Thursday, August 28, 2003

I've had a run-in with poison ivy. I loathe poison ivy. Fortunately, most of the affected area seems to be on my shins. There is a small patch beside my hip, but it hardly itches at all. There is a very small patch at the base of my neck, and it itches like crazy. I bought a topical treatment which reduces the itch and will supposedly help dry the rash. Given my experiences with serious bouts of poison ivy rash as a youngster, I consider myself lucky that it's not worse. Still -- I loathe poison ivy.

The first week of classes is mostly done, and I've enjoyed what I've seen. I've resolved to spend less time fiddling with the computer and more time on technical reading this semester. I've succeeded for the first week, at least. I hope I can continue to maintain this much reading time in my schedule; I've digested one paper on wave localization which has a significant impact on some of the applications I consider, and I'm struggling with some textbook material involving thermoelastic damping and general coupled effects between the electrical, thermal, and mechanical domains in solids.

My rear bike wheel has a broken spoke, but I think I'll wait until Tuesday to have it repaired.

Monday, August 25, 2003

Did someone say there's some rotini in the state of Denmark? What are we waiting for, let's eat!

The first day lecture of the semester went well. It looks like the first two weeks of the fluid mechanics course will be an introduction to the engineering flavor of tensor calculus. Ask a mathematician what a tensor is, and he'll start telling you about forms and multilinear algebra. Ask an engineer what a tensor is, and he'll hand you a symbol decorated with a pile of eyesight-ruining subscripts and superscripts. You need the superscripts and subscripts in order to actually perform computations, and sometimes they are even handy for doing some symbolic calculations. But I much prefer coordinate-free notation, if only for the way it preserves my eyesight for another day. Naetheless, this is something I can do.

I spent part of the evening looking closely at the thermoelasticity equations I copied from Sanjay on Friday. I'm not sure whether he made a minor error, or I copied something incorrectly, or I just am suffering a misunderstanding. Anyhow, right now I have two formulas for the stress power term in the thermal energy balance equations, and those two formulas cannot be consistent with each other. One or the other is wrong; now I just need to figure out if the error is in my calculation or if it's in the formula from the notes. Ah, the joy of continuum mechanics.

I've learned a few things in graduate school, but I think one of the deeper lessons was how important physics is. I so easily miscopy terms in equations and flip signs in my codes; without a picture of the physics of the problem, those errors are difficult (sometimes nigh unto impossible) to find and fix. Of course, I also know people who have an excellent intuition for the physics of their area, but who will bend over backward to avoid the mathematical descriptions that make those physics precise. If physical problems are stories, then the physics is the plot line and the mathematics is the language. It's possible to skimp one or the other -- people still make silent films, and there exist rambling stories full of fine language and no particular plot. I suppose it's possible to skimp both plot and language, actually, as evinced by some summer blockbusters driven by explosions and sex appeal and precious little else. But it's very satisfying to read a story with both plot and well thought-out sentences, and I find it very satisfying to attack mathematical problems that come from physics I understand.

    Currently drinking: Peppermint tea

Sunday, August 24, 2003

I frittered the weekend away. I spent Friday night and Saturday morning reading fiction. Saturday afternoon, I walked along Solano; Saturday evening, I watched a bad martial arts movie, which was dubbed into Spanish for extra incomprehensibility. The dialogue and plot weren't too complicated, so I was still able to follow most of it with the aid of my high school Spanish, hazy though the memories are. Today I cleaned and reorganized my room. I suppose that means that I was trying to avoid work; I always seem to end up cleaning when I feel like avoiding some other task. The shelves in my room are packed tightly, now, but there is working space on my desk once again, and I've stowed the little piles of books that were starting to gather around the corners of the room like seaweed around a pier.

Tomorrow is the first day of instruction for the fall semester. I have a class at 9:00, which is simultaneously a blessing and a curse. Since the course meets so early in the day, it will not interrupt me when I'm in the middle of something. On the other hand, if I want to get to campus at 9:00, I'll face less pleasant traffic than I usually encounter when I travel later in the day. There are other advantages to working from home in the mornings, too. I have a window at home with a pleasant view, and I can open the shades to let in the sunlight. When I take a break, I can go to the kitchen and prepare a hot lunch or a cup of coffee. The window in the office is always shuttered to prevent monitor glare; even were it open, we'd just look at the side of Etcheverry hall. Of course, I do enjoy visiting the cafes around campus.

In the interest of finishing research work, I took no classes last year. It was a reasonable choice, but I look forward to my classes this semester. I've wanted these courses for a long time, and as a graduate student I still have the scheduling flexibility to take them. Jim has a sabbatical this semester; I wonder if he'll audit any courses? I know he sometimes misses taking classes. Perhaps he'll choose instead to spend a few hours a week in self-study of algebraic geometry or some other topic that he's mentioned he'd like to pursue further.

The semester is starting, the semester is starting! I've failed to finish all the things I wanted to finish this summer. Now that the summer has passed, though, there's no sense in worrying further about what I might be able to fit in before the semester starts. The semester is about to start, and I'm happy about it.

