I wonder if kayoone realizes he submitted the same story about seven months ago. Ironically, I recall reading the story as well on HN when it was submitted yet another time here https://news.ycombinator.com/item?id=4652643.
Must be something about this story, or perhaps the SR-71, that grabs our attention.
> Must be something about this story, or perhaps the SR-71, that grabs our attention.
The SR-71 is a fantastic setting for any story, but I think what grabs me most about the story is the rationality of it. He wakes up in a strange situation, starts deducing where he is, what his situation is, how his safety systems are functioning, etc, and lands safely; then, a thorough investigation replicates the accident, finds all the root causes of the breakup, and addresses them one by one.
I know, but as nobody looked at it back then and i think it is relevant again this week i just posted it a second time. Its a really interesting read, so i hope people dont mind.
I for one have read this story before, and I'll happily continue to re-read it every time I run into it. Exciting and interesting, with plenty of neat details (100-mile turning radius?!). Thanks for posting it.
tl;dr - the plane disintegrated around him before he had the chance to eject, and he floated to the ground thanks to his parachute. Copilot, not so lucky.
Certainly something different than the daily "my framework foobar is the next successor to angular"-kind of posts. I'm actually happy for a bit of... rather unusual content.
As for the article: it's a pity. The US could fly to the moon with less computer power than a dumbphone, put people in things like the SR71 (how did they actually put enough processing power for all the flight computers in a plane back then?!)... and now?
Space Shuttle's been retired, no near-time replacement. No SR71 (or anything that can be seen as a replacement), and stupid Republican teabaggers caring more about their pockets instead of the population. Seriously, America what's wrong with you?
We Germans needed the Marshall plan after WW2, maybe it's needed the other way round this time...
Oddly, I think it's a side effect of the way that technology works. When we humans want to do something we find a way and usually it is extreme and capital intensive. Afterward, we find more economical and less extreme ways of doing it.
The Colosseum in Rome still exists because the only way to do it was to do it in stone and stone lasts forever. Since then we've built millions of arenas that won't last as long because we've used economic forces to develop just-good-enough structure. Likewise, the SR-71 has been replaced by the just-good-enough surveillance satellite.
There's another aspect too. Sometimes we do things just to see that we can. Going to the moon is in that category. If there was a compelling reason to go back we would've. The Concorde, one could argue, was in the same category - we had to have the thrill of taking commercial flight into the supersonic realm, but we found that they economics were such that the thrill of going supersonic and getting there a few hours earlier were not enough to support a market.
The SR-71 was a hack, where the goal was to use 1960s technology to make an areal camera platform that could photograph targets protected by SAMs.
The original design brief was that it should be a stealth U2, but as the A-12/SR-71 project developed it became clear that that wasn't possible with the technology of the time, but that they could make a platform that flew so fast the SAMs wouldn't have time to lock, launch, and intercept.
The problem was that the cost of the keyhole program was incredible, you needed a whole launch platform plus a whole recovery mission every time you wanted to take a picture. So the SR-71 was a cost-saving measure versus Keyhole.
Currently the us military can get the information that the SR-71 could provide more easily and with less danger and cost.
> The SR-71 has been replaced by the just-good-enough surveillance satellite.
Given that the primary motivation for the SR-71 (and, before it, the A-12) was to have a platform wasn't vulnerable to interception like the U-2, I think satellites aren't merely "just good enough", they are far better in that key respect.
They might be worse at "be where we want on demand", but if that's the overriding concern, there's all kinds of other platforms (drones, U-2). The problem with the SR-71 is that it is highly optimized for survivability through speed, but the speed it has doesn't do enough for survivability to be worth the cost given the other alternatives available.
> There's another aspect too. Sometimes we do things just to see that we can. Going to the moon is in the category.
Well, if you ignore the Cold War geopolitical context of the space race, sure, going to the moon was "just to see that we can".
The Colosseum (and much Roman construction) was built with stone but also concrete. The composition of Roman concrete is such that it's much more durable than most modern concrete, due in part to material properties not well understood until recently.
There's also a survivor bias built into structures which have survived the test of time: by definition, what's lasted are the structures which have lasted.
> Certainly something different than the daily "my framework foobar is the next successor to angular"-kind of posts. I'm actually happy for a bit of... rather unusual content.
I'm often extremely critical of off-topic political content, but think that "unusual" content that highlights some interesting bit of the world is great! It definitely fits "intellectually gratifying" far more than politics, or the latest "outrage story".
