Was the internet designed to survive a nuclear attack?
siliconfolklore.com
siliconfolklore.com
In this case, Wikipedia makes it pretty clear that while surviving a nuclear attack specifically might not have been an explicit part of the design goals of the designers (which may have started the myth myth), resilience against large-scale failures was exactly the feature that made the technology interesting to its military sponsors:
> Nonetheless, according to Stephen J. Lukasik, who as deputy director (1967–1970) and Director of DARPA (1970–1975) was "the person who signed most of the checks for Arpanet's development":
>> The goal was to exploit new computer technologies to meet the needs of military command and control against nuclear threats, achieve survivable control of US nuclear forces, and improve military tactical and management decision making.
(https://en.wikipedia.org/wiki/ARPANET -- where the quote is sourced directly to an article by this Stephen Lukasik.)
My takeaway is that while the ARPANET creators might not themselves have particularly cared about surviving nuclear disasters, without the completely decentralized packet switching that made this at least a possibility, it wouldn't have ever received the funding to get off the ground.
Shortening this fact to "ARPANET was funded to survive a nuclear attack" is IMO a fair sound-bite sized summary, not a "myth" to be debunked, even though the story has obviously more nuances.
It's also important to mention that Lukasik wrote this article in 2011, over 40 years after the fact, and that the claim you quote is contradicted by pretty much everyone else involved at the time, including Lukasik's boss at the time ARPANET was conceived, Charles Herzfeld (as cited in the same wikipedia article.)
Betteridge's law - "Any headline that ends in a question mark can be answered by the word no."
The military has to plan for the full range of the scale, from small brush fire near a minor ammo depo to full blown WWIII. But the part that is 75% towards full blown WWIII is so often overlooked by commentators it's like it doesn't exist. I could come up with 100s of scenarios where most of the internet works but there is a recently-started nuclear war. (India vs Pakistan, for example.)
However as I'm in the UK, a large nuke in Docklands would cause big problems. Most of my company's connectivity reaches the world through Telehouse and Sovereign House, both in Docklands, about a mile from each other. A large nuke landing on top of the Blackwall Tunnel - perhaps on top of Switch (another major building we have), would probably knock both out.
You could effectively knock out the UK's internet by sabotaging the Thames Barrier, let alone the damage just a single nuclear bomb in docklands would do.
How much of routing would then happen after that first strike? How many other centralized places of the decentralized internet would need to be targeted in the first strike?
1.1.1.1 and 8.8.8.8 are anycasted addresses, they don't have one location, and the location you are connected to differences based on your location. Anycast was one of the earliest things formalized by the IETF and originally created in 1989 as a method of addressing using IP, /specifically/ for resiliency and performance.
Maybe your ISP didn't have a fast enough backbone connection to support all of its customers, or maybe one of its internal routers became overloaded. Or maybe it literally didn't have enough phone lines to support all of the customers simultaneously dialing in.
It's also very likely that your ISP had backhaul issues, and myraid other issues existed related to capacity.
Ham radio is still used by many people, and will never go away.
https://arstechnica.com/information-technology/2019/06/bgp-m...
https://arstechnica.com/information-technology/2018/11/stran...
The fantasy that life could still be lived after nuclear attack is disturbing in itself without any further "strategy".
What use would the internet be at that point? Even if you survived, your body and food is now under non-stop attack from radiation, you have far bigger problems.
Look at how miserable the world is that everyone literally gave up on protecting themselves from a pandemic.
Now imagine widespread radiation with the same mentality "well you can't see it so it can't hurt you" and further fantasy "only old people die from it, I'll be fine"
Regarding that question mark, David Walden was actually the software designer and Severo Ornstein his hardware counterpart.
Compare:
> In 1968 BBN was preparing to bid on the contract from ARPA to develop the ARPANET Interface Message Processors (IMPs) [ARPA68]. Frank Heart, Bob Kahn, Severo Ornstein, and David Walden were the main members of BBN’s proposal team (although other BBN people participated) with Heart as the team leader, Kahn with prior background in the concepts of packet switching, Ornstein as the hardware designer, and Walden as the software designer. Shortly before the due date for BBN’s proposal [BBN68] to ARPA, Will Crowther was added to the team as another (more senior) software person; and after BBN was awarded the contract, Ben Barker was added as a hardware designer and Bernie Cosell as a third software person. BBN was awarded to IMP development contract with a start date of January 1, 1969.
