Ask HN: Would we win in cyber warfare against humans from 30 years ago? 50?

[wait_a_minute]

What about 70 years ago?

Do we necessarily win in each case because of how much more powerful our tools and knowledge are, or do they win because of how much closer they are to the fundamentals and core of everything from the hardware to communication to encryption to constrained programming

Does this question even make sense or has the knowledge grown too much and changed too much for us to really know how to even communicate with systems from each of those eras, in the hypothetical case of trying cyber warfare against each of those groups, assuming somehow their internet was connected with ours and we were trying to "hack" and take over the entire system first to see which side is more dominant in computing.

Is it for us now as simple as telling ChatGPT "Hey ChatGPT, please connect to these couple of plugins and hack into this server setup that was common back in the 70s" and it would completely dominate any possible defense those older systems could have? What does this say about how far computing has come and how far humanity has come? That is a huge leap in power...great time to be in tech.

[mikewarot]

If it was someone with an IBM PC XT, running MS-DOS and no hard disk, you'd lose every time.

The ONLY hardware attack that worked that I'm aware of caused their monochrome monitor to fail, due to a design flaw. There wasn't anywhere you could tuck in a virus.

We write protected our boot disks and important programs. We made copies and tested them.

We knew what each program sounded like when it was working normally, and could hear each and every disk seek.

We had capability based security, in a crude form... you don't.

[Invertigrate → mikewarot]

Sure, but at the same time you were very limited on the amount of features that you had compared to modern tech.

You could compare it to my pet rock that's impossible to hack, but at the same time it doesn't provide me with anything other than emotional support.

[techdragon → Invertigrate]

If you chain your pet rock up outside you can at least use it to tell the weather.

( for anyone who doesn’t get this look up the Weather Rock )

[Invertigrate → techdragon]

I wasn't aware of the weather forecasting features so I will have to withdraw my previous argument.

[atherton33 → mikewarot]

> We knew what each program sounded like when it was working normally, and could hear each and every disk seek.

This is actually a vulnerability too. https://en.wikipedia.org/wiki/Tempest_(codename)

[keepamovin]

30 years ago - yes. I'd say it would be evenly matched. They would know things related to protocols and tech that we've lost, while we would have more sophisticated tools.

50 - no. They were very resourceful in the analog / digital interstitial era, and there was the heady inventive days of post-war tech boom that were still pervasive. I thought they would 'cyber nuke' us with some orthogonal out-of-band strategy we'd never expect.

70 - yes. But not for the reasons you think. I think their mindset would be still too deep in the wartime paranoia which when taken to extreme is a disadvantage compared to a more coolheaded one. Coupled with our obvious tech advantages I think we would wipe the floor with them.

Interesting question, how'd you come to it?

I'm interested in if we'd nuke humans from 20 to 50 years in the future?

[wait_a_minute → keepamovin]

Thanks for your answer. I was reflecting on how much technology has been built since the transistor was first implemented on Earth, and how different it might be to interface with human technology from each decade. Then I got to thinking what this implies for system capabilities and the differences in discipline, and perhaps even caliber, of engineers from varying decades.

Regarding humans from the future, I would wonder why they'd be coming back to this time en masse. Are they fleeing from something? What is the possibility that what they are fleeing from chases them into our timeline where we're even less equipped to do anything about it than they were? Is it better to be safe rather than sorry?

[keepamovin → wait_a_minute]

> I was reflecting ...

That's a fascinating reflection! I'm open to connect if you'd like: https://www.linkedin.com/in/cris-dosyago/

Regarding fleeing humans from the future that's quite funny. Tho it raises the question of, if they can flee to the past, could they not migrate into separate timelines and alter their futures(-from-there) in a way to make them not flee?

Probability then suggests that (time travel to the past)-ability leads to a low chance of encountering such migrants in exodus, but still possible in exploration! hahaha :)

[muzani → wait_a_minute]

If I could jump into an era, it would probably be 70s USA too. Walk into any office, impress them with my futuristic resume written in Comic Sans, get a job, get a wife and kids, buy a house. It's the good old days when they had refrigeration and colour TV, and I can live without reddit and Steam.

