Does memory leak? (1995)

> CIA drone strikes all over the world

The US is known to have carried out drone strikes in Afghanistan, Yemen (including against US citizens), Pakistan, Libya, and Somalia; authority over the assassination program was officially transferred from the CIA to the military by Obama. That leaves another 200-plus countries whose citizens do not yet know the feeling of helpless terror when the car in front of you on the highway explodes into a fireball unexpectedly, presaged only by the far-off ripping sound of a Reaper on the horizon, just like most days. The smaller drones that make this tactic affordable to a wider range of groups will give no such warning.

> It’s much easier to defend soldiers against tiny drones than against heavy fire.

Daesh used tiny drones against soldiers with some effectiveness, but there are several major differences between autonomous drones and heavy fire. First, heavy fire is expensive, requiring either heavy weapons or a large number of small arms. Second, autonomous drones (which Daesh evidently did not have) can travel a lot farther than heavy fire; the attacker can target the soldiers’ families in another city rather than the soldiers themselves, and even if they are targeting the soldiers directly, they do not need to expose themselves to counterattack from the soldiers. Third, almost all bullets miss, but autonomous drones hardly ever need to miss; like a sniper, they can plan for one shot, one kill.

You may be thinking of the 5 m/s quadcopters shown in the Slaughterbots video, but there’s no reason for drones to move that slowly. Slingshot stones, arrows from bows, and bottle-rockets all move on the order of 100 m/s, and you can stick guidance canards on any of them, VAPP-style.

> If you don’t know who your enemy is, say terrorists mixed in the crowd of civilians, how would face detection help you?

Yes, it’s true that if your enemy is protected by anonymity, face-recognition drones are less than useful — that’s why the first step in my scenario is the end of any government transparency, because the only people who can govern in that scenario (in the Westphalian sense of applying deadly force with impunity) are anonymous terrorists. But if the terrorists know who their victims are, the victims cannot protect themselves by mixing into a crowd of civilians.

> Yes, terrorists and evil governments will keep murdering people just like they always have. No, this technology does not make it fundamentally easier.

Well, on the battlefield it definitely will drive down the cost per kill, even though it hasn’t yet. It’s plausible to think that it will drive down the cost per kill in scenarios of mass political killing, as I described above, but you might be right that it won’t.

The two really big changes, though, are not about making killing easier, but about making killing more persuasive, for two reasons. ① It allows the killing to be precisely focused on the desired target, for example enabling armies to kill only the officers of the opposing forces, only the men in a city, or only the workers at a munitions plant, rather than everybody within three kilometers; ② it allows the killing to be truly borderless, so that it’s very nearly as easy to kill the officers’ families as to kill the officers — but only the officers who refuse to surrender.

You say “evil governments”, but killing people to break their will to continue to struggle is not limited to some subset of governments; it is the fundamental way that governments retain power in the face of the threat of invasion.

Covering a city with nets is surprisingly practical, given modern materials like Dyneema and Zylon, but not effective against all kinds of drones. I agree that underground fortresses and off-planet camps cannot save very many people, but perhaps they can preserve some seed of human civilization.

> [1] https://doc.rust-lang.org/std/slice/fn.from_raw_parts.html

Thanks.

> having Option be the same size as &[T]

That is literally what I mentioned as a possible reason ("except possibly ..."), but what I overlooked was that you could take a mutable reference to the &[T] inside a Option, then store a valid &[T] into it - if NULL is allowed, you effectively mutated the discriminant of a enum when you have active references to its fields, violating some aspect of type/memory safety, even I'm not sure which.

> C generally chooses NULL, Rust happens to choose a different value.

It's not about what pointer value the langauge chooses when it's asked to create a zero-length slice, it's about whether the language accepts a NULL pointer in a zero-length slice it finds lying around somewhere.

In addition to what the other commenters pointed out about cat being able to concatenate, even using cat as the identity function is useful. Just as the number zero is useful.

Possibly the Ariane 5 failure?

https://en.m.wikipedia.org/wiki/Cluster_(spacecraft)#Launch_...

Solve the problem you have now, not the problem you may not have later. You can worry about that when the time comes, if it ever comes.

In the case of compiler, one solution would be to replace all calls to `malloc` with something like `ccalloc` that simply returns pieces of a `realloc`'d buffer which is freed after the in-IDE compiler has finished compiling.