Ken's Blog

In the future, we will have the option to wear unobtrusive blended reality glasses. As a user interface, these glasses might largely replace our phones, which will stay in our pockets, mainly functioning to help our glasses with computation and communication with the Cloud.

Those glasses will be equipped with cameras and microphones, and will use A.I. to help us navigate the world. Which means that they will serve as companions in helping us to build the latent spaces that will turn into useful memories.

As has always been the case, we will become intertwined with our technology. In essense, the technology will become part of our intellectual, emotional and cultural identity.

Once upon a time we mainly did this through books. Then we incorporated movies, then television, then the internet, and more recently social media.

When we speak to one another in conversation, the things we remember, the associations that we make, will be subtly reinforced by information that we are getting in real time from our own latent space, via the A.I. in our glasses, in communication with the Cloud.

This won’t just be information about the world around us, but also about ourselves. The A.I. will be have been trained on all of our past conversations, our life experience, our emotional reactions to everything and everyone we have ever encountered.

To children who are born into a world where people have this capability none of his will feel like technology, any more than it feels like “technology” when you read a favorite passage from a cherished book, or can see because you are wearing your prescription glasses. It will just feel normal. It will just feel like us, being ourselves.

Once you start to see human perception and interaction in terms of latent spaces, the entire concept of memory takes on a different meaning. It would be easy to think of a memory as some kind of internal recording, like a tape in a tape recorder, or a strip of movie film. But it would also be wrong.

Memories are not literal at all. They are reconstructed from patterns of thought that we have spent our entire lives building.

Your memory of a person’s face is not a literal recording of their face, but rather an impression that you carry with you, which relates their face to all of the other faces that you have seen in your life. In this sense, there is no such thing as a “memory of someone’s face” in isolation from the rest of our lived experience.

This way of looking at things makes it easier to understand how our conversations fall into patterns. When we talk with another person, most of what we say is framing, using our memory of conversations past to scaffold new thoughts.

That’s not a bad thing. In fact, we can better understand the meaning and import of a conversation by how it uses those conversational memories as a launching point to travel into new territory.

Thinking in terms of latent space helps to understand the human condition. When we see the face of another person, we never see just that face in isolation. Instead, we see how that particular face relates to all of the faces that we have ever seen.

We don’t do this consciously — it’s something that happens in our brain far beneath the level of consciousness. And the same thing happens when we see a house, or a tree, or hear a guitar playing.

All of our lives, up until that moment, our brains have been busy classifying everything, and taking into account all of the variations within any given category. That range of perceptible variation is our internal latent space.

Similarly, when you have a conversation with somebody about romance, or politics, or sports, or movies, some topic of shared expertise, you are both exploring your shared latent space. There are unspoken rules for a discussion of any such topic, and we all know those rules so well that we don’t even think about them.

If you were having a heart to heart talk with a friend about relationships, and that person suddenly started talking instead about stamp collecting, you would be right to be worried. This is because the parameters of a conversation about relationships are generally unspoken, but they are also generally well known.

When it comes right down to it, every human dwells within the latent space that they have spent their life constructing. The external world is really only raw material for the construction of that latent space.

And we we spend much of our time discovering the ways that our respective latent spaces fit together. From those shared connections, we create meaning in our lives.

The latent space of any collection of objects is essentially the behind-the-scenes set of control knobs that you could adjust to create any given object in the collection. For example, consider all of the chairs in the world.

You might imagine a magic chair factory with a showroom containing a single generic chair, and a control panel that contains hundreds of knobs. You can’t manufacture your dream chair, but you can pick up a pencil and make a sketch of it.

When you enter the factory, the factory operator looks at your sketch and starts turning various knobs right or left. Every time a knob is adjusted, the shape of the chair changes.

If the factory operator is really good, after a while the chair in the showroom starts to look a lot like the chair in your sketch. When the process is complete, that generic chair has turned into your dream chair.

Basically, the knobs in that control panel represent the latent space of all possible chairs. And any particular chair corresponds to a particular setting of those knobs.

More tomorrow.

I have been thinking recently about latent space. It is the key concept behind modern A.I., but I think it is much more than that.

If you look at latent space from a philosophical perspective, there is so much there to unpack. More tomorrow.

I finally forced myself to watch the press conference with President Zelenskyy from the White House. It wasn’t easy viewing.

I was quite impressed by the dignity and restraint shown by President Zelenskyy, considering the circus he found himself in. It was as though the two other people in the conversation were trying to convince all of us that they are complete idiots.

Or maybe they really are complete idiots. It’s possible that they simply don’t understand how high the stakes are for our own country.

The U.S. hasn’t been supplying arms to Ukraine out of generosity. We’ve been supplying arms to Ukraine because if Russia actually gets away with destroying a sovereign European democracy on our watch, the negative consequences for the U.S., both political and economic, are profound, and potentially devastating.

It’s scary to realize that we have a president who doesn’t seem to know that. Or maybe our president, and his little pet attack dog, just have a deep hatred and contempt for the United States of America.

Has anyone else noticed how quickly the Whitehouse has declared the U.S. to be one of the bad guys? Europe is scrambling to reorder its priorities, because now it’s pretty much alone in standing up for decency and in maintaining a moral compass.

It’s as though our Federal government just flipped that old Google directive on its head, the one that went “Don’t be evil”. It seems as though our government’s new motto is “Be evil”.

I realized, after certain political events of this past week, that one of the most patriotic things I can do is learn another language.

After all, the true strength of our nation has been our diversity. We are the great melting pot, the place where diverse cultures mingle to create something beautiful.

So if some idiot tries to attack our country’s greatest strength, we need to defend it. And what better way to do that than by learning another language.

So be a true patriot today. Take a Spanish class.

Suppose we ask the same questions that we asked yesterday, but we use only audio to annotate the objects in our daily lives. Will that be useful?

Suppose, for example, when we walk into a grocery store, we can choose to hear other people’s ratings of each product, with some sort of audio indication of upvotes and downvotes.

When we go to a museum or art gallery, we can choose to hear the thoughts of other art lovers who have been there before us. For example, suppose I want to know what works of art in some museum were of particular interest to computer science professors. And then will I be able to listen for directions guiding me to those pieces?

And maybe, after I have brewed a fresh cup of tea, my cup can tell me, at just the right moment, when the tea has cooled enough to drink. That would be nice.

In the future, when we’ve all replaced our smartphones by those future blended reality glasses, will we annotate the objects in our daily lives? And if so, how and why?

When we walk into a grocery store will we see other people’s ratings of each product, with some sort of visual indication of upvotes and downvotes?

Similarly, when we go to a museum or art gallery, will we see annotations telling us which art other people liked? And will we be able to filter those results?

For example, suppose I want to know what works of art in some museum were of particular interest to computer science professors. Will I be able to be directed to those pieces?

Will there be a rating of the quality our tea? Or perhaps there will be a bit of text floating over the cup to tell us when it is just the right temperature to drink.

The possibilities are endless.