Live data from Hacker News

NASA can't figure out what's causing computer issues on the Hubble telescope

npr.org

311–320 of 333 posts

Re: NASA can't figure out what's causing computer issues on the Hubble telescope

#311
post #7

pushes up glasses I would watch the heck out of a Twitch stream of their debugging/brainstorming sessions. I always loved the movie Apollo 13 , especially the technical troubleshooting parts.

I am halfway through the second season of For All Mankind, it's a series based on an accelerated (versus the decelerated) series of events around the time of the Apollo missions. I think you may enjoy it!

Re: NASA can't figure out what's causing computer issues on the Hubble telescope

#312
post #213
post #17

I know that, at different times, NASA has used Forth[1] and Lisp[2] in some of their space applications. Both of these languages offer REPLs that generally accelerate the debugging process, and while your "average" Lisp might be unsuitable for hard real-time applications (due to the presence of a garbage collector, usually without the hard real-time constraints that you can get out of garbage collectors with extreme…

Forth yes, lisp not. Lisp was only used on ground to simulate the rover. Also a repl in space only makes sense in earth orbit, but not farther away, with 8-20min waiting time for a packet roundtrip to Mars. Those machines really need proper and faster decision making (AI, think lots of `if` statements and proper modeling) on board, eg to perform landing or docking maneuvers. Or to detect and workaround radiation dama…

You might check on how New Horizons was implemented. They initially wanted to fly LISP, but (I think) they eventually used the LISP to generate C++. This was to allow dynamic, model based re-planning when the craft was in the outer reaches of the solar system.

Re: NASA can't figure out what's causing computer issues on the Hubble telescope

#313

Anyone else look at the headline and feel this is one of the dumbest headlines ever. It makes it sound like NASA's incompetent. 'Why haven't you fixed Hubble yet?'. There doesn't seem to be any nuance or respect that they're trying to repair an orbiting telescope that was launched 30 years ago and designed 40 years ago - and that people are patiently trying to sort through a fully autonomous system 400 miles above th…

I've worked on some legacy code in my time but debugging this thing has gotta be completely nuts.

Re: NASA can't figure out what's causing computer issues on the Hubble telescope

#314
post #305

Earlier quoted context omitted.

Interestingly, we humans actually have more than just Hubble floating around up there in the way of telescopes. https://en.wikipedia.org/wiki/List_of_space_telescopes

Missing from the list are the space telescopes that look inwards: https://en.wikipedia.org/wiki/KH-11_Kennen

Fascinating buried lead:

> A NASA history of the Hubble,[24] in discussing the reasons for switching from a 3-meter main mirror to a 2.4-meter design, states: "In addition, changing to a 2.4-meter mirror would lessen fabrication costs by using manufacturing technologies developed for military spy satellites."

Hubble was derived from these spy satellites.

Re: NASA can't figure out what's causing computer issues on the Hubble telescope

#315
post #104
post #7

pushes up glasses I would watch the heck out of a Twitch stream of their debugging/brainstorming sessions. I always loved the movie Apollo 13 , especially the technical troubleshooting parts.

In case you haven't seen it: https://apolloinrealtime.org/13/

Wow - so here I am an hour later... Thanks for that link!

Re: NASA can't figure out what's causing computer issues on the Hubble telescope

#316

Earlier quoted context omitted.

Most of my work could benefit from a journalist > Overworked Engineer misses semicolon. All night review session finds it, data gets loaded! > Management insists Jira stories be routed to new Epic. Team Lead spends hours learning Jira API before giving up and doing it "the hard way"

Completely not related to the topic at hand but ;) I love this "spend hours figuring out the API" then giving up. Maybe it's just me but I have noticed that a lot (and it mean a lot) of devs will overestimate greatly on anything manual they need to do that they don't like. And spend hours if not days trying to automate it. Which is fine if it's gonna be needed again soon or over and over. But really, what is so bad i…

At least automating it can be kind of fun and educational.

Re: NASA can't figure out what's causing computer issues on the Hubble telescope

#317
I had no idea that the Hubble had been serviced by shuttle crews (multiple times even). Exactly matching orbits with a tiny object zooming around the earth at 10s of thousands of kph sounds like a very difficult and impressive feat.

Re: NASA can't figure out what's causing computer issues on the Hubble telescope

#318
post #216

Earlier quoted context omitted.

Hubble is basically a repurposed spy satellite so it may still be sensitive. Although I doubt any of its sisters are still flying.

On the contrary I doubt that the US military would give up advanced imaging technology, like reading car plates from space, for nothing. That's the Hubble. Nothing else comes close.

I don't think anybody else believes that they would, either.

The question is not "is the US military still doing telescope-based surveillance?" it's "is the US military doing it with Hubble-class stuff, or are they doing it with newer stuff?"

Given the age and amount of hands-on maintenance required to keep the Hubble working, it's almost certain they're doing it with newer stuff.

Re: NASA can't figure out what's causing computer issues on the Hubble telescope

#319
post #305

Earlier quoted context omitted.

Missing from the list are the space telescopes that look inwards: https://en.wikipedia.org/wiki/KH-11_Kennen

Fascinating buried lead: > A NASA history of the Hubble,[24] in discussing the reasons for switching from a 3-meter main mirror to a 2.4-meter design, states: "In addition, changing to a 2.4-meter mirror would lessen fabrication costs by using manufacturing technologies developed for military spy satellites." Hubble was derived from these spy satellites.

What's mindboggling to me is the sheer number of spy satellites that were more or less each equivalent to Hubble.

There are estimated to have been something like seventeen of the KH-11 satellites, each one of which cost something like 2/3s of a Nimitz class aircraft carrier.

https://en.wikipedia.org/wiki/KH-11_Kennen

And there are an unknown number of its successor, but probably at least 3, where they updated it for stealth.

https://en.wikipedia.org/wiki/Misty_(satellite)

It feels like a parallel universe where what is a technical struggle and tour de force here, is churned out by the dozens.

Re: NASA can't figure out what's causing computer issues on the Hubble telescope

#320

Anyone else look at the headline and feel this is one of the dumbest headlines ever. It makes it sound like NASA's incompetent. 'Why haven't you fixed Hubble yet?'. There doesn't seem to be any nuance or respect that they're trying to repair an orbiting telescope that was launched 30 years ago and designed 40 years ago - and that people are patiently trying to sort through a fully autonomous system 400 miles above th…

Saying “can’t figure out” gives me the impression that they’ve tried their hardest and have concluded they are unable to diagnose or solve the problem, which is not accurate.
Post reply on HN