Monday, August 29, 2005

Paul Graham makes republican mistake #1

The problem with wealthy entrepreneurs who got lucky is that they attribute their success and wealth to their skill, intelligence and risk taking ability, while a lot of their wealth actually comes from luck. But because they can't admit that (since that would be an admission that they do not deserve their wealth), they have to come up with justifications as to why the universe made them a multi-millionaire.

The problem with Paul's thesis is that I can think of many many ways you can reduce inequality without reducing risk-taking behavior. Here's an example: supposed you provide universal healthcare. A large proportion of U.S. bankruptcies are driven by medical bills due to lack of health insurance. Suddenly, anyone who wanted to start a business can do so without losing their healthcare. I'd argue that more people would start businesses and take more risks than a national policy where your healthcare is tied to your job, effectively locking you to that job if you don't have a lot of money and have one or more pre-existing conditions.

There's an obvious place to tax people if you truly believe Paul's thesis (which I don't buy into --- I've seen too many folks luck into wealth to think that wealth is in any form "deserved" by most people who make that kind of money): inter-generational transfers. In other words, we should tax inheritances heavily and severely. But of course, the Republicans we see in congress are the ones most enthusiastic about eliminating the estate tax.

Sunday, August 28, 2005

Flight Vol 2

Not nearly as good as the first one, mostly because the large number of artists/writers contributing such disjoint visions brings into focus how the graphic strip art form needs many more pages to provide character and narrative. Nevertheless, there are a few good pieces: The Robot and the Sparrow, Jake Parker, Destiny Xpress Jen Wang, The Orange Grove, Kazu Kibuishi, Dust on the Shelves, Banniester, The Flying Bride, Giuseppe Ferrario.

Once again, I don't feel like I want to own this book, but I'm glad I did see read it once. If the really good artists here start producing their own books, it might be worth following up on them.

Saturday, August 27, 2005

A tough ride

Went on a Western Wheelers ride today. Everyone was scary fast! Sure, it's the end of the summer, so folks are as fit as they're going to be, but I'd just done a Tour of the Alps not that long ago, and it's scary how much fitter everyone else was. Of course, I could use the excuse that I planned my peak for that tour, but seriously, I'm a recreational rider. I don't peak.

I did meet someone I rode with 13 years ago, when I first started riding with the Western Wheelers, but before I dropped out. It was funny when I first said, "oh we rode together 10 years ago", she interrupted and said, "No, it's longer than that. It was 6 boyfriends ago." Thinking back upon it, I'd had 6 jobs (well, 2 at the same company) since then as well, so it's nice to know that some people switch boyfriends as often as I switch jobs.

Anyway, I'm now incredibly exhausted, despite having only ridden 72 miles with 7500' of climb. As they say, it's not the distance or the climb, it's the pace --- we did it all in 6 hours, and that's including stops!

Friday, August 26, 2005

Dealers of Lightning, Michael Hiltzik

Xerox's Palo Alto Research Center (PARC) invented the concepts that went into the modern computer that we use today. Mice, icon based bitmapped displays, WYSIWIG word processors, ethernet, file servers, object-oriented programming, and the laser printer. Yet Xerox did not make money from any of these inventions except the laser printer, and even then fumbled it by introducing it too late. This book is an indictment of Xerox's corporate culture, its short-sighted executives, and is the perfect illustration of my thesis that a technology company (or any company affected by technology) that is not mature must be run by engineers for it to be successful beyond the short term. Of course, that's not the only factor. Everything else also has to be done right (marketing, sales, etc), but if the key executives do not have the vision to pursue technological break-throughs and turn them into product, you might as well not bother funding an R&D lab.

Xerox's top executives were for the most part salesmen of copy machines. From these leased behemoths the revenue stream was as tangible as the "click" of the meters counting of copies, for which the customer paid Xerox so many cents per page (and from which Xerox paid its salespersons their commissions). Noticing their eyes narrow, Ellenby could almost hear them thinking: "If there is no paper to be copies, where's the `click'?" In other words: "How will I get paid?"

Wednesday, August 24, 2005

Google driving up the cost of Silicon Valley talent?

Google, Mr. Hoffman said, has caused "across the board a 25 to 50 percent salary inflation for engineers in Silicon Valley" - or at least those in a position to weigh competing offers. A sought-after computer programmer can now expect to make more than $150,000 a year.

