Author: philg

Yesterday’s New York Times article has an article entitled “Blacks Less Likely to Take A.P. Exam.” The article has two points. Blacks are less likely than other high school graduates to take the Advanced Placement tests. Blacks who take the test are much less likely to pass than non-black students (8 percent of those taking the exams checked the “I am black” box; only 4 percent of those passing the exams had checked the “I am black” box).

Does the utility of this information justify its collection? We’re not willing to do anything drastic to change the way that our schools are run. We will still have unionized teachers, teachers who jobs are guaranteed (tenured) after a few years, and all but the richest kids forced to attend their local public school. If schools are failing black students, we’re apparently more comfortable with that than we are in confronting the teachers and administrators who benefit from the status quo.

Medical doctors say that “Never order a test unless you know what you’re going to do with the result.” What are we going to do with this result? Fire teachers if most of their students fail the A.P. exam? That would violate the union contract and we’re not going to do it. Increase the budget for schools with a lot of black students? Many of these schools already spend close to $25,000 per student per year (much of that is spent on administration rather than classroom instruction). Absent an economic miracle there is no way that taxes or debt can be increased to enable an increase beyond 25,000 current dollars per student annually.

Let’s consider the negative aspects of collecting these data. A student walks into the testing room. She thinks that she is going to be judged on the accuracy of her answers. The first question that she faces, however, is “What is your race?” The second will be “Are you male or female?”

Unless we’re going to guarantee today’s 17-year-olds that they will enjoy a lifetime of race-based university admissions, public and private employer hiring, and government contracting, why introduce them to the idea that the color of their skin is going to be the primary thing that matters to other people?

Another negative to collecting and publishing these data is the effect on employers and people involved in college admissions. The simplest one-line summary for the New York Times article would be “blacks are not as smart or well educated as people of other races” (it might have been written by William Shockley!). That’s probably what the average reader will remember in his or her subconscious for the next few weeks. The black guy who goes looking for a job today will definitely not be helped if the person interviewing him has read yesterday’s Times. If his interviewer has a Ph.D. in Statistics, perhaps the interview will go okay as the interviewer can recognize that an individual sample may fall anywhere within a probability distribution. But for most of us the last thing that we need are unconscious racial biases introduced by non-profit organizations (the College Board) and the media (the Times).

Full post, including comments

Consider a university teacher. She wants to hold a class in which the students work together to solve a problem. To make the example concrete, suppose it is an economics class in which the students have to come up with a $1 trillion stimulus plan for the U.S. Or suppose it is an engineering class in which the students have to come up with a design for an exercise machine that can function as a first-person shooter videogame controller (my personal favorite solution to America’s obesity problem).

During the class, the best student ideas will need to be highlighted and expanded upon. The unworkable ideas will need to be pushed into the background so that they don’t clutter the discussion. If you ask Jane Professor how many students she can accommodate in an “active problem-solving seminar”, she would probably say no more than 15. Why so few? Jane is going to have to be the one to direct the discussion, suppress truly bad ideas (like my personal economic recovery plan!), highlight the best ideas, and summarize some of the results.

The students should enjoy a course like this because their efforts might in fact be productive. Perhaps the engineering course is sponsored by a company that can produce the product if the design is good enough. The students should also learn more from this course than from a standard “here’s some knowledge handed down from on high” because they have to be active thinkers. We’re so enthusiastic about this method of learning that we tell Jane Professor we’re giving her 100 students intead of 15. What does she do? Go back to the traditional lecture and homework format. She can’t moderate a discussion among 100 students. Even if she were to run a standard Web-based discussion forum system for the 100 students it would be too much work to moderate. The loudmouths and first answers would dominate, not the best answers.

Two questions then, for the assembled readers…

1. would it be practical to use software to assist a group of 100 students develop a solution to a problem?
2. if so, how much of this software already exists?

At a minimum, the software would need to provide distributed features to allow some ideas to be flagged as more important. Perhaps the authors of those ideas would thereby earn some “karma points” (slashdot style) that would enable them to moderate up or down other ideas.

One big area where I think traditional discussion forum software falls short is that there is a strong presumption in a discussion forum software that the most recently posted content is more important than older stuff. If the challenge for the semester is a complete solution to a design problem, every part of the solution is equally important, whether that solution was developed early in the semester or later. Also, we need something that lets a group of people collaboratively develop a hierarchy.

Wikipedia seems like an example of a collaboratively developed solution to the problem of “how do you write an encyclopedia”. There isn’t very much identification of content with authors, however, and it becomes tough to know who should get credit for various pieces of the solution (in a university course, you need to make sure that each student is doing something!).

