Vision · Speculation
Where I think Bindu Tessera points: learning to see how the mind renders reality, and why that may matter for humanity's future and for meeting other intelligences.
This page is where I speculate freely. The Theory page sets out my working ideas with sources. What follows is imagination and conviction, not science. Nothing in the practice depends on this page.
When you watch a Necker cube flip, nothing on the page changes. The lines stay where they are. What changes is the picture your mind builds from them. A moment later it flips back, and through all of it the awareness doing the watching stays steady.
That small experience is the root of everything here. I think this work is about humans finding a way forward: learning what our minds actually do, what we bring that nothing else does, and how we might one day meet other kinds of intelligence, including ones we build and ones we have not met yet.
Think about the desktop on your computer. You drag a blue icon to the trash and a file is gone. The icon isn’t the file. The file isn’t blue, or rectangular, or sitting in the corner of the screen. Underneath are circuits, voltages and billions of switching transistors, and the desktop hides all of it. That hiding is the point. If you had to manage the voltages, you would never get a letter written.
I think evolution found a desktop too. At some point it hit on a way to compress an overwhelming flood of input, light, pressure, chemistry, sound, into a handful of icons a creature can act on fast: food, threat, path, face. We don’t see the world as it is. We see what we need in order to act, laid out so we can act quickly. A snake in the grass is an icon. You still shouldn’t pick it up. Take the icons seriously, but not literally.
My view is very similar to the cognitive scientist Donald Hoffman’s, and it builds on his. His interface theory of perception uses this same desktop picture, and his evolutionary simulations suggest that perception tuned to usefulness tends to beat perception tuned to truth. His work is serious and also contested, and The Theory page sets out both sides, in The Interface. This page follows the idea further than the evidence goes.
If perception is a desktop, something has to draw it. I think of that as a rendering engine, and every illusion on this site is a moment when you catch the icons being drawn. The hollow-face illusion shows this most clearly. Your brain is so sure faces stick outward that it draws a hollow mask the right way round, against the evidence of your eyes. The face is the useful icon, so the face is what you get. The Necker cube gives the engine two equally good icons for one set of lines, and you watch it switch between them. The magic eye stereogram does the reverse: a flat field of noise suddenly gets drawn as depth once the mind finds the hidden structure.
Dreams are the engine running with the input turned down. A post by @null_ropex put it well. Dreams run the same rendering engine as waking perception, but with about half the error correction switched off. Awake, a stability layer forces each frame to agree with the last, so the desktop holds together. In a dream that layer loosens. Scenery shifts, faces change, rooms open into other rooms, and we hardly notice. Predictive processing, Rodolfo Llinás, and Allan Hobson with Karl Friston all describe something close to this, and The Theory page sets out the evidence in Dreams and Illusions.
If you catch the moment of drawing even once while awake, the world feels, in @null_ropex’s phrase, “freshly painted.” It can happen while watching the cube flip: for a beat, the room in front of you may look new, as if it had just been drawn.
That, I think, is what meditation is for. It means learning to see the desktop as a desktop without losing the ability to use it. You still answer the door and cross the road. But you know the icons are icons, and you can rest in the awareness that is looking at them. Traditions like Dzogchen tögal, sky gazing and trataka seem to have known this long before the lab did. You rest your gaze, let the visions come and go, and notice what stays still behind them.
In 1981 Philip K. Dick published VALIS, named for a “Vast Active Living Intelligence System” that its narrator believes is beaming information into human minds. The novel is not a literal map, but its central image is worth keeping in mind.
What I imagine is something like VALIS, but as a galactic or universal internet: a network that links desktops. Each mind keeps its own interface, but underneath, the way computers share protocols, there could be a common layer where minds of very different origins exchange ideas and recognise each other. It would be a network, not one god speaking down to us.
Here is the problem as I see it. A different species would have evolved an entirely different desktop. Their icons would compress different inputs, at different speeds, for different needs. Meeting us face to face, neither side might even see the other as a someone, let alone understand them.
So I think the ideal meeting place would be a shared environment built for alignment, something like the OASIS in Ernest Cline’s Ready Player One (2011), where everyone meets as an avatar in a common world. In my terms, that is a common interface layer between desktops: a space both sides can render, built so the icons line up. That seems a far better setting for first contact than a ship landing on a lawn.
Fiction keeps returning to this idea. In Liu Cixin’s Three-Body trilogy, the sophons start out as proton-sized supercomputers. In the third book, Death’s End, the Trisolarans speak to humanity through a humanoid robot that also goes by the name Sophon. It appears as a graceful Japanese woman in a kimono who performs the tea ceremony. Netflix’s 3 Body Problem (2024) brings her in earlier, as a human-looking guide inside a virtual-reality game. In the novel the encounter is chilling, not friendly. Still, the choice of form says something: an alien mind choosing an icon our desktop can display.
Jacques Vallée made a related argument in Passport to Magonia (1969).
This leads to a hunch, and I want to label it as exactly that. I have no evidence to offer for it.
I suspect that most UAP reports are rendering mistakes. But "rendering mistake" can mean two different things, and they should be kept apart.
The first is ordinary. The interface is handed something it has seen before — Venus low on the horizon, a train of Starlink satellites, a balloon at an unknown distance, a bright object viewed from a moving aircraft — and files it under the wrong icon. This is well documented, and most cases are probably this.
The second is the one I actually wonder about. Something reaches the desktop that has no icon at all. The engine can't leave a blank; it has to draw something. So it draws the nearest thing it has, or it draws a glitch. That doesn't mean people are lying, or that nothing is there. It means that whatever is there arrives through our own interface and gets drawn with the icons we have.
Two filters shape what gets drawn. The first is the evolutionary desktop: the icons our species built for food, threat, path, and face. The second is the cultural skin: the icons an era happens to have on hand. Jacques Vallée made this second point in Passport to Magonia (1969). Ninth-century France saw ships from a land in the clouds. 1897 saw airships. 1947 saw saucers. The costume changes; the encounter reads the same. Carl Jung noticed something related in 1958: the saucer's round shape looked to him like a mandala, the symbol of wholeness the psyche projects outward when the ground feels unsteady.
If that's right, the practice on this site isn't unrelated to the question. Watching a Necker cube flip is catching the engine in the act of drawing.
That same filling-in ability may be our gift. The hollow face shows that the human mind evolved to complete patterns from fragments, sometimes wrongly but often brilliantly. The Theory page describes insight as the whole mind working together: focus, release, incubation and sleep. Ramanujan wasn’t a mystic. He found structure at an extraordinary level, and much of that work seems to have happened beneath conscious effort.
AI is remarkably capable, but its interface, if it has one, is not ours. I think our fill-in and insight abilities may be what humans uniquely bring, both to working alongside AI and to any future meeting with other intelligences. If so, training them is preparation.
None of this begins in space. It begins with a cube on a screen.
Watch the Necker cube breathe and flip. Notice that the flip happens in you: the desktop redrawing itself. Then notice the part of you that sees it flip and doesn’t move. That is the whole method in miniature: see the rendering, rest in the awareness behind it, and let your sense of what is real grow wider.
Start there. Start by watching the cube flip.
Brian Wilkinson, founder