SCIENCE OR COMPUTERS?
Some career advice from Jon Claerbout

Advice below was written well before the 1999 dot com bust.

It is also said ... `Go not to the Elves for counsel, for they will say both no and yes.'     `Is it indeed?' ...
`Elves seldom give unguarded advice, for advice is a dangerous gift, even from the wise to the wise, and all courses may run ill.' -- Tolkien

Fifteen years ago, after I had a beer or two, a few students might have gotten this advice:

Try a job at the U.S.Geological Survey. There you never need to worry about competition. Academia has a reputation of being an ivory tower, but some of the most academic people I know are settled more comfortably in large companies. At the opposite extreme are geophysical contracting companies. Your competitors are always trying to undercut you, and they will often succeed.

Hindsight taught us that employees of government and giant companies have suffered massive dislocations and layoffs too. Life in a survey contracting company though, has proven to be the roller coaster that we always thought it was.

Should I make my career in computers?

Reflection seismology is strongly infused with computation. That suggests that maybe you would rather have an exciting career in computers. You might choose a career in computers because

There are many good opportunities because the computer world is always changing, and that puts young people on an even footing with older people.

During my life, computers were always undergoing revolutionary change. When I was a sophomore I studied vacuum tubes. As a senior, I studied transistors, and wondered if I would ever understand them as well as I understood vacuum tubes. Before long, transistors were superceded by integrated circuits. When I arrived at Stanford I realized that much of Stanford's EE faculty was out of date because their field had changed so fast. Students in EE were well advised to get an MS and then leave, because industry moved so rapidly ahead of the PhD programs, many of which were hidebound with their professors. Then computer science split from EE. I was astonished to become the owner of a Fortran compiler (previously the University had only one) and to acquire the obligation of maintaining an operating system. Then followed a long succession of languages.

I was at the cutting edge of interactive graphics programming with Sunview. In a few short years, Sunview was dead; everyone needed to learn Xwindow. Now we should learn TCL/TK or Open GL or Java. I learned text processing in troff. Then everyone at universities switched to tex, (I am typing this in raw html) and soon it will be something new, probably MathML. (To say nothing of people urging us to learn MSWord, Frame, ....) To maintain my sanity with reproducibility of computational research results I first learned make, then I had to convert to cake, then gmake. [Five years later I hear we'll be switching to Phython and SCONES.] One language that I learned for administrative tasks is AWK. It has served me well, but to any younger person, I would say, ``Learn PERL instead.'' [Five years later, now it is Python instead.] I have become reasonably expert in C and Fortran77 but wish I knew better Fortran90, and sometimes I dream about C++ and Java. Besides all these languages, some colleagues urge us to work instead in Mathematica or Matlab, and they may turn out to be correct. Such a buzz of languages! Would you like to learn to speak Chinese and Russian too? It is already too late. You should have begun in your childhood.

So to anyone who thinks that computers are a good career choice because young people are on an equal footing with older people, I need only to ask,
Are you planning to stay young forever?

Math, Engineering, and Geophysics have their eternal verities: Fourier analysis, Maxwell equations, elasticity, finite differences, operators, eigenvectors, adjoints, conjugate-gradient solvers, expectation and covariance, moveout corrections, acoustic imaging, prediction error, spectrum, the list goes on and on. Learn these things and learn them well, because they can serve you for a lifetime. Of course you'll need to know computers too, in order to put these fundamental principles to use. IMHO fundamental principles (not the latest computer fad) should make up a full half of your education. Then, as Andrew Long says, "As long as you are enthusiastic about trying new ideas, you'll always have opportunities."

Computers are a good thing, but my advice is to build your career also for its middle and later years by building on a more permanent structure of science, engineering, or other skills.

Now let us take a serious look at the kind of career that results from doing your PhD with the Stanford Exploration Project. I will not answer this theoretically, but with a summary of the careers of my graduates.

48 careers in reflection seismology

At the 25th anniversary of the Stanford Exploration Project in 1998, we had an opportunity to take stock, to accumulate the results of 25 years of experience of 50 people. Of 48 PhDs at SEP, none were dead, 36 graduates showed up, and we were able to communicate with all the others. I think it is fair to say that all continue to benefit directly from their SEP education including those few who drifted off into computers or academia. Nobody was unemployed at the time of the reunion, but many had been forced to change company, some several times.

Below is Rick Ottolini's analysis based on his alumni data base, and on the SEP thesis data base .

Type of organization

16 oil company
15 exploration contractor
7 university
5 startup company
2 consultant
2 computer company
1 government agency

Where they live now

13 California
11 Texas
8 Europe
5 Colorado
5 Other North America
3 Latin America
3 Mideast/Asia/Australia

Thesis topic breakdown (multiple counts)

21 Migration/Modeling
10 Velocity Estimation
8 Signal Processing
7 Dip Moveout
4 Estimation/Inversion
4 Slant stacks
4 Tomography
3 Elastic waves
3 Geologic targets
2 Near surface
2 Anisotropy
2 Sources
1 Attenuation

Polish translation of this page by Abdul Sattar
Czech translation of this page by Valeria Aleksandrova