From Mel Pheasant · 1996-03-26 12:51 · #73129 in FREE25
In reply to An political question by Paul Edwards
LE>It was still a mess.... It didn't help that the original programmer LE>normally wrote COBOL programs for the mainframes. (ever see a Pascal LE>procedure 12 *pages* long? <shudder>) :) You want a real "shudder"? See if you can acquire a copy of the GTDS (Goddard trajectory determination system) software from JPL labs. The software is now in the public domain. It is the very same software that NASA used for all the earl space missions, from low orbit trajectory and maintenance to complete interplanetary missions. The version I had to evaluate for a sub-contract was ported from the ancient IBMs NASA used to a VAX, and then later from VAX to PC. Can you say UGH? This puppy was, at one time, written in Fortran on cards (probably originally on an even older format). There are no comments, and there are miles and miles of complex mathematical integrators chalked full of tape I/O low level programming and other crap. Not to mention many different mathematical models for computing orbit perturbations due to things like oblateness of the Earth, upper atmosphere models, planetary positions, time epochs, nutation of the Earth's axis, solar radiation pressure, drag, and a whole pile of other stuff! Unfortunately, somewhere in the port from the VAX->PC, a time or date calculation was wrong, so the software didn't work anymore. Still, I'm amazed that the thing even worked originally! Now *that* is scary software! --- msged 1.99S ZTC * Origin: Talking To A Wall, Ottawa, Canada (1:163/207)