I like to point out that we are blind to the extraordinarily common "colors" of nitrogen gas and water vapor, blindness which is beneficial, because otherwise we'd constantly stumble through fog until we go over the edge of a cliff or get eaten by a tiger.
Yet if some aliens insisted that our planet was a very boring featureless nitrogen-colored ball, we would probably object that their viewing strategy is naive and incorrect.
We are born seeing the universe based on pragmatic decisions about signal versus noise. Our evolved tuning utterly fails in new places, so we should pick new tunings.
gchamonlive · 13h ago
Rather than new tunings, they are more like tune mapping, from one possible set of fitness functions to our set of fitness functions.
And I'd argue that it's all signal. Noise is something very human -- maybe even for carbon-based lifeforms, but I'd be careful to make such broad assertion --, that which is unnecessary or harmful for the needs of the body.
Vox_Leone · 3d ago
To the unaided human eye, space is largely a muted void. Stars pierce through the darkness like pinholes in velvet, but much of the grand tapestry—the swirling colors of nebulae, the fiery birthplaces of stars, the delicate filaments of distant galaxies—are invisible. Our vision, evolved for survival under a sunlit sky and on a green-blue Earth, isn’t tuned to perceive the vast electromagnetic chorus the universe sings in.
Although space may appear subdued to our eyes, that doesn’t mean it isn’t beautiful—only that we need to borrow better eyes to truly see it
godshatter · 2h ago
I've seen the sun described as being really mostly white and that the yellow color we normally see has more to do with what light makes it through the atmosphere rather than the actual color of the star and that the light is a continuous spectrum and so on. But then the sun is also classified as a G-2, a yellow dwarf. Is the color that we see in the sky just coincidentally match the color the star is classified under?
kelseydh · 13h ago
A bit of a tangent, but this excerpt from a Sam Kriss rant about science content on social media always stuck with me:
Those multicoloured nebulae are not real objects, they exist only in fantastic pictures overlaid with Neil deGrasse Tyson’s face and some vague sentiments about how wonderful the universe is when it’s very far away from human life. The images are digitally stitched together, the colours are fake, the shapes are not anything that could actually be seen out the window of your spaceship, a real-life nebula is about as exciting as a damp fog.
Images are not real, whether astronomical or not...the colours coming from your display or printout have a very different spectrum from that of the real object
gchamonlive · 13h ago
The image you see is also very different from the images that hit your eyes, since they travel through complicated mechanisms before being consciously assimilated by the mind, but at this point what are we even talking about...
Etheryte · 14h ago
Ceci n'est pas un commentaire.
gherard5555 · 9h ago
Ceci n'est pas une réponse
kunzhi · 4h ago
Ceci n'est pas une pipe
Simulacra · 12h ago
I understand the necessity to approximate colors, but this article doesn't really explain the answer to the question. How much of the colors are simulated?
TheBigSalad · 47m ago
Just assume everything in space is invisible to your eyes because they are tuned for Earth only.
AStonesThrow · 5h ago
The article attempts to explain the difference between "colors" perceived by human eye, vs. wavelengths and signals received by artificial and scientific instruments.
"How much" isn't a question that makes sense, unless you are looking at a specific set of images. Astronomical images can be generated in many, many ways from all sorts of processes. Sometimes there is natural color included, but enhanced; sometimes a wider spectrum is mapped into visible colors so humans can appreciate it; sometimes, the colors are codes for different features, and there is a key provided, so that the colors can be interpreted to what they mean, rather than corresponding to visible light.
It's not just astronomical images, but all disciplines of science use color keys to depict things that are not colors. Look at a topographical map, and it may have green and brown areas that are simply abstracts. A political map uses 5-6 colors to highlight the shapes states or nations, but these are not "real" colors either. But usually, you're provided a color key, for interpretation.
That sounds a lot like dodging the question. Obviously correcting for things like red shifts, camera limitations etc is fine. But as far as I know there are many instances where the coloring is pushed beyond anything remotely realistic. I think the pillars of creation are a popular example
creatonez · 13h ago
Keep in mind that Pillars of Creation is not supposed to be intentionally beautiful. It's a false color image for maximizing research value. Green for hydrogen band, red for sulfur band, blue for oxygen band. The goal of the processing is to minimize the error in these three readings, to study the ionized gas in the cloud as it truly is. If it were engineered for the public to look in awe, you could probably come up with an even more striking coloring than a simple RGB assignment. This is something a lot of people get wrong about the topic of false color images -- usually, there is not a separate "for the public" and "for researchers" image. The media outlets just pulled the coolest looking processed images out of actual research papers, and some of them were so cool looking they became cultural icons. (The exceptions to this, ironically, are when an image is intentionally processed to be as close to true color as possible -- which makes it less useful for research purposes, but fantastic for writing explainers for laymen on the topic of false color images.)
Though, you could say it's indirectly designed for beauty. When astronomers process an image, they are looking for interesting features to study. And once you've narrowed in on the most scientifically interesting way to process an image, the sorts of striking color contrasts that creates are inherently beautiful because they are unveiling the interesting parts & letting us see the unseen for the first time. And there is selection bias on what is interesting to look at. A violent explosion is both more aesthetically inspiring and teaches us more about how matter works than a picture of a random stairwell, and astronomers love their violent explosions.