When I came home Friday night, I thought to myself, I'll take this evening off, then work for the remainder of the weekend. I'm excited! I'm motivated! And I was wrong. I spent Friday evening reading and Mars-watching. I spent Saturday morning reading and listening to the radio. I wandered to the top of Solano on Saturday afternoon, and watched a bad Steven Segall movie with my neighbors on Saturday night. The movie was dubbed into Spanish, so following the plot was a little trickier than it might otherwise have been. Fortunately, the plot was not particularly deep and the dialogue was not all that complicated.

I spent most of today's motivation on clearning and reorganizing in my room. With some judicious packing, I cleared the piles of books that were starting to take over every elevated flat surface in my room. My shelves are full, but at least my desk has free space for writing once more.

My first lecture of the semester is tomorrow morning at 9:00. I'm looking forward to the lecture, but I'm not looking forward to going to campus at 9:00. There are advantages to working from home in the morning: I have a larger desk here, I have a kitchen where I can prepare hot food or coffee, and I have a window that lets in sunlight and lets me look out at the world. The window shades at the office are always drawn to prevent glare on the monitors; anyhow, the only thing we can see through the window is the side of Etcheverry hall. It's also sometimes easier to work undistracted when I'm at home. Or perhaps it's just a different set of distractions, sometimes easier to resist than the distractions at the office, sometimes not.

Tuesday, August 19, 2003

I've decided to take the PDE sequence this year. I've intended to take Math 222A-B every year since I came to Berkeley, but I always ended up deciding another course or activity was more important. But now I'm registered for the class, and I have the book. I'm looking forward to my classes this semester, both in partial differential equations and in fluid mechanics. I'm past my required coursework, but I want to learn the material. Besides, there is something relaxing about mathematical classwork. Perhaps it's that homework problems tend to have answers that are shorter, neater, and easier than the answers to problems that arise in my research. Or perhaps it's just the change of pace.

The semester has officially commenced, though instruction doesn't begin until next week. The undergraduate population has returned to campus, along with the fraction of the graduate student population that spent the summer at an internship and those just arriving. It's quite a change from the summer campus: now, the majority of the people I see when wandering out of the office are not staff, graduate students, faculty, deranged, or some combination thereof. Of course, I only see the difference when I wander out of the office and pay attention to the world around me.

The start of the semester also means the return of free food. I ate a dinner of leftover noodles, rice, and tofu from some activity or the other. Now I have the energy to sit in my office and work on this problem again for hours!

Time to head home, I think.

Sunday, August 17, 2003

I spent my days this week reading, writing, and thinking about solution methods for structured eigenvalue problems. I spent my evenings reading about political and legal matters, particularly those relating to copyright. I finished reading Lessig's The Future of Ideas, which is an extended essay on and recent history of copyrights, intellectual property, and the tension between open use of resources and governmental or commercial control. I've also been reading about the SCO vs. Linux circus (see summaries at Computerworld or ZDNet). Briefly, SCO claims the copyright to code released by IBM into recent versions of the Linux kernel. They've initiated law suits against IBM and RedHat, and are making threatening noises toward Linux users in general. Whether or not they legally hold rights to pieces of the code -- and the history of Unix and its relatives is sufficiently tortuous that I'm unsure -- SCO's general strategy seems indefensible. But I'm not a lawyer, and I suppose SCO's antics are no sillier than the dynamics of the California recall election.

I strive to write directly and clearly when I describe my work. Writing well is difficult, but it's ultimately rewarding: not only do I gain insight by writing and rewriting my abstracts, talks, and papers, but I have the opportunity to influence someone else. I want the ideas and codes I develop to be understood and used. That's a powerful motivating force for many: scientists and engineers, programmers and philosophers, and doubtless many writers and artists. I read about strategies in which businessmen and lawyers try to exploit or defend exclusive use of intellectual property which they do not understand or want to understand, and which the developers would probably prefer to share, and I feel frustrated.

And after I spend some time feeling frustrated, I eat and sleep and wake to write some more.

Monday, August 11, 2003

My CD burner arrived today. After I upgraded cdrecord and xcdroast, it worked fine. I've backed up my home directory now. And I got my code working today. As far as bugs go, this one wasn't too bad: it was sufficiently tricky that I didn't feel stupid for missing it the first time, and it was sufficiently untricky that I was able to diagnose and fix it in relatively short order. The office was full of people I have not seen for a while. The weather is pleasant, and I had a good bike ride.

But mostly my drive is backed up and my code is working.

    Currently drinking: Vanilla hazelnut flavored black tea
We are what we pretend to be, so we must be careful what we pretend to be.
-- Kurt Vonnegut

I remembered that quote from the beginning of my high school psychology course in eleventh grade. But I forgot that it was Vonnegut; I thought it was William James. A Google search remedied the error.

So if I pretend to be done with my code and my paper, and to feel bright and chipper in the face of another semester, will it take?