Full post, including comments

My friend John and I want to make some instructional videos for streaming from within my “how to fly” Web pages. A first attempt is available on YouTube: “Briefing an Instrument Approach.” I have a few questions about the quality of this video for those of you with powerful PCs. On my Dell XPS from 2006 (two CPUs), the video is slightly choppy and the audio seems out of sync of the video. The video was captured with a 3-CCD Sony HDV camera in 1080i (state of the art in 2006!) and exported to 720p with Adobe Premiere Pro CS4. The audio was captured at 48 KHz but converted to 44.1 KHz for the export, which is the default behavior when you say “YouTube HD” to Premiere. We used a Westcott daylight-balanced fluorescent Spider Light TD5 and soft boxes for lighting.

So.. for those of you with modern quad-core PCs… how does the video look? Do I have Parkinson’s?

My second question is whether YouTube is the right service to use. I don’t want to pay for the bandwidth to stream these videos. At the same time, I don’t want the videos to be interrupted with a lot of ads. Even worse, it would be bad if competitive flight schools ran ads inside these videos. Aside from the satisfaction of helping others learn, it would be nice if we got a few new students at East Coast Aero Club. Nor would we want other flight schools to embed these videos in their own sites, chop off the opening and closing titles (with our URL), and use them to promote their own operation. (I recognize that it is unrealistic to ask someone else to pay for all of the streaming infrastructure and not eventually stick ads in the videos.)

My third question is what do video nerds think about the overall quality of production? Should we give up before we embarrass ourselves further?

My final question for video experts comes from my friend Ken. What consumer-grade HD camcorder should he buy to make videos of his kids? Was there anything interesting announced at CES? I think his budget would stretch to $1000 but it would obviously be better to bring this in for closer to $500. And what’s the deal with hard drive camcorders with 80 GB drives? Aren’t there 64 GB flash cards now?

Full post, including comments

Congress completed its attack on the U.S. economy today by voting (in the House) to delay the shutdown of analog TV broadcasting. If you were hoping to get a job building, installing, or maintaining new services that used the freed-up spectrum, hope no longer. The spectrum, and therefore any new jobs created by its new licensees, won’t be opened up until mid-June. An auction for a tiny portion of the TV spectrum raised $20 billion (source). Let’s assume that auctioning the entire spectrum would generate at least $100 billion in revenue for the Federal Government. Let’s further assume that the spectrum license is one third of the total capital investment required for a new wireless service. That means that, had analog TV been shut down, companies would have invested a further $200 billion in planning, design, equipment, software, marketing, customer service facilities, etc. That’s real capital investment that should be creating long-term jobs. There probably isn’t that much actual investment in the latest $800+ billion “stimulus” bill going through Congress. We would need to expand the current bill to $1.6 trillion in order to undo the damage done by the delay in the shutdown of analog TV.

Let’s also look at the global warming and air pollution angle. A VHF station transmitter will typically deliver more than 300 KW of output power. Assuming some level of losses, we’re talking about 0.5 megawatt per TV station to run the analog transmitter. Multipled by the more than 1000 television stations nationwide, even accounting for the fact that many are lower power UHF stations, we would need a good-sized coal-fired power plant running 24 hours per day to supply this load. Given that we’re running these huge transmitters for no particular reason, we’re going to need even more elaborate schemes to combat CO2 pollution.

Finally let’s look at the electronics retailers, some of which are going through Chapter 7 liquidation proceedings right now, notably Circuit City. Under the original plan they would have had their best day ever on February 17 as the analog TVs stopped receiving signals through their rabbit ears. They would have sold converter boxes, new digital TVs, cable subscriptions, satellite subscriptions. What do they get now? The same temporary boost in sales, but three or four months after most have gone out of business.

Can anyone think of a more effective way to shrink the U.S. economy?

[To those who worry about the effects of a TV shutdown on the poor: more than 97 percent of U.S. households classified as being in poverty own a color television and this number was about the same even when TVs were much more expensive. Americans at all income levels seem to be able to obtain necessary television gear.]

Full post, including comments

What would you call people who pay $25,000 for a nine-week course of study with a collection of Silicon Valley optimists? “Singularly Stupid”? That may explain the name for the new Singularity University (SF Chronicle article).

The idea of the singularity is that technology, especially in the form of artifically intelligent robots, will solve all of our problems and technological advance will speed up exponentially starting roughly around the year 2030.