[1] https://walden-family.com/bbn/imp-code.pdf
BTW, as I've read it, the use of packet switching was all about connection times, connecting from hop to hop versus the time required for establishing a cross-continental circuit switching line. (And yes, packet switching for telephony had once been pitched for its military robustness, but this is not why it was considered for ARPANET.)
If their hypothesis was correct, that moving matter in relation to other matter generated a field exchange, that meant we could each have a "gravity transmitter" and "gravity receiver" that could act independently of any wired or wireless systems.
Having point to point high speed communications from anywhere on the planet to anywhere else on the planet, which themselves weren't governed or use restricted, would be game changing. Alas, current theories and decades of research haven't given us much to go on.
If you could develop a sensor that picked up on just a small segment of the fourier vibrations going on, identify it, and then respond, all of this theory still seems like it would hold true. It's just practically impossible.
In reality, they worked on radiation-hardened point to point multipath comms that allow them to evade jamming and know who they are talking to.
I'd put more weight on Baran's claims. He has primacy, it was clearly documented in the Rand Techical Reports and then retold well in the Wired article with Stewart Brand.
Yes, Cerf, Kahn et al revived and invented a good amount of it, but the ideas had been circulating for a decade by that point, at the phone companies and US military, which is where the academic money came from anyways.
Primary and Secondary were running 24/7 with Secondary being in exact sync with Primary. If Primary or Secondary went down, Tertiary could be brought up in sync within 24 hours. Primary and Secondary were geographically separated (east/west coast). Tertiary was in the mid-west. After that, the additional facilities would take days to bring up, and then a bit of time to sync.
The specific bank is no longer alive as it was consumed by one of the "too big to fail" institutions.
The insanity of humanity all here on HN...
Many people fail to see how dangerous this war is. It could readily lead to an all-out nuclear war between all the nuclear powers on the globe. We've been stirring the hornets nest and Russia is now calling our bluff.
Some believe that if Putin is removed Russia will somehow become a model Democracy, but that couldn't be farther from the truth. If Putin is replaced it will be by some ultra-nationalist which will spell even more trouble.
As far as Russia's concerned NATO is already a combatant in this conflict. It will only take a slight nudge to light the fuse.
I don't know about "inevitably" but I'd agree to "very likely." On the other hand, NATO not being directly involved in the Ukraine makes it less likely that nukes will be used in the Ukraine. When the other side has nukes, the fog-of-war means you may wrongly think that they were used, and retaliate with nukes.
I'm not an amateur.
But I guess that is why we also in the middle of climate disaster.
It is funny though, the very basics eg. climate control, clean water, unspoiled food... More important than internet.
https://www.webpal.org/SAFE/aaarecovery/1_farm_recovery/uv.h...
As long as there's electricity and as more of the internet moves into outer space, it'll work.
https://en.wikipedia.org/wiki/Square_Leg
I appreciate that anyone who has seen Threads might find that difficult to believe.
>It was assumed that 131 nuclear weapons would fall on Britain with a total yield of 205 megatons (69 ground burst; 62 air burst)[1] with yields of 500 KT to 3 MT[2] That was felt to be a reasonably realistic scenario, but the report stated that a total strike in excess of 1,000 megatons would be likely.
But also that the scenario was overestimated:
>Square Leg was criticised for a number of reasons: the weapons used were exclusively in the high-yield megaton range, with an average of 1.5 megatons per bomb, but a realistic attack based on known Soviet capabilities would have seen mixed weapons yields, including many missile-based warheads in the low-hundred-kiloton range.
In other words, more small bombs does way more destruction than one big bomb if we assume the same total yield. To give a specific example, imagine dropping one 1500 kiloton bomb vs five 300 kiloton bombs. In the first scenario the destructive power would be about 11.5 units of damage. In the second example it would be 5 * 6.7 = 33.5 units - assuming no overlap.
In other other words, the report assuming a small number of high yield weapons (instead of larger number of lower yield weapons - but an otherwise comparable total yield) was understating the impact of nuclear war, potentially dramatically.
In other other other words - the report was being criticized for understating the impact of nuclear war in both cases.
In the last months, we've seen how the Red army performs in conventional war. They were supposed to have 10,000 - 14,000 tanks, including modern ones, but we now see that they barely had any modern tanks and that after losing 2,500 tanks from the 60s, 70s and 80s, they started begging Belarus for their old tanks. They also claimed to have a modern air force, but until now, they have been unable to establish air dominance against a country that barely had any air force. And they were supposed to have a navy, but also that was an exaggeration. They claimed to have precision-guided missiles, and even supersonic missiles. They ran out of those long time ago, and they weren't that precise or fast. What they had was a lot of empty posture.