Saturday, August 09, 2003

Last night, I went with Mike and Tracy and Mike's father to see a jazz show at Yoshi's in Oakland. It was Pharoah Sanders and his group, and the show was excellent. There was a drummer, bass fiddle player, pianist, vocalist, and saxophonist. I was particularly impressed by the piano and bass fiddle.

It was an evening well spent, particularly since the alternative would probably have been to spend more time debugging. I thought about spending some time debugging today, but I opted to use my time and energy writing instead. The mathematics is interesting, and it seems as though broader implications become clear daily. Being able to say to someone I can solve that problem using an order of magnitude less time and space than you used before is very satisfying. It's useful to read and write and remember how curious and interesting these structures are when I feel utterly frustrated trying to diagnose problems with the code that works on them.

One day, I'll learn to present my work in a way that sounds interesting and useful to anyone, regardless of technical background or lack thereof. I think one key is to describe the physical things described by my equations, and say what I'm trying to predict in an informal way. Another key is to simply change the subject when the asker could really care less. I'm getting better at that. Hearing from friends that my work sounds incomprehensible and boring doesn't decrease my interest, but over time I think it has worn away some of my tendency to launch into minor lectures about whatever I've been thinking about at the slightest cue. I can always deal with the frustration by delivering longer lectures to those who are actually interested. I feel like people who ask me questions are doing a favor to me more than vice-versa. They give me ideas! And they let me tell them about interesting things!

  • Currently drinking: Water

Wednesday, August 06, 2003

Testing numerical software is hard. It takes a certain philosophical -- or perhaps vengeful -- frame of mind to find problems that will cause a mostly-correct code to behave badly. Not only is it difficult to find truly pathological problems, it's hard to write the code to verify the computed answer. And the immediate reward? Glaring evidence of one's own fallibility, along with more work. Testing and debugging are fun, but they're more fun when taken in small doses.

Taking a break from debugging my test code, I looked at the SF Chronicle web site and found out that Arnold Schwarzenegger will run in the recall election. I know there are other people out there thinking of the antics of Jesse Ventura in Minnesota. I suppose there are also people thinking of Ronald Reagan, who was more glib and less associated with smashing chairs into the heads of wrestlers and movie cyborgs. The recall election is foolishness. It's taking time, money, and political resources to replace a governor in a bind with another governor who will be in the same bind. California finally passed a budget, one based on heavy loans. Davis protested that he would have preferred an increase in the sales tax, but if he had refused to sign, it would have meant minimum wage for state employees, starting at the end of August and continuing until a budget was in place.

However, the whole mess does provide entertainment.

  • Currently drinking: Miso soup

Tuesday, August 05, 2003

Garlic is wonderful stuff. But when it goes bad, it's fearsome. It's like a computer, that way.

Sunday, August 03, 2003

David's wedding was Friday. We drove to Tahoe, attended the ceremony and reception, and drove back. The ride was a little under four hours in one direction, so we were in the car a long time. Both the wedding and the ride were fun. I see Rich and Jason often enough; we share an office, after all. And Jimmy drops by frequently, too. But conversations on a long car trip evolve differently than break-time conversations at work.

I finished Dawn to Decadence on Saturday. Barzun writes vigorous, engaging prose, and I enjoyed the book thoroughly. I also enjoyed closing the cover and thinking Done! The book is about eight hundred pages, and I did not read those pages quickly.

I did no school work this weekend. Instead, I read and played, ate lunch with Elaine, visited Barnes and Noble, and did small chores around the apartment. I also bought an external USB CDR drive to use with my laptop. I've been lax about making backup copies of data which I could, in principle, recover from another source. But if I lost my laptop drive now, I would lose the organization of much of my data; and that would be painful.

  • Currently drinking: Vanilla-hazelnut tea

Friday, August 01, 2003

The follower is underappreciated.

Leaders -- those who administer, who influence, who lead by example, or who explore new areas -- are important. But the word leader is multi-faceted. The sense in which a competent administrator leads need not be the same as the sense in which a scientific explorer leads. Members of the cult of leadership, though, hold that leadership is not a collection of activities, loosely related, which share a common term only by the vagaries of language evolution. Nay, leadership is a virtue! These are the people who write high school graduation speeches in which they address tomorrow's leaders, or who insist that we should stand behind our leaders solely because they lead.

But what of the follower? From the rhetoric of the cult of the leader, a follower might be a sheep-like animal, though without wool and perhaps with a milder odor. But the researcher who studies the literature is a follower. So is the student engaged by a lecture, and the scientist who verifies an experimental result. The engineers who detail the construction of a structure may follow the plans of an architect. Judges follow precedent. Done well, following is not passive and unquestioning. It is active and critical, the pursuit of the man still humble enough to believe his fellows and forebearers may have some ideas of merit.

Whether or not anyone chooses to follow my ideas, whether or not I'm ever stuffed into an administrative role, may I never lose the perspective to remain a follower.