I think most people have it wrong: software engineering talent has been too cheap for too long. Today, to entice someone to study software engineering, you would have to offer sufficient incentive to:

  1. overcome the fear that 4-6 years studying computer science would be rewarded by all the good jobs going to India
  2. medical school and law school all have potentially higher rewards, and are also more highly regarded professions (with also a more balanced male/female ratio)
  3. the fact that winner-takes-all is even more prevalent in computer science than anywhere else
  4. the average career of a Silicon Valley engineer is around 7 years, shorter than that of many pro atheletes!

Given all those disadvantages, the question should be why there are smart, talented, hardworking folks in software engineering at all. Well, the answer is typically that there is a chance for a big pay-off, if you work for the right company! The rising incomes of top Silicon Valley engineers needs to reflect that when all is said and done, Google is the only software company I've worked for in the last 10 years that can be said to have been almost completely engineering driven. That means that a similar offer from Google would be preferred by the discerning, intelligent software engineer than an equivalent offer from anywhere else. The following quote from the same article highlights the similarities to Microsoft in the 1990s:

Bill Gates certainly sees similarities between Google and his own company. This spring, in an interview with Fortune, Mr. Gates, Microsoft's chairman, said that Google was "more like us than anyone else we have ever competed with."

I will note that Microsoft was also in many ways, managed by smart engineers more than its competitors (like Borland, Lotus, IBM, Apple, or Netscape). They had more technical people in the upper ranks than their competitors, and perhaps that is why they had an edge --- the deep understanding of technology is important when technology is the battleground. Companies that pick non-technical CEOs too early in their lifecycle risk stunting their future growth.
Thoughts about the Canon 5D

With a full frame sensor and nearly 13 megapixel, this is the first digital SLR that I will consider dumping my film cameras for. Previous SLRs were either too expensive, have the nasty tiny sensor, not compatible with my EOS lenses, or were too heavy. (Or a combination of the above) At $3300, it's still expensive, but Moore's law will get it to about $2000 in 18 months, so that's probably the point at which the case for switching over to digital entirely becomes compelling.

The 24-105/4L, however, is almost certainly a must have. When the price drops a bit next year, I will definitely buy one and ditch my long loved 24-85.

Sunday, August 21, 2005

Lacing a new rear wheel Posted by Picasa

Thomas Marsten Posted by Picasa

Cyclists enjoy the sunrise from the Jeanne Meadows memorial bench Posted by Picasa

Bob & Betty Posted by Picasa

Self-portrait: Piaw, Lisa, and the Tandem (aka Yellow Totoro) Posted by Picasa

Piaw & Radek enjoying the Sunrise Posted by Picasa

The San Antonio Valley shrouded with Fog Posted by Picasa

Piaw & Lisa on Mt. Hamilton during Civil Twilight Posted by Picasa

Saturday, August 20, 2005

Flight Vol 1

An anthology of short comic strips. The problem with a short comic strip is that it's very difficult to develop all but the simplest characters in a few pages, even with the help of illustrations. While a lot of the strips in this book are very good, many are just mediocre, or unsatisfying. I like it, but not enough to buy it.

I never paid attention to people like you, either, although I knew who a lot of you were. Your names were all over the halls...
A Fire Upon The Deep, Vernor Vinge

I first read this book about 10 or so years ago when it won the Hugo award (and was unjustly denied a Nebula award). It is still just as good a read today as it was 10 years ago. The space opera is fun, the characters are fun, and the universe as postulated entertaining, even though it's every bit as far fetched as the E. E. Doc Smith universes. (At least the writing is much better)

Somewhere barriers slipped aside, the final failing of Old One's control, or a final gift. It did not matter which now, for whatever the ghost said, the truth was obvious to Pham Nuwen and he would not be denied:

Canberrra, Cindi, the centuries avoyaging with Qeng Ho, the final flight of the
Wild Goose. It was all real.

Thursday, August 18, 2005

Planning the next trip

The next trip is likely to be an inn-to-inn walking trip in England's lake district. As usual, I'm starting out by planning this as a nice long walk, without any such distractions as museums or other cultural artifacts and let the folks who are coming along force me into visiting Wordsworth's birthplace or some such.

Suggestions, etc are welcome. And due to Lisa's school, the timing is forced: late May/early June. That makes planning and buying plane tickets easy.

