Hacker News Reader

Comment by unwind | Hacker News Reader

Clearly, section 3.2.1 is the most important part:

All dimensions are in millimeters. All angular dimensions are in degrees

Implementors in non-metric countries, take note. Nothing is more annoying than arriving at a place where you might want to use your docking interface, only to find out it doesn't work. :)

View on HN

>Nothing is more annoying than arriving at a place (...) to find out it doesn't work. :)

Which has sort-of happened already in history of space exploration, when Mars Climate Orbiter mis-handled Mars orbitial insertion and was lost. The problem was traced down to mis-match of measurement units used in software (Newtons vs. Pound-force).

http://en.wikipedia.org/wiki/Mars_Climate_Orbiter#Cause_of_f...

Yep happened once in 50 years of space travel.

Ugh, we get it, everyone thinks they're being cute and its a way to mock US units. Yes, yes, this isn't getting incredibly tiresome.

But you have to admit the "guy insists on doing things unnecessarily complicated and falls on his face" trope has a strangely entertaining aspect to it :)

The most famous smallest but most significant human error was Mariner I:

"The error had occurred when a symbol was being transcribed by hand in the specification for the guidance program. The writer missed the superscript bar (or overline) in [the formula] by which was meant "the nth smoothed value of the time derivative of a radius R". Without the smoothing function indicated by the bar, the program treated normal minor variations of velocity as if they were serious, causing spurious corrections that sent the rocket off course. It was then destroyed by the Range Safety Officer."

http://en.wikipedia.org/wiki/Mariner_1

As an American, I would prefer that the stubbornness continue to be mocked in a hope that someday we might get rid of this silly system.

IIRC, it was an accumulation of errors when converting between units, not a unit mis-match.

Root cause is still a unit mis-match though.

Not exactly, ultimately it was just that. The problem was that there was a spec for a file format for data from the spacecraft that was supposed to be in metric units, but some of the data was in imperial units, against the spec. This resulted in a course deviation which ultimately doomed the spacecraft. That said, there were warning signs that something was wrong which the ops team didn't pick up on but should have.

Implementers in metric countries need to take note as well, since the degree is not the SI unit for angular dimension.

I think you would notice if you built a dock confusing inches for millimeters.

Reply on news.ycombinator.com