punishment for parole violators”). Ms. Bullock will likely have to share her prize with current and former Miss Americas and therefore may not be in a position to relieve the citizens of Massachusetts of the burden of paying for President Obama’s aunt’s apartment.

Full post, including comments