Monday, August 15, 2005

Sunday, August 14, 2005

Climate Crash: Abrupt Climate Change And What It Means For Our Future

A few years ago, I read an article about climate change: the concern was that climate change might not be a continuous process as most folks might believe (naively), but that it could flip-flop between cold and warm states for unknown, chaotically driven reasons. I was extremely skeptical that such theories would in fact pan out, given the recent history of climate change, but this book (which is more about the theory behind abrupt climate change and less about the implications) shows how the theory has become more than just speculation and is now the primary theory behind the history of Earth's climates.

Of course, my pessimism about the human race says that we won't do anything about it until it's too late.

Down in the core below 2750 meters, in ice that the Europeans were confident represented the period of Eemian warmth about 120,000 years ago, oxygen isotope data showed two especially large and sudden plunges towards ice age cold. In one episode, average temperatures apparently plunged 25 degrees F for about 70 years. The only period of relative stability during the Eemian came during the last 2,000 years of its warmest stage.

"The unexpected finding that the remainder of the Eemian period was interrupted by a series of oscillations, apparently reflecting reversals to a `mid-glacial' climate, is extremely difficult to explain," the Euroepans wrote. "Perhaps the most pressing question is why similar oscillations do not persist today, as the Eemian period is often considered an analogue for a world slightly warmer than today's." Given the history of the last 150,000 years, they wrote, the past 8000 years "has been strangely stable."

At the granite portions of the ten lakes hike Posted by Picasa

At the top of the climb! Posted by Picasa

Stream feeding into the main lake Posted by Picasa

The second highest of the ten lakes Posted by Picasa

Piaw & Lisa waiting for the evening alpenglow Posted by Picasa

Evening reflection on the first of the ten lakes Posted by Picasa

Ridges from the climb up from ten lakes Posted by Picasa

Lisa enjoys fields of flowers on the ten lakes hike Posted by Picasa

Tuesday, August 09, 2005

Fooled by Randomness, by Nassim Nicholas Taleb

This is probably the best financial book I've read all year. (Which doesn't say much, I admit) It's not for anyone who hasn't read any of the classics (like Malkiel's Random Walk Down Wall Street), but it's very much worth reading since even for those of us who have a decent mathematical background, interpreting the results of the mathematics in real world terms is a rare and useful skill, and Taleb has a good knack for explaining how to do it.

I have no large desires to sacrifice much of my personal habits, intellectual pleasures, and personal standards in order to become a billionaire like Warren Buffett, and I certainly do not see the point of becoming one if I were to adopt Spartan (even miserly) habits and live in my starter house... Becoming rich is not directly a moral achievement, but that is now where the severe flaw in the book lies.
Note: the book in question is The Millionaire Next Door.
Why I am a die-hard Canon loyalist

After Lisa broke the camera on a sailing trip, I thought that was the end of it. My brother gave me the customer service # for Canon, and I called them. They gave me an address to send the camera to, I printed out the receipt (from, and sent the camera, box and all to them, expecting a hefty repair bill at the end.

I got the camera back yesterday. The lens unit had been replaced, the camera had been cleaned (though the scratches from my tour of the alps are still in evident, so it was clearly not a new camera), and the camera had been restored to working condition.

The charge: $0. They considered this a warranty repair.

Wednesday, August 03, 2005

A fabulous Joss Wheldon interview

You own a “Speed” poster on which your writing credit remains.

I do.

Was it a misprint? Was a teaser poster issued before the Writers Guild arbitrated that credit away?
It was “the” poster. And they put it out and then the arbitration happened kind of late. And so they pulled it and changed it.

So there are maybe a lot of those floating around out there somewhere?
I don’t know if they were actually up or if this was just the final mock-up. I just know that I have a copy of it. The arbitration was a great sticking point with me. I’ve always just disagreed with the WGA’s policy that says you can write every line of dialogue for a movie – and they literally say this – and not deserve credit on it. Because I think that makes no sense of any kind. Writers get very protective of themselves. They’re worried that some producer will want to add a line so he can put his name on it. But what they can do is throw writers at it forever without putting their names on it because of this rule. So I actually don’t think it works for writers. It certainly didn’t work for me.

