[userbinator]

What I find more amazing is the fact that a lot of what could be referred to as "space-age technology" is actually many decades old, and thus was accomplished with a fraction of the processing power and knowledge we have today. Voyager was launched in the late 70s, but based on technology of the 50s and 60s. We visited the Moon almost 50 years ago, using that technology.

If you look at old spacecraft hardware, one thing that stands out is its apparent simplicity and down-to-earth (no pun intended) design --- and I'd argue that this is at least partially responsible for its extreme reliability.

From that perspective, I feel as though developments in modern technology just can't compete for impact; we constantly search for new ways of designing things, wrapping ourselves in endless layers of abstraction and high-level thought, yet aren't really "getting off the ground" and accomplishing something concrete, so to speak.

[matthewmacleod]

From that perspective, I feel as though developments in modern technology just can't compete for impact; we constantly search for new ways of designing things, wrapping ourselves in endless layers of abstraction and high-level thought, yet aren't really "getting off the ground" and accomplishing something concrete, so to speak.

I’m not really sure this is true, though I understand why it might feel that way sometimes.

I’m currently travelling at about 180mph on board a high-speed train in Japan. I flew here on a jet which is something like 20% more efficient than the equivalent from a few years ago. Using the ubiquitous LTE network, I can make a real-time HD video call to my family back in the UK, using my palm-sized, battery-powered computer. I used the same device earlier to do some research about cities as we passed through them, and also to check the CCTV system at home. Over the past couple of weeks I’ve used a similar technology stack to locate my position to meter-level accuracy, to read and translate foreign language text from images in real time, and to record hours of 4K video.

Modern technology is astonishingly powerful - and in some ways, the examples I described above are even more impactful to me on a day-to-day basis than space exploration is. Don’t get me wrong - the latter is still important and exciting! But it’s sometimes too easy to forget the impact of the somewhat more mundane technology that’s all around us.

[coldtea]

>I’m currently travelling at about 180mph on board a high-speed train in Japan.

Which is also (high speed trains) an 30+ year old technology, even if the train you're on was built more recently.

That's what the parent means about those older technologies having more impact. Which is trivially true: earlier low hanging fruits give substantially more bang for the back to progress, and then you get incremental progress and finally marginal returns on any technological fields.

The airplane was a huge development. The modern commercial airliner (50+ years old by now) as well. A jet "which is something like 20% more efficient than the equivalent from a few years ago"? Not so much.

>Using the ubiquitous LTE network, I can make a real-time HD video call to my family back in the UK, using my palm-sized, battery-powered computer.

Which again, compared to the initial impact of the internet and mobile communications it's just an incremental improvement. Being able to send messages and talk from Japan to the UK instantly -- great impact. Being able to send HD video on top of that? Not so much.

[ghaff]

I agree that LTE/HD is incremental. It's probably fair to say that ubiquitous mobile smartphones are something more. (Initial mobile communications less so. When I had a feature phone I often didn't even carry it with me although certainly some people probably found them more transformational.)

[Teichopsia]

You may have missed the last word of his previous sentence. reliability. Which is at least to me, the way I understood it.

Something built back then and still works.

Edit: fix markdown syntax.

[djsumdog]

But they're different systems, designed very differently. A high speed train and a space craft are both designed carefully to maximize their possible life. A cellphone is, unfortunately, been phased into the economy of consumption and planned obsolescence.

Consumer electronics from a few decades are not quite cellphones, but not quite high speed trains or nuclear reactors or space rockets. Many old C64 systems still work or can be restored, and I bet most of our current high end laptops will continue to work a decade from now (you might need to replace the battery).

The OP might have been talking about efficiency, and we have gotten a bit sloppy with that in the consumer world (why does Slack/Atom/Discord need to be a 100MB+ app bundled with its entire web browser and framework? It's like we're in the 2000s with 15 copies of the JDK on your system again!), but once again .. different uses.

A modern SpaceX craft is going to have custom real time operating systems designed specifically to preform much more complex calculations than we've done in previous space missions, hopefully increasing reliability and the amount of sensors we can read, record and transmit data for. The software engineers might be less space efficient in their code than the previous generation, but if the hardware is cheaper and we can increase readability at the expense of memory, why not do it?

In Kim Stanley Robinson's Mars trilogy (highly recommend; best Sci-Fi I've ever read), humans eventually create AI so complex it can manage space factories designed to build from asteroids. The most advanced AI ever created is used to maneuver an asteroid into orbit of Mars while also mining the interior and constructing the cable that would eventually turn into the space elevator over the course of a decade.

[Teichopsia]

Fair enough. I haven't had my morning coffee yet, but I do agree. You make a good point. Thanks for correcting me / explaining.

[nojvek]

Will deffo check the book out.

Electronics aside, voyagers nuclear energy supply fascinates me.

I'm still very optimistic that someday we'll figure out safe micro nuclear reactors.

The energy density of nuclear fuel is just amazing.

[branko_d]

Marry such a nuclear reactor to something like VASIMR engine, and suddenly the entire Solar System opens-up to us.

[tener]

The reliability of these systems is indeed quite impressive, yet it simply isn't a requirement for most of our day to day equipment. OTOH some heavy duty machinery can be very reliable.

[IshKebab]

I'm pretty sure today's spacecraft are very reliable too.

[nojvek]

We always building on what was previously there so one can make that argument for almost about anything.

But we've definitely made progress on a number of things. My rav4 although a lot more complex than Toyotas of 30 years back is a lot safer and efficient.

[Radim]

Focused human attention is still by far the most powerful optimization device we have.

AI hype notwithstanding, no machine learning comes even close -- it's generally a parameter search in a space that's too pedestrian, leaving the real hard work (defining the problem context, goals / objective function, viable tradeoffs and shortcuts) to the human.

Cute AI demos aside, when reliability comes knocking on the door, you end up looking for ways to simplify or avoid the whole bloody mess.

If you spend the time to understand the problem well enough (as you must with space tech), the number of degrees of freedom aka model parameters shrinks. Then suddenly computing power and large-scale parameter searches don't buy you as much; their trade-offs against increased complexity aren't as appealing.

"With four parameters I can fit an elephant, and with five I can make him wiggle his trunk." -- John von Neumann

[jvanderbot]

I've lived through a couple of design reviews for flight software and hardware, and I can tell you this is spot on. The emphasis is on minimal delta from previous products, absolutely nothing remotely risky, and complete assurance of functionality before launch.

[zitterbewegung]

Have you heard of the Opportunity Rover? I believe it has outlasted its designed lifetime a few times now. Also, its still operational.

The only way we can even contact Voyager is the fact that we have better radio signal processing on earth and that has been improving since we launched Voyager.

https://en.wikipedia.org/wiki/Opportunity_(rover)

[coldtea]

>What I find more amazing is the fact that a lot of what could be referred to as "space-age technology" is actually many decades old, and thus was accomplished with a fraction of the processing power and knowledge we have today.

Yes, but then they weren't using it to power immense social networks with pictures of people's food.