I've always wished the planet descriptions weren't so dry, the RNG generates some pretty compelling combinations sometimes.
So I spared an hour or so and vibecoded this quick tampermonkey extension for the game. It scrapes the page and assembles a prompt from planet stats for AI image generators.
Try it out if you wanna give it a try, it supports both Gemini and OpenAI API, or you can just copy the prompt it generates and input it manually: https://pastebin.com/tGcJTBQw
Example of what it does - https://imgur.com/uIpthQS
I'd say Gemini(Nano Banana) definitely generates more interesting stuff than gpt-image. It's output is less realistic and it messes up often, but it actually manages to synthesize some pretty neat novel environments.
My prediction: pairing AI with high-resolution satellite data will allow us to find untapped reserves of high-quality timber in the Amazon basin, enriching the local populace and the wise investors who get in to this metatrend on time.
But seriously, this list reads like an eerily accurate vision of the corporate hellscape that is to come.
I can't tell if the author is genuinely naive to the effects of near-perfect surveillance and profiling in the hands of governments and megacorps, or if it's just dressed up in pleasantries because there's money to be made there.
In some materials, this hidden configuration might have interesting properties. It could even be (quasi)stable once the original pattern is suppressed. So, novel superconductors or metamaterials, maybe?
The art doesn't even need more detail or realism, it just needs consistent detail.
Even just a simple palettization/pixelation preprocessing for every sprite could be enough - https://i.imgur.com/oPH7paD.jpg
And making sure light is consistent. In the first picture some objects like banners, notice board, bathtub shaped thing have clear shadow below them, others like fence, green plant, mushroom don't. Buckets have light coming from left side hay stack from right, other from top. I wonder if it would look better if some objects were flipped horizontally.
That's at least 50% of it, if he had given his artists any direction it should be this: Light coming from top left, soft shadows. Give the artists a palette and let them create. When you don't have any limits, your creativity is unbounded, and that can be bad for delivering something on a time budget.
That is a surprisingly good result. It literally changed my perception of those screenshots from "that is some gross and sloppy artwork, I'm keeping my money" to "hey not bad, I might buy that"
That's a motte-and-bailey argument.
The easy to defend idea of CoC is of course that one shouldn't discriminate and act unprofessionally.
But declaring a CoC is just the first step - it needs to be enforced.
And by design the enforcement seems to involve backroom committees and no chance for the accused to even know about specific allegations, let alone defend against them.
It can result in downright kafkaesque situations and unnecessarily forces identity politics into the mix.
University code of conducts have lead to some pretty egregious violations of common sense.
For example, I know of a black girl who was harassed using the code of conduct. Her harasser was white, and constantly complained of being made to feel unsafe—-no actual accusations mind you, just a feeling she had.
Unfortunately making someone feel unsafe is against the university code of conduct, so each incident resulted in a long tribunal before everyone concluded that nothing had happened.
It was a waste of time, and lead to the humiliiation of the black girl, whom was always innocent, but to people who didnt know her only saw that she was being accused of something.
Universiities have more, not less, infrastructure to deal with adjucating code of conducts. They get to do it in person, and have written guidelines of how to conduct enforcement. And yet situations like I described are not uncommon.
How do you think a software group will fare in comparision?
So what you're saying is a racist find a way to abuse the rules and no one thought to change the rules in the face of obvious abuse?
This just reminded us we need to be vigilant for folks abusing the rules, and be prepared to alert them in a public & justified way to prevent CoC Snyder by truly dedicated and despicable racists and sexist. Good moral!
The definition of enforcement is left extremely vague in standard CoC, so let's extrapolate a bit.
Biased enforcement will follow quite naturally.
The generic call for project leadership to resolve all issues will eventually become unworkable for larger projects. This process can be hastened by swamping the project leadership with a large number of reports.
It's likely that a subcommittee or enforcement team will be formed to deal with this at some point. These teams will naturally attract people interested in moderating the project and enforcing CoC. Mostly left-leaning people who are interested in achieving "social justice", I'd expect. Thus, a separate branch of leadership with mostly homogenous viewpoints and considerable power to prosecute is formed.
Unless the enforcement is done by randomized jury, this outcome is almost inevitable.
There should definitely be an additional axis for alternative, non-mainstream sources.
It doesn't really work if you're just lumping everything together as either Democrat or Republican.
You're absolutely right that this is a multi-dimensional issue. We're starting out with just one dimension but want to add more in the future. What are the alternative sources you're thinking of? We're considering how to add libertarian to the mix, but there are other alternatives, of course.
It's a shame that the whole matter of global warming has been politicized and muddied so much by special interests that average person simply doesn't care anymore. That's why it's especially weird to see something like this published in Daily Mail. The positive feedback loop is definitely happening though, but the magnitude and extents of it could still vary.
To be honest, I don't believe we can stop this, not with the current population size and the current economic model.