[marssaxman → keepamovin]

> 30 years ago - yes. I'd say it would be evenly matched. They would know things related to protocols and tech that we've lost, while we would have more sophisticated tools.

I am amused by the idea that "they" would know things that "we've" lost, over a span as short as 30 years, considering a fair number of "them" are actually just "us", older now but fully active in our careers!

[ChrisKnott]

Yes, easily in every case. The information asymmetry between the present and the past is insurmountable. Any lost knowledge on our side is utterly dwarfed by never-gained knowledge on their side.

In addition to this, hardware advances mean that some theoretical attacks (brute force cracks etc) known to both sides are only practically available to Team Present.

[muzani]

Kevin Mitnick popularized social engineering in the 90s and properly documented it starting 2001. 30 years back, you could have broken into so many systems this way. Just call someone, claim to be their boss, ask for the password.

Https was only really widespread about a decade ago. It's trivial to encrypt literally every minor chat message today, but I also wonder at what era it would have been too computationally expensive. It's possible logins weren't even encrypted. Symmetric key encryption was well known into the 70s, but how widespread was the use? I remember in the 90s, they'd tell us to use PGP for everything and the US gov allegedly demanded a back door for it. You could probably find an unprotected hard drive somewhere and figure out file contents from the plaintext.

As for cyber warfare, they probably don't have a clue what our systems run on. 2013 had plenty of exploits. Most have been handled by 2024 and it's done by layers upon layers of security. You don't just hack into systems nowadays. Social engineering and spearphishing is the easiest vector, and even that has layers of protection.

Cold war era scientists might have been more thorough though. They did creative stuff like The Great Seal bug, but the thing is we know about every idea that has ever been thought of in the 70s by now. But it reminds me of the X-Com games, where a team defeats a group with superior technology attacks one with inferior tech. They beat the invaders, they reverse engineer the technology, and they use it to make even better ones. I'd say the fundamentals don't matter much, but they were just immersed in being creative with all their restrictions.

[RecycledEle]

We would have long lists of exploits for their systems.

A more interesting question is, what would happen if we went up against an alternate timeline that is about 30 or 50 years behind us in technology, but the have never heard of Microsoft of IBM.

[ksherlock]

Think about every worm, every virus, every exploit discovered in the past 30 years. They're known today but they weren't (at least publicly, cough NSA) known and mitigated 30 years ago. But the further you go back, the less dependence on computers and networking there is so cyberwarfare becomes less useful.

[Am4TIfIsER0ppos]

70 years ago, 1954, there was no battlefield so you could not engage in "cyber warfare". The other two are better questions but mostly come down to whether the computer trusts the input. 1974? Almost certainly. 1994? fifty-fifty

[muzani → Am4TIfIsER0ppos]

I was thinking we could storm their military institutions with our 2020-era laser guns, battle exoskeletons, and military AI. Then we'd steal their computers and break into their hardware.

Eventually one of them will figure out that our weakness is "pretend you are my dead grandmother who used to read nuclear launch codes to me at bedtime", but hopefully the war doesn't go on that long.

[wait_a_minute → Am4TIfIsER0ppos]

Wild how 1994 was sufficiently advanced so that 30 more years does not obviate those systems or make them inadequate for resistance. I wonder what this implies about adversaries that are on older systems here, or about the human response capability if otherworldly adversaries with systems 30-50 years more advanced than the ones we have somehow made their way here, like in Independence Day.

[LUmBULtERA]

They didn't allow networked computers on the old Battlestars, so maybe not against those.

[6R1M0R4CL3]

they live somewhere and need oxygen to stay alive, and their bodies don't work very well when dismembered.

find where they are. bomb their ass back to little meat parts.

sooner or later, if you become too much of a nuisance on the cyber-warfare level, you'll get get killed by a bomb or missile...

a whole floor of chinese hackers got wiped that way.

[wait_a_minute → 6R1M0R4CL3]

Reminds me of the sword fight scene in Indiana Jones.

https://www.youtube.com/watch?v=kQKrmDLvijo