The very first text editor 3690

Elliott Roper

--snip

i think there is some of this in all the great high energy physics labs, and in several other physics departments

in the early 80's i was fortunate enough to work at Fermilab for a while. our little party of three arrived in a rental truck shortly before the winter solstice, with a load of 8x3 ft aluminum drift chambers that we had designed and built.

it was a dark and stormy night, the rain just turning to ice, in that exquisitely midwestern way, as we began unloading. we discovered that we had one pair of gloves between us, whereupon the tenured professor in our group disappeared.

he mysteriously reappeared, as tenured professors do, just as we finished.

shortly after that discouraging introduction, i received a five thousand volt electric shock while replacing a defective drift chamber, causing me to drop the 300 pound piece on said tenured professor. a proposal was was immediately tabled and unanimously carried; we repaired to yon friendly pub by the entrance, where we remained until ejected in the wee hours.

but despite my initial experiences, i must say that Fermilab was one of the best places to learn about a great many things, and physics was just one, albeit perhaps the most important.

the particular experiment i was working on was littered with PDP-11s in various stages of vivisection, which they bore uncomplainingly and well. the supercomputers were balkier, apparently due to preparations then being made to put in a great wall of tape drives, an early forerunner of the Petabyte project.

i was greatly impressed by the responsibility given even lowly grad students on the team. this occasionally backfired, as when one of them shut a beam door while the beam was on, causing much clucking and careful examination of radiation badges.

i think this comes with the turf...

i am glad they found interesting and rewarding employment. i know many who never did, and i know many who are struggling today.

snip--

many grad level courses are graded S(satisfactory)-U(unsatisfactory)

snip--

i dont know about this...in my experience we were thrown in the deep end, and expected to master whatever combination of hardware and software existed in the labs. most succeeded and some throve in this environment. in one case, the hardware i was working on failed, and i was expected to design and budget all of the replacement electronics and computers, install it, write the software, and prove the final buttembly functioned as expected. the final step involved (among other things) repeating previous experiments for which the results were known, in order to demonstrate consistency. the analytical software had to pbutt its own, similar tests.

i know very few physicists who ever took any kind of CS clbutt (although there were some that took EE courses to gain experience with fabrication processes.)

quite...

snip--

the improving bit is very important. also, one needs to have some discernment as to which references to trust, and more important, when not to code.

i was once working on reconciling some experimental data with my calculations, using an algorithm stolen from a very popular book. after i had already put in some considerable effort, my advisor wandered by, noticed the book, sweetly informed me that the algorithm in the book was flawed, and directed me to another source. i repeated my thievery, and with great energy began to bludgeon the new routines into doing what i thought i wanted. upon seeing the carnage, my advisor pointed out that while stealing code was good, knowing what to steal was better. he then produced an ancient and dusty tome wherein was inscribed the solution to a very similar problem. some thirty minutes later, with the aid of a pencil and two pages of paper, he produced the result i was seeking. i was left with the thought that (to paraphrase one of Uncle Al's favorite expressions) several weeks of coding had saved me an afternoon in the library.

sidd