Hacker News Reader

Comment by d_burfoot | Hacker News Reader

One of the most unrealistic things about depictions of futuristic computer systems and UI design is that they almost always do exactly what the character wants with a minimum of hassle. No, technology in the future is going to cause as many problems and as much frustration as it does today. And those problems will be more keenly felt, because in space, your life depends on the technology working properly. There should be episodes where, for example, the main challenge and risk doesn't come from aliens or hostile humans, but from the fact that the OS running the ship's engines crashes and needs to be restarted every 5 minutes.

View on HN

There's one scene where a guy dies because his voice control malfunctioned and he couldn't use the manual controls because he was under too many Gs of acceleration.

It's a flashback to the invention of the fusion drive that allowed humanity to colonize the belts and outer planets, no less. It took place over a century before the events of the series, which itself is estimated to be around 2350 AD [1]. Clearly they've had plenty of time to not only work out the kinks but integrate the technology into society at a level far beyond what we see today with kids and their smartphones. The rock hoppers likely start developing the muscle memory needed to survive in space by grade school.

[1] https://expanse.fandom.com/wiki/Timeline_and_chronology

Unless you're more of a city belter...

Which is probably why they show Detective Miller dropping critical gear first time he's on a space walk (he had never left the Ceres station before the events in the series).

Many of the Millenium Falcon scenes capture frustrating, malfunctioning technology in space very well.

> will be more keenly felt, because in space, your life depends on the technology working properly.

The look on Han's face when he goes to gun it into Hyperspace and there's that failing engine sound effect, as the Empire bears down on them.

Excellent point. Note that Solomon "Sol" Epstein dies as a direct result of his engine working too well and pressing him against his seat such that he couldn't turn it off.

True but if I ever wanted to watch people fumble with software I'd do some pair programming or do a demo.

Also we know pretty well how to build stable rocket control software.

> but from the fact that the OS running the ship's engines crashes and needs to be restarted every 5 minutes.

I would hope by the 23rd century that they’ve figured out that formally verified micro-kernel OSs are the way to go for mission-critical use cases like that.

Of course they figured it out, unfortunately aint nobody got time to rewrite all the software

Haha, yes I’m sure that will still be true a hundred+ years from now. We’re stuck with our initial conditions.

You can’t formally verify behavior with subtle and specific hardware failures. Or in general interaction with reality. You can formally verify only internal logic.

Modern aviation disagrees with you. Yes, systems fail, but they fail very rarely, and usually disasters are long chains of failures, not "need to restart the engine every 5 minutes".

It's likely to assume the same will true for space travel, should it ever happen. Even in our current "duct tape and baling wire" stage of space flight, computer problems occur, but they're not regularly and frustrating, they are unique and pretty catastrophic.

Your inflight entertainment system still won't work with your headphones because you have the wrong plugs, though ;)

What gets me is not the interfaces to control things -- that seems believable -- but rather the ease/speed with which they answer questions about reams of data. Like, I work on a system with a database and to answer any complicated question I am like, "OK let me use the shell and write a query and check some stuff ... and then make sure that I'm asking the right question, and verify that this data looks right...", and here they are in a dozen (or 30) seconds finding all the craft that could have a telemetry that would intercept with some orbit, or cross-referencing passenger/crew manifests with other data.

It bugs me but not enough not to enjoy the show. ;)

It’s just sql with intellisense /s

In a similar vein, I also get bothered by the unrealistic fact that there are never compatibility issues between any given device/technology.

Since we are talking about The Expanse, I think a great example is whatever protocol allows for characters to physically fling data from their devices to other people's devices with a simple swipe.

I'll ignore the fact that sometimes they do this in city streets where they are surrounded by many potential recipients and somehow always deliver the payload to the right person...

But how likely is it that every single device's manufacturer implements that protocol in the same, non-proprietary way?

Because somebody, somewhere, finally got their shit together and decided that it's not a protocol without two independent implementations and a continually updated conformance test suite.

It's the future. It might happen.

That was more or less the plot of 2001

Dave: Open the pod bay doors, HAL.

HAL: I'm sorry, Dave. I'm afraid I can't do that.

Dave opens the debugger and starts stepping through 900 million lines of code

I doubt they’ll be running Windows 98. Now, I could see some poor interfaces that allow you to delete safety precautions by mistake…

Of all the technology featured in The Expanse, I want the holographic displays that perfectly interpret your hand gestures the most.

That's not specific to technology. Applies also to dialog and action. I don't mind this, but I'm getting tired of the transparent screens and random graphics.

I've always thought it would be really interesting to write a show where the characters speak in a way that mimics how real human being communicate with each other on a daily basis.

It might end up being really bad and hard to watch but I think it could be a refreshing alternative to what we have now where every character is able to give an on-the-spot motivational speech with the confidence of someone who rehearsed it for months before delivering it.

Just a real fictional story that includes realistic human dialog.

Check out https://www.imdb.com/title/tt2866360/ Coherence.

This style of improvised natural dialog is called "mumble-core". It's refreshing, but unfortunately not preferred by the highly polished products of TV and cinema today.

If the character is a professional software operator, someone like a programmer who spent years on his Emacs setup, then it actually might be realistic.

Reply on news.ycombinator.com