DiogenesKynikos · 12h ago
Astronomers rarely look at color images. These color images are produced by press offices for the public.
That being said, the complaint that the images are "unrealistic" are off-base. The images accurately portray the information that was captured by the telescope. Your eyes often can't even see the wavelengths imaged by space telescopes (like the James Webb Space Telescope), so how you map those wavelengths to RGB is largely arbitrary. As long as you map them in the same order (shorter wavelengths appear bluer, longer wavelengths appear redder), I would consider the images accurate.
xioxox · 2h ago
That's not right. I'm an astronomer, and I often look at colour images and colleagues do, too. For example, the galaxies in a cluster of galaxies follow a relation of colours - brightness (the "red sequence"), which can be used to detect the cluster. The eye is also quite good at helping confirm a cluster by spotting the galaxies following this sequence. I also use colour images to help identify spectral changes that change across an image, in my case in the X-ray waveband.
creatonez · 12h ago
This is just not true, though. Astronomers look at color images all the time, in certain areas of research at least. Many color images in the media are copied directly from research papers, since image processing can be rather labor intense to be done just for PR purposes (mainly because you have to essentially calibrate the positioning and noise characteristics of the sensors from scratch -- the color assignment part can actually be done in vanilla photoshop). This is something I feel the article skipped out on explaining, that these false color images are actually research tools the same way that the grayscale images are.
It is true that they also look at a lot of grayscale images, though. Differentiating chemicals in nebulas or on the surface of planets isn't always necessary, and you can always take the simpler (but sometimes less useful, since it doesn't make full use of the human visual system) route of looking at multiple grayscale images. A lot of the areas where false color is most useful are well-trodded ground, so we don't necessarily need more images of the same objects when the existing body of research is already rich. And sometimes, a researcher includes an image in their paper just because they thought it looked cool, not because it was useful for the methodology.
on_the_train · 12h ago
Color Images imply that they are real colors. If they are not, that's misleading.
IAmBroom · 8h ago
Nonsense.
A screenshot of this page would capture a title bar in orange.
The REAL title bar is displayed as dots of green, blue, and red; no orange is present.
Yet if some aliens insisted that our planet was a very boring featureless nitrogen-colored ball, we would probably object that their viewing strategy is naive and incorrect.
We are born seeing the universe based on pragmatic decisions about signal versus noise. Our evolved tuning utterly fails in new places, so we should pick new tunings.
And I'd argue that it's all signal. Noise is something very human -- maybe even for carbon-based lifeforms, but I'd be careful to make such broad assertion --, that which is unnecessary or harmful for the needs of the body.
Although space may appear subdued to our eyes, that doesn’t mean it isn’t beautiful—only that we need to borrow better eyes to truly see it
"How much" isn't a question that makes sense, unless you are looking at a specific set of images. Astronomical images can be generated in many, many ways from all sorts of processes. Sometimes there is natural color included, but enhanced; sometimes a wider spectrum is mapped into visible colors so humans can appreciate it; sometimes, the colors are codes for different features, and there is a key provided, so that the colors can be interpreted to what they mean, rather than corresponding to visible light.
It's not just astronomical images, but all disciplines of science use color keys to depict things that are not colors. Look at a topographical map, and it may have green and brown areas that are simply abstracts. A political map uses 5-6 colors to highlight the shapes states or nations, but these are not "real" colors either. But usually, you're provided a color key, for interpretation.
Though, you could say it's indirectly designed for beauty. When astronomers process an image, they are looking for interesting features to study. And once you've narrowed in on the most scientifically interesting way to process an image, the sorts of striking color contrasts that creates are inherently beautiful because they are unveiling the interesting parts & letting us see the unseen for the first time. And there is selection bias on what is interesting to look at. A violent explosion is both more aesthetically inspiring and teaches us more about how matter works than a picture of a random stairwell, and astronomers love their violent explosions.
That being said, the complaint that the images are "unrealistic" are off-base. The images accurately portray the information that was captured by the telescope. Your eyes often can't even see the wavelengths imaged by space telescopes (like the James Webb Space Telescope), so how you map those wavelengths to RGB is largely arbitrary. As long as you map them in the same order (shorter wavelengths appear bluer, longer wavelengths appear redder), I would consider the images accurate.
It is true that they also look at a lot of grayscale images, though. Differentiating chemicals in nebulas or on the surface of planets isn't always necessary, and you can always take the simpler (but sometimes less useful, since it doesn't make full use of the human visual system) route of looking at multiple grayscale images. A lot of the areas where false color is most useful are well-trodded ground, so we don't necessarily need more images of the same objects when the existing body of research is already rich. And sometimes, a researcher includes an image in their paper just because they thought it looked cool, not because it was useful for the methodology.
A screenshot of this page would capture a title bar in orange.
The REAL title bar is displayed as dots of green, blue, and red; no orange is present.