I'm happy to read well-written articles about things that are new and interesting to me.
A plane doesn't need flight computers, it just needs control systems that react to inputs. These kinds of systems can be totally mechanical, totally digital, or a mixture of both. In the 60's, many of these systems would have been mechanical, with the digital circuitry operating at a higher level to provide the pilots with operational control.
You also have to remember that many of the cycles on a modern desktop are wasted, dealing with the gui, and lots of abstraction layers. Most embedded control systems are much simpler, and are closer to the baremetal, so they don't need GHz of computing power.
Too bad that thing is unmanned. For sure, it massively reduces the complexity (no life support, no need for high-temperature-stable glass, more agility in flight due to much higher g-force tolerances)... but just imagine flying it!
Software developers and entrepreneurs like problem solving and advanced technologies. Stories about aviation and other seemingly unrelated topics could spark ideas as to how to improve something or inspire a different way to solve a problem.
Nothing wrong with day-dreaming. Many of us aspire to be like Elon Musk or Richard Branson, and they certainly spent their fair share dreaming about what's possible with the latest technologies.
It's also interesting to get different groups of peoples perspective on different subjects. If you posted this question on multiple sites dedicated to a subset of the population (developers, entrepreneurs, moms, croquet enthusiasts, stock-car racing enthusiasts, sales-people, 5th graders) it would be interesting to see the different responses, and to see how much group-think there would be depending on how the first few comments sounded.
Some news topics are inappropriate for HN in my opinion. The current news topic about the missing airplane doesn't really need to be here unless there are people whose only source of news is HN.
It's the weekend, quite a few people do like aviation, and the SR-71 is 'futuristic' by most people's standards; arguably the single coolest vehicle ever built.
In the mid-70s an SR-71 was deployed to the UK and of course it was a sensation in that part of the world. Uncharacteristically, I bullied my mother into collecting breakfast cereal box coupons so I could get a snap-together plastic model of the plane.
Also, it's much more of a hacker's plane than the 'waterfall' approach of things like the F-35. The fact that you start the engines by setting off a small explosion on a special truck surely appeals to a lot of people here.
It could be related to StackOverflow displaying "Hot Network Questions" in a sidebar, some of which are from the aviation StackExchange, combined with the usual phenomenon of upvoted stories attracting similar submissions.
Incidentally, I find this sidebar extremely distracting; maybe it's time for a plugin/script to fix it.
I am in software because aviation engineering was more than I wanted to handle at the time. I needed to be less challenge adverse in my younger days and that is one bit of advice I give to the children of friends who ask we what I do.
The ST-71 is the result of groundbreaking engineering which flies three times higher than most other aircraft and goes so fast that it needs protection from heat even that high up in the atmosphere. And it looks awesome.
Is anyone here familiar with how flight testing is actually performed?
One thing I wonder about after reading this is why, if they were able to recreate the entire flight profile in a simulator the next day and see the same result, would they not just test everything in the simulator first?
Because testing every sequence would take exponential time relative to the length of the sequence.
For the post-accident simulation, they say "ok since we know we are in this modification of the SR-71, in this phase of the flight, and this accident happens, and this backup fails... ok yeah it looks like that's pretty bad.
I misunderstood the article I guess. I thought they knew specifically what they were testing, i.e. 'we have altered the airplane this way and we will now do these things.' After the failure, they were able to essentially reconfigure the simulator to match the test parameters, rerun the test and observe the same failure in simulation.
You can't test everything. There are so many independent variables, to try to test every possible scenario is a combinatorial nightmare. More so when you factor in transient events like engine unstarts and so forth.
However, "the right engine inlet's automatic control system malfunctioned".
That probably wasn't part of their simulation. While it sounds like they could have erred on the side of caution and cancelled the remaining tests after that occurred, they must have been confident in their manual control abilities.
Even if they had simulated with the inlet(s) under manual control, real life airflow and reaction times are going to vary.
They are cool. A bunch of them were stationed at an Air National Guard base near my house and they would frequently fly overhead. Eventually they moved elsewhere; which is just as well. They are noisy too.
amazing--"Touching the Void" for the aviation world.
"[E]xtracting myself from the parachute harness, I discovered the source of those flapping-strap noises heard on the way down. My seat belt and shoulder harness were still draped around me, attached and latched."
in the blockbuster film adaptation of this account, (starring Keanu Reeves) the pilot is still holding the stick when his parachute brings him to the ground.