Why do we believe that they are any better at nuclear war than they are at waging conventional war? Whatever is true for a simple conventional war, is even more true for a nuclear war: Their missiles and warheads are old and unmaintained, and far fewer in numbers than they pretend.
The effects of corruption runs so deep in their society. Just like you can't buy any complex Russian products in your local stores, and just like you never wanted a Russian car, Russian military is completely incapable of successfully waging war against the West.
Can Russia launch a nuclear attack against NATO? Probably, yes. But it would be much smaller than the official numbers suggest, and layered Western missile shields would intercept a percentage of those missiles. Some missiles would miss their target and some would not detonate.
Different studies give different estimates ranging from tens of millions to hundreds of millions dead in the immediate aftermath of a strike, but studies assume that Russia is actually capable of carrying out a strike and penetrate our defenses with a large number weapons that work. Estimates of dead afterwards, are usually from starvation, not from radiation.
It's been 40-50 years since Russia actually had strength comparable to NATO. Even back then, humanity would have survived a nuclear war. Nowadays, both Russia and the free world has fewer weapons, and we now know that Russian military capabilities are routinely exaggerated. A nuclear war with Russia would be terrible, but it would be over quickly, it would end in Russian defeat, and most of us would live on afterwards.
I meet a lot of people who think their life will be short and nasty if Russia drops a nuclear weapon over Ukraine (which, at a distance of 2000km, would affect us not at all).
My point was only to reply to the person thinking that a nuclear strike in Ukraine affects them not at all somewhere else in Europe, which I think is extremely shortsighted.
Lack of water could be an issue in some areas but simply walking to fresh though not clean drinking water is viable in most areas.
Violence after disasters is more trope than reality. People will loot after a disaster, but it’s not like people are going to show up to Best Buy and work after a nuclear war. Looting is more useful to humanity than simply letting food rot because people can’t work the self checkout machine.
And these tangential issues would emerge in spades after nuclear war. For instance one of the most basic is logistics. Think about how insanely much is just involved in getting an ear of corn from a farm to you. At the most fundamental you're relying on machinery, drivers, oil, gas, refineries, roads infrastructure, food processing infrastructure and a whole lot more - before that ear of corn is anywhere near you. And then there needs to be a means of distribution, and more.
Our entire economy came on the verge of collapse because of COVID, and nuclear war would have an impact with so many orders of magnitude greater than COVID that I think even dreaming of anything even vaguely resembling normalcy afterwards is unrealistic. Violence happens after disasters when the food runs out. In practical disasters that tends to not happen due to rapid aid, delivery, support, and organization. In nuclear war, the food will run out.
Saying the current social order breaks down isn’t enough to suggest Anarchy will take it’s place. A farmer who wants to collect a harvest without oil will want manpower, and survivors will want to eat. That’s the basis of civilization.
Sure suburbs in their current form don’t work without cars and oil. But, people can migrate quite far on a half a tank of gas and their feet.
And within these stories one can also see what happens to societies once normal social function starts to collapse. While this [1] article reads like a propaganda piece, the tales it recounts are neither unique nor exceptional given the situation. Of course it does feel weird to be here kind of arguing that, "Yes we will starve." I think my main point and motivation is that the consequences of what we're discussing simply cannot be overstated. There is a dangerous amount of misinformation that increasingly seems to be headed towards, 'Maybe nuclear wouldn't be 'that' bad.' Perhaps not misinformation. Because it's true that it wouldn't be 'that' bad, it'd be vastly worse.
[1] - https://www.theguardian.com/world/2013/jan/01/china-great-fa...
The Irish potato famine occurred in part because of the potato blight, but it was a multi year issue that at it’s core related to growing exports (cotton, grain) in favor of local food. Russia’s industrialization created famine because so many farm workers where sent to factories. China had similar root causes, the 18year old worker in that story went home to see their family and many fields where fallow. Steal all the food or workers from a village and people starve.
That’s not to say drought etc didn’t play a role, but the American dust bowl isn’t remembered as a famine because there was feedback between the amount of food crops planted and the need for food.
On the other hand WWII involved most of South America, Africa, and Asia to some degree. If WWIII got similarly out of hand then nukes might take out all of the worlds chip making capacity, oil refineries, and deep sea ports as strategic assets making thing far more unpleasant.