Graham Yost [who received the sole screenplay credit for “Speed”] has always been very polite to me and very sweet but he did say to me, “You would have done the same thing.” And all I could say to him at the time was, “Well, I guess we don’t know if that’s true.” Because I’d never been in his situation. Then more than a year later John Lasseter called me and said, “I want to give all the animators who worked on the story credit on ‘Toy Story.’” And I said, “Sure.” And there are entire episodes of “Buffy” that I have written every word of that my name is not on. Which is gratifying to me because it means I finally have an answer to that. Which is, “No, I wouldn’t.”

Tuesday, August 02, 2005

S&S couplers considered useless

My issue with the couplers are that they're expensive and take forever to use. My first tandem had couplers and disassembly took 2 hours (with two of us on it), and assembly took an hour. And this was after we'd gotten good at it! And of course, take pictures when you first get the bike or you'll never be able to pack it away in the boxes ever again!

Anyway, after I'd had enough of this, we finally sold the bike, and bought an uncoupled tandem. We took it to Europe, and this time, wrapped pipe insulation around the tubes and took it with us on the plane unboxed. They bent the deraileur hanger, which cost me 5 Euros and 15 minutes to fix at a local bike shop. On the way back I removed the deraileur and had no damage to the bike. Packing time: 5 minutes. Unpacking time: 5 minutes. (Removing and putting pedals back on) All through Europe the trains took our tandem and us, no problems. All the
hotels also found a place to keep our bike.

When we flew to Colorado last year for a tour, we had to box up the bike. On the way there, we got a Santana cardboard box (cost: $45), and took it with us on the plane ($80). Packing time: 20 minutes. On the way back, we rode to the Denver International Airport, bought 2
United Airlines boxes ($20), packed the bike there and then (20 minutes), and brought it back to San Jose International. The airline scratched the front saddle and gave us a $20 United airlines coupon for a minor scratch, we reassembled the bike (10 minutes), and rode
home. In Colorado we just kept our bikes in the hotel room with us.

Both my experiences travelling with the uncoupled tandem were superior to travelling with the coupled tandem in a hard case. We spent less time packing, and more time enjoying the biking. (but note: we don't have a carbon fiber bike --- those might be more fragile, in which case you might have no choice but to go S&S)

Finally, this year, I went to Europe again on my single. One of my friends brought a single bike with ritchey breakaway coupler. It took him more than an hour to assemble the bike! And more than 1.5 hours to take it apart at the airport. For myself, I would not pay the extra cost in either time or money, just to save the 15 Swiss Francs (30 for a tandem) it costs to take a bike on the train.

One of the couples in the club I'm in has both an S&S Santana and an uncoupled Calfee. They're signed up for 2 Erickson tours this September, and guess which bike they're bringing? The uncoupled Calfee. Even though they're retired and have lots of time, it's still less fun assembling and disassembling bikes than riding them, and they'd rather risk damaging their $10,000 bike.

I ran into a cyclist last year who said he surveyed every tandem couple he met while riding because he was shopping for a tandem. He said that whenever he asked about couplers for couples that had them, the usual response was "We thought they'd be a good idea, but we've
never used them." For those couples, maybe the cost of the couplers meant they didn't have any left over to take a vacation, but still.

There are a few good applications for couplers:

  1. Private plane owners. You're not going to fit a full sized tandem on a propellor plane, no matter what. (When travelling, every time I could have taken a prop plane I could also have taken a ferry, so it's not an issue that comes up unless you're a private pilot)
  2. Cruises. This was what S&S was designed for --- so you can bring your bike into your stateroom disassembled, get off the cruise boat, and assemble your bike and ride around town. A folding bike is probably cheaper for this task than adding S&S couplers to your Calfee.
  3. Frequent domestic flights. At $80 a pop, if you take 13 domestic flights, you've made back the cost of the couplers. ($2000 after you throw in the suit cases) Of course, after you've done the assembly once or twice, you might discover (as some folks I know did), that you'd rather *drive* to your domestic destination than go through the assembly/disassembly process again.
  4. You can't bear the thought of a $10,000 bike being protected by a $20 cardboard box, even though in my experience the $20 cardboard box in 10 years of travelling has never failed me.
I suspect that in practice, reason #4 is why folks buy S&S couplers.

[Update: there have been recent reports that even coupled S&S tandems are no longer escaping airline domestic charges. In addition, real world experiences have vindicated this article over and over again, as described in a later blog entry]