The Box is way more fascinating than it has any right to be. It touches on basically every aspect of post-1950 world economy, history, urban planning, social changes, migration, etc.
One interesting take I’ve read on that quote is that you need to consider the source. If you’re a general for a country that routinely deploys troops to the other side of the world, in the service that specializes in landing those troops on a hostile beach with no friendly infrastructure, in a milieu when it is no longer considered acceptable to supply your troops by allowing them to pillage the countryside, you are going to spend a lot more time worrying about logistics.
It's not just remote deployments where it matters.
There is no deployment scenario that is not heavily dependant on logistics. This is more true in protracted land deployments of troops across large distances, not less so. No matter the strategy, logistics must match it or success is significantly more difficult.
Take Napoleon w/ Russia, where Napoleon's ambition out ran his logistics. His strategy and tactics had yielded results until then, and despite a large effort to supply his troops, he was woefully under prepared and in the end it was his logistics that failed him: Russia's retreating scorched-earth strategy meant Russia was retreating into friendly territory with resupply, the French were extending into a no man's land. The Napoleonic forces originally outnumbered Russian forces about 2.5 to 1, but forced marches through barren terrain and cities left stripped of resources, ahead of their supply lines and the limited supply buffer they'd planned, resulted in failure. 200,000 troops, about 1/3 of his total, died from starvation or froze to death, far more than actually died in battle. Napoleon won
or fought the Russians to a standstill in pretty much all battles, yet lost the war for lack of supply and other planning for the rigors of campaigning in that area of the world.
A few thousand years of military history offer plenty of examples of what happens when a force fails to consolidate gains and outruns or otherwise has inadequate supply lines. This is has not changed from ancient times through to modern warfare.
Sure, but there are circumstances where logistics become more and more challenging, and I’d say Napoleon’s invasion of Russia was another clear case of such a circumstance. The more challenging the logistics are, the more you have to worry about them. American Marines need to worry about logistics a lot, and Napoleon probably should have worried about logistics more than he did in 1812. But, to pick an example off the top of my head, I don’t think the Finns needed to worry about logistics quite as much in 1940; logistics were the least of their challenges.
Bringing it back to software, I find it interesting that the word "deployment" is used in both contexts. And indeed, there is something to be said about the "logistics of data" - as in, how it flows through your system(s) to combine, in the end, into a rapid fire, accurate, useful, valuable response to specific request.
Notice how many posts it took us to get from shipping containers to relative ancient-modern military strategic considerations. I think we've all been playing too many strategy games.
Nah, I just study history. Many strategy games nearly completely ignore logistics-- certainly any semblance of realistic logistics. Though actually when I do play strategy games I avoid those more realistic simulations: I want to focus on strategy, not the fiddly logistical details that actually make it all possible. Sort of like choosing a higher level programming language where many fiddly details are handled behind the scenes and I rarely have to bother with them.
Or in the case of Napoleon in Russia when they ran out of supplies, it marches on the 200,000 dead corpses of starved and frozen to death soldiers, with some help from countless slaughtered horses killed for food; and boot leather boiled soft enough to chew.
That's what you get when your supply plan buffers 50 days for rapid assault & victory while the enemy retreats, retreats, retreats, scorched-earth all the way. Hell, the Russians torched the entire city of Moscow to deny it to the enemy so it couldn't be used as a stop over on the way to St. Petersburg.
That was pretty much the end of Napoleon's campaign in Russia, and the beginning of his downfall. You don't get 500,000 troops killed in a single campaign and come home to fanfare and accolades.
In fact his failures during that campaign foreshadowed his downfall with a (failed) military coup. This was actually slightly convenient for Napoleon, giving him a reason to get the heck out of Russia before the final end came: extremely fortunate since the very small remaining body of troops retreating back west were not especially happy with their failed commander.
The running, value-producing software that lives in the data-center requires many kinds of resources, like people do. Electricity is the air of the robot armies. The network is something like having legs and eyes. Data is food because it has to be gathered, stored, and moved to the right place at the right time, and it is certainly the hardest thing for a process to get!
On a related note, there is an episode of Connections that ties the width of the early railcar's wheelbase to pretty much all subsequent forms of transportation, ending with modern space exploration capability. Pretty awesome to observe the actual reach of decisions made long ago.
Semi related: “How big should a railroad track be?” Is actually an extremely complicated question that took a long time for us to figure out. It wasn’t unusual for different states to have different gauges, often requiring that passengers change trains at state lines.
Also, Russia to this day uses its own gauge, because they’re terrified of being invaded via rail from Europe.
It reminds me of a quote about warfare: "Amateurs talk about strategy and tactics. Professionals talk about logistics"