John G. Fletcher RIP

I sadly note the recent death of my stepfather, John G. Fletcher. Like my father he had a doctorate in physics and had a long career at the Lawrence Livermore Laboratory. Unlike my father he quickly moved away from physics to work in the lab's computation department. Here is a 1983 video in which he talks about the lab's computer systems. And here are transcripts of interviews with him in 1993 and 1995 looking back at his career at the lab. Finally here is the wikipedia article on Fletcher's checksum which he invented.

He shared my interest in mathematical games and puzzles and continued to work on them after he retired in 1993. He submitted many solutions to IBM's monthly Ponder This puzzle from November 2001 through June 2011, before, during and after the period I was puzzlemaster.

I took the photo in 1991 at Big Sand Lake, near Park Rapids Minnesota, where his parents lived. I have fond memories of vacationing there in the early 1990s.

Moving on

As any long term readers of this blog know I was laid off from my job at IBM's Watson Research Lab in Westchester New York about 3 years ago. I was out of work for a while but then got a job in central New Jersey which is about 90 miles away from my Ossining townhouse. This is too far to conveniently commute but I was reluctant to move as the job was originally temporary and I liked my townhouse. So I stayed in a motel during the week and returned to Ossining on weekends. But this wasn't ideal and since my position was made permanent early last year it seemed best to relocate. After a somewhat prolonged search I bought a townhouse near Princeton late last year and after some more delays I had my stuff moved last week.

This took three men two days. About 10 hours the first day to pack it up and load the truck, about 5 hours the second day to unload the truck (leaving the unpacking to me). I found the process a bit tiring. Not in a physical sense as I wasn't doing any of the work but somewhat emotionally draining as it marked a rather definite end to a big portion of my life. At least I didn't have to worry about the cost as my employer paid. Apparently quite a bit, hopefully a reasonable amount trickled down to the men doing the work.

Despite my doubts about moving I am settling in fine in the new place.

Kodak bankrupt

Long ago I inherited small amounts of a few stocks from my great uncle. Over the years their fortunes have varied. The one that did worst was Eastman Kodak which last week filed for bankruptcy after years of decline. My investment wasn't actually a total loss (although I expect the shareholders will receive nothing in any reorganization) as Kodak had paid a dividend for many years and also spun off Eastman Chemical which while not doing so terrific itself isn't bankrupt. Still not one of my better investments.

While it is of course difficult when your main product dies it is a bit hard to believe that better management couldn't have salvaged something. It is a little hard for me to understand why the current CEO still his job.

Kodak should have had plenty of warning, while the early digital cameras were expensive and not very good their problems were clearly fixable while film was a mature technology without much upside potential. Which brings me to the subject of the Kindle Touch e-reader which I received for Christmas. I like it a lot despite an annoying user interface. The problems seem fixable and I expect over time that e-books will dominate the market.

One up on Wall Street

I recently read "One up on Wall Street" by Peter Lynch (with John Rothchild, 1989). Peter Lynch was a very successful manager for Fidelity's Magellan mutual fund which he directed from 1977 to 1990. According to wikipedia he achieved an average annual return of 29% which is of course terrific. One might expect he could write a worthwhile book on personal investing but in my view this book is not it.

The main problem with this book is its premise (see pages 240-241 in the paperback edition) that with moderate effort following his advice the average person can expect to beat the market averages by 2-5% a year (so if the market averages 10% average annual return you can expect to achieve 12-15%). In my opinion this is totally unrealistic and potentially dangerous advice for the average investor. More realistically an individual investor should expect returns (before expenses) which roughly track the market. Of course if you aren't widely diversified you won't match the market exactly and might in fact achieve excess returns of 2-5% annually for some time. But these will probably just reflect good luck and be no more likely than lagging the market by an equivalent amount.

So if you don't enjoy researching and picking stocks but want to be invested in the market the practical thing to is to invest in an index fund. This takes minimal effort and you should achieve the market return. Index funds are also relatively tax efficient. If you do enjoy picking stocks and have a bit of a taste for gambling then picking out and holding a portfolio of 20 or so stocks is fairly harmless as long as you don't trade too much and spread your picks around. Your expected return won't suffer, you can root for your stocks and you may get lucky and beat the market by a bit. And by managing your own portfolio you can adjust for your personal tax situation. Active mutual fund managers often only think about pretax returns and give up some after tax return by being too willing to take gains. Lynch in fact totally ignores taxes in this book.

A large part of Lynch's book consists of war stories, stock picks of his that did well or poorly. He has explanations and advice based on these examples but to me they just seem to illustrate the adage that it is easier to predict the past than the future. They don't appear to add up to a reliable method for beating the market going forward.

The book also come across as rather dated. References are made to looking up company information in local libraries which subscribe to the Value Line service but I believe this type of information is now widely available on the internet. And 1990 time frame stock picks are mostly of historical interest at this point.

So in summary I would skip this book.

Habit

When I joined IBM I opened an account at a bank with a branch in Millwood. It was convenient as I could stop by on my way to work. When I got laid off I went on banking there out of habit although it was now considerably less convenient. About a month ago it rather belatedly occurred to me that my bank has many branches. I have since been using one in Croton and it is rather startling how much easier this is. It makes me wonder how much other suboptimal behavior I engage in out of inertia.

A Song of Ice and Fire

I recently read the five existing books of George R. R. Martin's fantasy series, "A Song of Ice and Fire". It is perhaps better known by the name of the first book, "A Game of Thrones" (1996, 704 pages in hardcover). The following four books were "A Clash of Kings" (1998, 784 pages), "A Storm of Swords" (2000, 992 pages), "A Feast for Crows" (2005, 784 pages) and "A Dance with Dragons" (2011, 959 pages). As can be seen this is a very long series, over 4000 pages so far, and reading it represents quite an investment in time. Martin certainly has story telling talent and there are good things about the books. But overall I found the series seriously flawed and I can't really recommend it.

My first objection is to the overly ambitious scope of the story. It appears to me that someplace in the second book the author lost control and began multiplying characters and subplots beyond reason. I like a book or series to tie things together at the end and at this point this seems unlikely to happen.

A related issue is the way the series is written. There are numerous narrative threads and viewpoint characters and the books jump around among them. This is of course a common technique and is ok up to a point. But in my view this series has gone well beyond that point. Many of the narrative threads are largely independent and chopping them up and intermixing them just seems confusing and irritating to me. The problem is aggravated by the author's propensity for abandoning threads in cliffhanger situations rather than at natural stopping points which makes it harder for the reader to pick up the new thread.

My final objection is to the general tone and content of the series which is quite dark. The books assume a medieval level of technology (swords and crossbows) with some magical elements added. Considering what this period of human history was actually like it would have been unrealistic for the series not to have dark aspects. But the series emphasizes them in a way that I found off putting. It is not so much any one description of the strong mistreating the weak (often with sadistic cruelty) or of betrayal and treachery but the cumulative effect of many such descriptions over 4000 pages. When reading for enjoyment and entertainment I would prefer a bit less realism.

In my opinion the first book was the best. According to wikipedia the series was originally envisioned as a trilogy. I think Martin would have done better to stick with this plan. Authors sometimes have trouble coming up with satisfying endings to their stories. Greatly expanding the scope of the story postpones dealing with this problem but does not make it any easier.

So in summary, Martin is a talented writer and I certainly found parts of the books entertaining even compelling but ultimately I found the series disappointing. I am the sort of person with a compulsion to find out what happens next so I will probably read any additional books as they appear but if you aren't already invested in the series you might think twice about embarking on it.