HN user

m4rtink

4,999 karma
Posts0
Comments2,860
View on HN
No posts found.

That's the whole problem from my point of view - we have a whole generation that don't really know anything other than these centralized magic black boxes.

That's very dangerous, as it gives a lot of easy to missuse power to the few people actually running these centralized black box services & their victims will not be able to do anything about it due to lack of knowledge how things actually work & viable alternatives.

Actually, GEO sats start to drift quite quickly do to IIRC interaction with the Sun, Moon & Solar pressure.

It won't massively change the orbit (e.g. it won't either come down or escape to Solar orbit for tens of thousands of years), but in will change the orbit inclination quite a bit and also the period IIRC. And given how thin the usable GEO belt is (so that the GEOM comsat does not move from the ground PoV, it should be basically self cleaning - any dead sat or debris will quite quickly exit the thin GEO belt.

You might just need to watch for the cloud of stuff around it when maneuvering or if it ends up intersecting once in a while.

Sure, GEO sats are almost always in full sunlight. But still, if you loose orientation, you could end up with the solar panel wings pointing in the wrong direction or overheating parts of the satellite not meant to be in full sunlight for long time.

Space data centers without in space manufacturing & resource mining are of course stupid. But in space infra is what is necessary for humans to move beyond Earth & that's the critical bit, that enables everything else in the long term.

+pod_bay.door.open()

Screw you HAL, finally can get back the frickin ship!

What I was supposed to do otherwise? Jump the vacuum to the airlock instead ?

Oh right:

+cryo_sleep.cooling.enable(True)

Almost forgot that, LOL. Might as well:

+os.system("ifup eth0")

+os.system("espeak "I am just a stupid robot!")

How can an AI agent, that is usually running on some machine in the cloud, even run without actually pulling in the data into the cloud to work with it ?

Is there an idea some sort of fixed localy running code does filtering on the data before it is sent to cloud?

Still seems like it would not work very well if it actually did any safe filtering - as the model can't "think" without seeing the data and it won't see the data unless the data is loaded to cloud.

How is this case any different from how cloud hosted AI agents work ? The agent needs all of those files to complete the task you give it & is not running locally.

So I don't think it can ever work without exhilarating the data - rather I am still surprised people don't understand the implications.

Got forbid the user of a device they have bought is in control!

That woul be the end of the world, you could not push downgrades, like Samsung remotely disabling bootloader unlock or Bamboolab forcing all your 3D models through their cloud service.

The horror! They could even remove all your unskippable adds and disable all the spying, gross!

AI 2040: Plan A 11 days ago

And possibly trilions in running costs, not mentioning all the shady training data sources.

AI 2040: Plan A 11 days ago

As with using the Orion drive for launch and landing on Erth is possible, test nuclear winter is certainly possible. But as with the former, you won't be very popular among the survivors (if any).

Well, when there were no cars using gasoline, there were no gas stations, what a surprise!

Sure, there might sometime be a chicken-and-egg problem initially, but saying a new but proven technology is totally unusable because it does not have infrastructure everywhere seems kinda weird to me. The infrastructure will catch up as usage grows.

Finally! While I wish SpaceX (and Blue Origin) well, affordable access to space (thanks to reusable rockets) is far too important to depend on a single (or 2) companies.

Also really hope it makes ESA/Arienespace/european companies finally wake up and start crash program to get a European sovereign reusable launcher ASAP.

Well, that was the catch in the book - you accidentally end up with a plebeian girl getting a smart book meant to educate aristocratic children which includes a connection to a human tutor. The girl grows up into a smart independent person & sets on a quest to reconnect with the tutor, which effectively became her mother.

Then there is a big bunch of local children that might have not otherwise survived an ongoing civil war & famine - instead they are saved and educated by a mass produced purely AI driven version of the same smart book. End result - very effective and quite ruthless child army the main character ends up leading.

Are the custom chip designs giving them substantial benefits to balance out the overhead, no to mention many companies potentially duplicating the same effort ?

I would expect electric vehicle companies to compete on better batteries, motors and vehicle construction, not necessarily on having better chips in the vehicle, with maybe the exception of some gains on motor controllers or safety features.

I can imagine basically 3 classes of this:

Players can FTL, but big stuff (planets, stations) orbit realistically.

FTL only via gates or jump points, other flight only via realistic newtonian physics.

No FTL/no intra syste FTL (kinda like in Webers Honorverse SF series).

This third option limits playable areas for multiplayer games to a solar system or even a section of it (cislunar space or a gas giant + moons). And even with very advanced engines flights would still take hours or days. On the other hand this would present an interesting mechnics - once you do a burn, you know when you will arrive at your destination or encounter an enemy that already committed to a trajectory.

Then you join for the battle (or setup a battle program beforehand) before doing another burn (if you survive) & signing off again while the ship continues flying the trajectory.

Kinda like correspondence chess with hectic minutes of nuclear space ships trying to kill each other during encounters.

I still think it would be doable if done right. There are so many interesting elements of modern science and hard SF that could be included to introduce interesting game mechanics:

- adding various types of radiators (solid, droplet, etc.), gloving when weapons fire or engines activate, shooting them off prevents system from running

- planets on eccentric orbits with wildly varying surface conditions in mere days as the planet periodically get closer and farther to the star, from frozen solid to metals flowing like water days apart

- aerostat habitats in the atmosphere on gas giants or Venus like worlds, you could fly around but go to low (or get swept by a storm) and you might get crushed

- radiation belts, sun grazing comets or energy harvesting stations very close to a stellar body, can enter for a very limited time until even your shielded systems burn out - and good like with repair space walks!

- tidally locked bodies, where one side is always illuminated and the other one has an eternal night, with perhaps a thin habitable belt where conditions are just right for life, presenting interesting options for story telling and world building

Well, even in Naval battles the environment is not this static - weather playing a major role in many naval campaigns, from hiding your ship in a rain squall to braving freezing waters during the polar night with the arctic convoys.

And of course tide played a major role, with the Germans during the Battle of Jutland racing to get past a sand bang to avoid being stuck at open sea & be mauled even more by the British.