> It's a shame that the whole matter of global warming has been politicized and muddied so much by special interests that average person simply doesn't care anymore. That's why it's especially weird to see something like this published in Daily Mail.
The Daily Mail profits from sensationalism, it's not really like Fox News in the sense that it only pushes one ideological position. So you'll get it reporting sensationalist articles for and against action on climate change (although historically it has been mostly anti, because it makes a better story for the purposes of invoking outrage).
This article is sensationalist by overplaying what is happening - the 200 figure actually refers to the concentration of methane in the leaks relative to the atmosphere (it doesn't refer to the actual scale of the leaks in any sense).
As an aside, I'm not sure these sensationalist stories about extreme climate change actually help, they're still basically pandering to readers by offering a way to be outraged without having to do anything. The narrative shifts directly from 'this is not a problem, we don't need to do anything' to 'this is the apocalypse, someone should have done something about it, but now it's too late for me to make any difference'. Then perhaps a few years later, when the apocalypse hasn't happened, it can go back to 'we're still okay, the problem was overblown, we don't need to do anything'. And the boring work of actually working out a solution falls by the wayside.
The best thing for Earth and possibly the thing for homo sapiens is a nuclear winter. Hillary will just continue the drone war and statist business as usual. Trump on the other hand, total wild-card, and possibly unwitting instigator of nuclear war. The long term ecological choice is Trump.
Is it a little presumptuous for you to decide, individually and on humanity's behalf, that widespread death and destruction is the best thing for everyone?
The idea that the Universe is this sort of discrete lattice always makes me suspect simulation.
I feel like true reality should not become grainy at some point, it feels like a mark of artificial origin.
I see some hints too, but that's probably because my mind is wired to be programmer :) So it's probably just a sort of confirmation bias.
However seeing the quantum effects like a version of compression, dark matter and dark energy which could be an ugly hack to stabilize matter and galaxies, the fractal self similarity of the universe is just too amusing to ignore :)
Or you can simply think of it as the best "resolution" that macroscopic instruments rooted in 3 dimensions can reduce it to.
I feel like the whole "wave-particle duality" is already a hint that the true nature of reality may always remain outside of our observation capability.
In fact, I think that humans may be a little too biased towards picturing everything as made up of discrete parts ("particles"), and that is probably a byproduct of having visual eyesight. Can't wait to meet alien intelligences with completely different senses and see what they think. :)
QM can be defined by the fact that the reality you observe isn't the "actual reality", but overall it doesn't really matter.
Wave-particle duality is a bit of a misnomer it's more of a pop-sci concept, all particles are defined by their wave function and there isn't a "particle function".
And if you think that we are too biased towards picturing everything made out of discrete parts then you won't like QM which is basically what the Q stands for which is "quanta" as in the smallest clump-thing-w/e of something.
That said most people that choose to study physics in high school (year 10-12) and or college would be exposed to Quantum Field Theory which makes everything make considerably more sense when you start thinking about everything in terms of fields.
Once you no longer think of an electron as an individual thing but as a electron field that extends through all space and the particles are the quanta of that field as in localized excitations of the field which when they reach a certain amplitude or energy level bring forth something we can measure and identify as an electron.
I wish QFT would be popularized more because it would bring so many important concepts into the common sense realm and it would actually be easier to explain things like why do particle accelerators work.
I read about this phenomenon where the two slit experiment could be replicated with drops bouncing on an oil bath (here's an article on it: http://resonance.is/news/quantum-weirdness-replaced-by-class... ). And one about how a higher-dimensional mathematical object could dramatically simplify certain quantum mechanical calculation ( http://www.wired.com/2013/12/amplituhedron-jewel-quantum-phy... ). I realize this proves nothing, but I'm still holding out hope that the randomness we observe is just a result of the interaction that higher dimensional "matter/energy" does when it intersects with our 3/4 dimensions. Maybe god doesn't play dice after all, maybe there's just a whole lot going on before we get to make our measurements.
I think that gets it backwards: the universe can move information around and perform computation because it has a level where things become both discrete and stochastic.
FWIW, my own intuition it precisely the opposite. I would be very surprised if reality turned out to be continuous, because of the resulting infinite information density.
I've had similar issues doing remote work for people who aren't really good at managing projects.
The solution depends on how much you care about the product itself - you can either take charge, send constant emails, follow up on missed promises and generally push people to actually do something, or you can just ride this thing out. The management probably won't improve in time, and the product will most likely crash and burn.
I've "saved" a few projects by doing most of the work myself, but often there are factors beyond your control that will drag the project down anyway. Another thing to think about - even if you build it, your CEO isn't guaranteed to make good use of the product and all your honest effort might be wasted by bad leadership anyway.
I'd say Gemini(Nano Banana) definitely generates more interesting stuff than gpt-image. It's output is less realistic and it messes up often, but it actually manages to synthesize some pretty neat novel environments.