So far technology innovation hasn’t outstripped Malthusian human population growth. We can grow more food more efficiently, but the number of human mouths to feed has grown just about as fast, so that we struggle to feed everyone. A lot of what we’ve done over the past few hundred years has come at the cost of using up the Earth, e.g., clearing forest for farmland or digging up coal and oil and lighting it on fire, taking all of the Cod out of the North Atlantic. Far from freeing us from cleaning the house, Artificial Intelligence thus far has failed to live up to promises made by professors seeking research funding in 1960 (that reminds me I need to do laundry!).

Given the track record of tech as a mixed blessing and as a slower agent for change than predicted, do young people need to prepare for 2030? Can they prepare by listening to Ray Kurzweil, or anyone else born in 1948? Should they fork over $25,000 for nine weeks or simply watch old Jetsons episodes?

Maybe I will kick off the comments section with a realistic tech innovation that would change the world in a positive way. My pick: A better battery (cheaper, lighter, higher power density). That would enable the use of renewable energy in every kind of portable application, e.g., cars and airplanes, and also make it much more practical to use wind and solar generation.

[Special offer: If you come to Boston this summer and pay me $25,000, I will spend 9 weeks telling you all of the places that a better battery could be used, starting with my Super Walkman design that can play 2300 cassette tapes before requiring a recharge.]

Full post, including comments

I’ve drafted an article comparing Verizon FiOS with Comcast, mostly looking at Internet service. Please comment with typo corrections here and use my main server’s “add a comment” link at the bottom of the page to contribute alternative perspectives and things that you think will help other readers make the best use of FiOS or Comcast. Thanks in advance.

Full post, including comments

I had dinner this evening with two Harvard undergraduates. I asked how they felt about our government borrowing trillions of dollars that they would have to pay back. “It’s a good time to be in school,” they responded. What did they expect would bring the U.S. out of this depression? “When Americans realize how tough it is going to be, they’ll start working a lot harder. I spent last summer working in a lab in India. The Indians worked much harder than any American because it is much more competitive over there.”

This might be a more sensible explanation of how we might plausibly return to economic growth than anything that I’ve heard from politicians or economists.

[Alternatively, the kids could get jobs as public school teachers upon graduation. This New York Times article talks about the lives of some teachers in the Rochester, NY public schools, which has doubled its real-dollar payroll expenses over a couple of decades even as the student achievement has continued to slip. The Times story concentrates on the benefits enjoyed by the unionized public employees and doesn’t mention the fact that the schools are considered failures and that an employer would find an ample supply of better educated workers in most parts of Mexico, India, and China.]

Full post, including comments

Michael Phelps, the heroic swimmer from the 2008 Olympics, apparently has been smoking some marijuana lately. Let’s compare his situation to the average American’s. Phelps has a high-paying job doing something that he loves, millions of dollars in endorsement revenue just in the last few months (source), and his own charitable foundation. All of this has been achieved at the age of 23. Phelps has made money so fast he probably didn’t have time to invest it in the stock market, so we can be fairly certain he isn’t depressed about his personal finances.

Joe Typical American, by contrast, is about 35 years of age. His Olympic gold medal count is zero. He is probably overweight, if not obese. Unless he works for the government or in health care, he is unemployed or worried about being unemployed. He probably didn’t enjoy his job that much when he was employed. His retirement savings have been confiscated by the Wall Street bonuses of 1995-2007. His future earnings have been confiscated by the Wall Street bonuses of 2008-2015 (to be paid out of TARP and other taxpayer funds, which will inevitably result in debt).

If Phelps needs to smoke dope, what do the rest of us need to get through the next decade or two?

[On an unrelated note, a few readers pointed out that Phelps was convicted of drunken driving four years ago, shortly after winning 8 gold medals, and the world didn’t get its panties quite as twisted. Driving a monster SUV and running the risk of killing someone isn’t as exciting as inhaling some marijuana.]

Full post, including comments

… because Microsoft will do it for me.

http://philip.greenspun.com/business/mobile-phone-as-home-computer

was published in September 2005, describing a dock for a mobile phone that would serve as the user’s desktop computing environment.

A friend emailed today to point out that on December 7, 2007 (Pearl Harbor Day), Microsoft filed “SMART INTERFACE SYSTEM FOR MOBILE COMMUNICATIONS DEVICES” (serial number 952152).

Can someone explain to me how these are different? Is my product proposal prior art for this Microsoft patent application?

Full post, including comments

… because none of the cabinet secretaries pay federal taxes anyway… (see the latest from the New York Times on Tom Daschle’s unpaid $140,000)

Full post, including comments