Tag: manyblinkinglights

Humans Are Weird




So there has been a bit of “what if humans were the weird ones?” going around tumblr at the moment and Earth Day got me thinking. Earth is a wonky place, the axis tilts, the orbit wobbles, and the ground spews molten rock for goodness sakes. What if what makes humans weird is just our capacity to survive? What if all the other life bearing planets are these mild, Mediterranean climates with no seasons, no tectonic plates, and no intense weather? 

What if several species (including humans) land on a world and the humans are all “SCORE! Earth like world! Let’s get exploring before we get out competed!” And the planet starts offing the other aliens right and left, electric storms, hypothermia, tornadoes and the humans are just … there… counting seconds between flashes, having snowball fights, and just surviving. 

To paraphrase one of my favorite bits of a ‘humans are awesome’ fiction megapost: “you don’t know you’re from a Death World until you leave it.” For a ton of reasons, I really like the idea of Earth being Space Australia.

Earth being Space Australia

Words cannot express how much I love these posts







I love Temeraire but man, I really want to see how that world continues to develop with dragons. What happens when people start building airplanes? Do they even bother to build planes when they can just hop on a dragon? How do dragons influence the development of the American military-industrial complex? What do dragons think of the Internet? Is Temeraire still alive and posting selfies in his best jewelry please I need to know

Okay but am I the only one who has imagined dragons *in planes*. Dragons want to go over the sea too, and ships are awful and take months to get anywhere. Planes with a section in the back for one or two little courier-weights. Planes with reinforced sections for baby dragons. Giant, giant carrier-planes for larger dragons.

Dragons hate planes. It is a Fact. They whine so, so much about flying without wings.

(Dragons going through Customs, guys.)

Whoa, you’re right–the most sensible solution for all concerned would be to have shipping switch over to airmailing small dragons, who could finish many of the final hops sending people/goods to last-stop distribution centers.

what about humans in hangliders and squirrel suits? a talented hanglider guy can coast around for ages. then a small dragon could tow them higher, or a middleweight dragon could simply be landed on. the humans would have to get the hang of dealing with the turbulence from enormous flapping wings, but if they wiped out the dragon could just grab them. 

the tactical advantages in dragon combat are obvious: instead of forcing the enormous and expensive dragons into direct combat, their boarding teams would glide at each other from a distance. if they miss, they just spiral down rather than falling to their death, unless encountered by an enemy glider with a big ripping hook or swatted by the enemy dragon. the human combatants can then continue the fight on the ground, or sit tight and radio for extraction. 

i’d equip a human boarder with a pistol, sticky explosives, a compass, a map of the area, and a big sickle-bladed grappling hook on a chain, for either grabbing on to a dragon’s hide or slicing through an enemy’s glider fabric. the trick would be to make something that caught in hide but went right through a glider, since you’d want to stick to the dragon, but not a plummeting enemy. i suppose it could be detachable at the chain…. then again, if you have a pistol, you might just want to shoot your gliding enemies directly. 

the explosives could be as simple as tape and a stick of dynamite, if it’s early 19-20 stuff. glider over, hook on wherever you can, tape the dynamite down while shooting whoever you can, and dive away before it blows. later plastic explosives could be made even stickier and harder to peel off, with remote detonators, so you could just brush by, throw it down, fly off, and then blow it. 






dragons have lips

this isn’t a shitpost. in skyrim’s dragon language there are pairs of distinct words such as: ‘nid’ (no/none), ‘mid’ (loyal/loyalty), ‘mu’ (we) vs ‘nu’ (now), and ‘aan’ (an) vs ‘aam’ (to serve). this indicates that dragons are able to distinguish between the sounds ‘m’ and ‘n’. 

the only difference between ‘m’ and ‘n’ is that the latter is formed with the tongue, while the former is produced with the lips 

therefore, dragons have lips 

This is some solid linguistic field work and I look forward to the followup post.

what happens if the robots achieve sentience and then need rights of their own?







We’ll build them (or have them build each other, eventually) to not do that.

Which I don’t think will be difficult.  A robot isn’t an organism with an unfathomably complex meat brain, it’s got a processor where all of the processes can be viewed and analyzed.  If you look at the log and you don’t see an “experience consciousness, desire rights” process, that means it doesn’t have one.