What military targets there were would be significantly reduced today. We were targeting military targets, not to target their military - but to try to destroy their means of nuclear delivery (bombers). Today in the ICBM era, that means of delivery is largely nuclear fortified silos spread far out and in the middle of nowhere. Poor targets.
Nuclear war isn't about defeating the enemy, it's about destroying them. If Russia nuked 100 military targets today, then tomorrow there'd be 20 million Americans looking to enlist, and every single aspect of manufacturing we have domestically or can influence abroad would be directed towards rebuilding at an unprecedented rate.
The way you overcome this is by focusing on that rebuilding: population, industry, economy, and even medical. You probably have a much better chance of surviving a nuclear war if you live out in the middle of nowhere, even beside a nuclear silo, than you do living in a common population center. Though with the goal of destruction, even that may be insufficient since there's no reason to expect things like salted bombs [2] or other doomsday type weaponry would not come into play.
[1] - https://nsarchive2.gwu.edu/nukevault/ebb538-Cold-War-Nuclear...
[1] https://www.bbc.com/news/magazine-30500591
It'll likely be the end of civilisation as you know it though.
I mean, perhaps they do have solid and tested contingency plans for events like that, or the military has some sort of access they could use to rig it into something useful, but I don't really believe it.
I also have no particular reason to think that Russia would use their very limited ICBM stockpile to target data centers rather than military facilities.
The goal of nuclear war, absent any chance for a realistic "win" conditions it to ensure that absolute destruction of the opposing country - beyond the ability to ever rebuild to a comparable relative strength. Our nuke targets from during the Cold War have been declassified [1], and it's likely that the Soviet Union's were similarly themed: population, industry, economic, medical. There's an extremely strong overlap between those areas and data centers.
In any case this is also all slightly tangential. One of the first things that will happen during WW3 is the widescale mutual destruction of satellites in the sky, for both offensive and defensive measures.
[1] - https://nsarchive2.gwu.edu/nukevault/ebb538-Cold-War-Nuclear...
Not sure how useful that would be though.
If a groundstation continues to work then even better, that bgan in your bag may well still work - assuming there's not too much radiation and dust in the atmosphere
Again, still not sure what the point is.
Hardlines between local command areas though - far more useful, and less likely to have been taken out.
That is what you said. Think about that and everything you do from now on and everyone you vote for and every piece of hate and division you have in your heart.
Who do you think pays you?
That might be what you read. In that case work on the hate and division in your heart before you preach.
In the time when the ARPANET was designed, the ICBM's accuracy was far lower, so a packet-switched network with lots of redundant links had a better chance of surviving for more time than a circuit-switched.
The idea was to have a communications network that fared a better chance to survive time enough to be able to be used to coordinate a counter-attack for a few hours or days. A network that would degrade more graciously but that would surely degrade the same in the end.
The type of person who jumps at the chance to recite "nuclear stuff" from back when the school I was in still had East Germany on the class globe while seeming to actively refuse to accept new information is the reason there still exists a risk of nuclear exchange.
A lot of these Americans (parent seems American, I'm also American -- I can use the word) keep forgetting: there is no draft, Iraq was a war crime, you could have chosen a different path.
I can chart out how one year in the 90s, the parallel evils of refusing to hire more auditors nor upgrade infrastructure cascaded through the federal government, culiminating in the OPM breach.
I should stop making these long Slashdot-esque rants under two people I've probably never met, because the last year it was gonna not be weird for me to trick or treat was the one after 9/11.
We are in a multipolar world now, in every sense of the word, and folks should adjust to the fact that after many conversations, with many people, in many places, it's my understanding folks, if they support nukes at all, support a robust second strike policy paired with what I would term "armed neutrality".
I can remember in college, I'd say we should get rid of the nuclear triad, instead focus on taking the existing stockpile, checking there's no... leaks or anything like that... and loading them onto cruise missles to be put on submarines or in those fancy ass stealth planes and the same homophobic, classist, ableist folks who used to use the draft to force people to fight their wars would repeat the same tired talking points from the cold war.
There's a feeling in the air like there was in 2002. They talked about whether they'd let kids trick or treat, as if some asshole was going to... what? Anthrax the candy?
The real threat was the type of person who gets an increasing edge in their voice that every year you come to their house, politely take one candy, and make your way in a big loop around the block but won't drink their kool aid or whatever. Like, yeah lady, I get it... you want me to show I trust you. I don't. I'm gonna take a candy bar and pretend I care if you live another year since you told on yourself -- and I won't be back.