That’s… really not how AI or coding in general works. 

Example: a major research problem in machine learning is the interpretability problem, which essentially boils down to: “How the fuck do deep nets work?” Like, we actually don’t know. And that’s for relatively simple programs. For anything complicated enough that we’re wondering “Is this thing conscious?”, we are almost certainly not going to know how it works.

Okay, fair, point, and I have to admit my knowledge of CS doesn’t go far beyond “computers do math!”

However, I’d counter that we could eliminate a lot of human labor, and the most unpleasant kinds, with relatively simple machines.  A therapist or general-manager robot might have to be so complex that it risks consciousness; a fruit-picker or bathroom-scrubber robot would not.

So fine, we’ll have some humans (who work obscenely short hours under delightfully comfortable conditions) do the high-level management of a farm or factory, but there’s no reason to have humans out in the fields getting up at 3 AM and wrecking their backs to execute the management decisions.  This is obviously already happening in a lot of industries anyway, but what I’d like to see is a society aimed at doing it for “nobody needs to wreck their back” reasons instead of “robots don’t ask for money” reasons.  (And a society where the resulting unemployment is rightly celebrated as the advancement of the human race into a life of leisure and freedom.)

This is getting more towards semi-automated luxury socialism, but it’s better than what we’ve got now, and it’s achievable with mostly extant technology.

Alternately, we could just decide that we don’t care if robots are sentient, we’re the fucking humans here.  We already do that for animals, and, well… I’m not a vegan.

I feel so bad about those last two sentences.  Like, I fully realize you could use that reasoning to justify human slavery too.

But there’s no escape from “I only care about things that are a certain amount like me.” I’m not going to shut down my immune system to honor bacteria rights.  And I really doubt that I’d change my mind if you proved bacteria were sentient. Maybe it would be better if I just admit that I’m doing it because I’m me and the bacteria aren’t me, instead of coming up with some convoluted reasoning why my survival serves The Ultimate Good.

“I’m not trying to maximize the total happiness of all beings, I’m just cheering for my team here" is sort of the root of all evil?  And sort of necessary to exist?

…I did not expect a post about my “robots do my job for me” fantasies to go this deep.

I read a Young Wizards book where some robots were affronted and disturbed by humans having things like… microwaves, and calculators, because all programmed machines were Enslaved Quicklife or something. That… Science fiction is absolutely full of sentient thinking robots oppressed by their creators, but it seems like there’s quite a slope between those and the commercially available machines we have today. Microwaves and calculators have no autonomy. They are glorified levers in that regard.

I’ve got a Star Wars novel where an R2D2-style droid going into an elevator in a hotel talks to the elevator AI, which used to be on a warship, and this elevator trashtalks the hotel’s other, snootier hotels and has long boring stories it wants to tell about the battles it’s been in. That ain’t right. Who’d harness something that intelligent to run a single elevator in a hotel? It’s an elevator!

Is something built to have more autonomy than a calculator but the kind of limited capacities suited for the relatively simple tasks they are set to – a Roomba, that one robot that brings room service, ELIZA the conversation chatbot – enslaved? I don’t think so, but what if it’s considerably more complex than that? A fruit picker, provided it wasn’t like that one robot that spreads these umbrella wings under a tree and shakes the trunk so ripe fruit falls and is caught, would have to be very maneuverable and able to get around obstacles and make fine judgments.

I can imagine robots themselves deciding the metric is, “If ten humans, in ten years, can comprehend it completely, it isn’t High AI.”

And then you could have a book or an anime or what have you about such a team of ten (a reasonably viable career-path, presumably), and the system(s) they were called in to evaluate.

Like jury nullification, the humans refusing to report complete understanding would let a lot of sociable edge cases (by the standards of other High AI) slip by and be certified as beings. Meanwhile if it were something that would be direly inconvenient for everybody if it turned out to be a person (for instance, something that was very big, very important, and/or had done a ton of murders), the humans would do their very best to comprehend something considered well beyond their faculties, to the point where perhaps existing, allied High AI could tell each other ghost stories and feel genteely, evening-party “threatened” about nascent mammalian capabilities, when whatever murderous edifice it was got dismantled afterwards.