Author: cgerrish

Unemployed philosopher

Symmetric Difference: Human and Machine Intelligence

Published January 2, 2010 by cgerrish

As we move into the teens, the decade of 2010, the speed of change will continue accelerate. With the advent of the Android operating system, cloud computing, advances in nanotechnologies and a resurgence in the field of artificial intelligence — all the ingredients are assembled to establish the primitives of what some call the singularity. That moment when intelligence becomes detached from humanity and operates independently and at vast scale. Of course, none of this may happen. It’s a possible future that was imagined in the 1950s and 60s. And although its characteristics were generally dystopian, it’s a future with an irresistible pull.

The hidden question in all of our work on injecting a form of intelligence into our algorithms is what can’t be virtualized? What is the symmetric difference, or the exclusive disjunction, between human and machine intelligence? Will these two sets become identical at some point, or will there always be a remainder? And if there is a remainder, what is its importance and value?

Philip K. Dick’s last interview was transcribed and published in a book called: “What If Our World Is Their Heaven.” The interview took place shortly before the premiere of the film Blade Runner, which was based on his novella “Do Androids Dream of Electric Sheep.” Dick spent a lot of time thinking, and exploring through his fiction, where the boundary lies between human and machine intelligence:

You notice an actual change in her thoughts. They begin to get rigid, brittle, and cold, and mechanical. And she turns into that android figure which is my metaphor for the dehumanized person, as you know, who is someone who is less than human—that essential quality that distinguishes a human being is essentially compassion or kindness, that—it’s not intelligence. An android—or in the film Blade Runner it’s called “replicant“—can be very intelligent, but it’s not really human. Because it’s not intelligence that makes a human being; in my opinion it’s the quality of kindness or compassion or whatever—who know, the Christians call it “agape.”

Minimizing the difference between human and machine intelligence is at first a competitive advantage for the businesses that build and sell machines. Later, discovering that difference becomes an essential imperative.

The gender of machines, and especially when we project gender on to machines, provides a lucid visualization of our fantasies and desires. The machine is a blank slate, the android is a particular kind of fantasy expressed through design, engineering and manufacturing.

There’s a moment when the line blurs, where the humanity of the replicant surpasses that of the human charged with hunting it down and terminating it. It’s just that blurriness, the absence of the remainder that seems to threaten humanity.

Of course this kind of narrative is all speculative fiction, the stuff of comic books and graphic novels. Isn’t it?

One Comment

Our Blakean Year

Published December 26, 2009 by cgerrish

William Blake

The Winter Solstice brings us the day with most darkness and the least amount of light. As that moment passes, the days begin to grow longer again. There are many markers for the end of the year, but this one seems the most significant to me. Sometimes called midwinter, it’s the moment when Autumn and Winter touch and the momentum of the seasons begins to change. We think of Spring as the time of new beginnings, but Winter is when the light begins to return.

This last trip around the Sun was a rough one. While the light has been growing since November 4, 2008, we’re only starting to feel the warmth. Surviving through the tough times requires an excess of spirit— where the material world withdraws, the imagination begins to fill in the gaps.

What I learned from William Blake is, don’t give up. And don’t expect anything.

– Patti Smith

In thinking about a way to sum up the past year, Patti Smith’s song My Blakean Year kept returning to my thoughts. Smith takes heart in the story of William Blake’s perseverance and faith in the face of utter rejection during his lifetime.

While Smith has learned to play the guitar, she has a limited range. The song My Blakean Year seems to be a sort of talisman, a point of connection for so many threads of her life.

The song gives her a solid foundation on which to stand and perform. It’s the thing that makes rock and roll portable for her. She can scale her performance to a small gathering, a large stadium, or a network television audience.

Having survived her own Blakean year, the song serves as a reminder to honor the past, but engage with the present.

This last year was Blakean for so many around the world. Times are rough, and while we can see the light beginning to grow, we know there are still the tough Winter months ahead. The song serves as a reminder to keep the faith, don’t give up, and don’t expect anything.

my blakean year

In my Blakean year
I was so disposed
Toward a mission yet unclear
Advancing pole by pole
Fortune breathed into my ear
Mouthed a simple ode
One road is paved in gold
One road is just a road

In my Blakean year
Such a woeful schism
The pain of our existence
Was not as I envisioned
Boots that trudged from track to track
Worn down to the sole
One road is paved in gold
One road is just a road

Boots that tread from track to track
Worn down to the sole
One road is paved in gold
One road is just a road

In my Blakean year
Temptation but a hiss
Just a shallow spear
Robed in cowardice

Brace yourself for bitter flack
For a life sublime
A labyrinth of riches
Never shall unwind
The threads that bind the pilgrim’s sack
Are stitched into the Blakean back
So throw off your stupid cloak
Embrace all that you fear
For joy will conquer all despair
In my Blakean year

Written by Patti Smith
© 2004 Druse Music (ASCAP)

What I learned from William Blake is, don’t give up. And don’t expect anything.

One Comment

Sensing the Network: The Sound of the Virtual

Published December 23, 2009 by cgerrish

Over lunch with Steve Gillmor the other day, the topic strayed to the dubbing of foreign films. It linked up to an earlier conversation with Aron Michalski about the digital editing of recordings of live music. Our live experience goes virtual as it moves into the past, sound and vision are no longer linked. They become arbitrarily coordinated streams of media. The soundtrack of a film can be completely replaced and the language spoken by the actors can be localized to particular audiences. Wrong notes or timing in a live music performance can be fixed in post production before a quick release to the Network. The period of latency between the live moment and its distribution through a channel provides the opportunity to match our desires with the physical artifact of production. We get a second bite at the apple.

The other instance where separate streams of sound and video are synchronized to create the appearance of a natural experience is when we have the expectation of sound. This is a common practice in science fiction films set in space. Floating through space, we hear the roar of the engines, the blast of the weapons, and the explosion of the enemy ships. Of course, space is a vacuum and sound vibrations can’t occur without a suitable medium. We dub in the sound that makes emotional sense— desire and experience are synchronized.

The mechanical vibrations that can be interpreted as sound are able to travel through all forms of matter: gases, liquids, solids, and plasmas. The matter that supports the sound is called the medium. Sound cannot travel through vacuum.

While we may consider outer space to be the final frontier, there’s another frontier that has opened in front of us that is being explored every day by ordinary people. The virtual space of the Network is all around us. When we type messages on our iPhones, we hear the sound of clicking keys; when we take digital photos we hear the sound of the shutter clicking; when we drive certain kinds of electric cars, we hear the sound of a gasoline engine.

The haptics of the virtual replicate the physics of the physical world. Events in the virtual space of code trigger a sound stream that has an experiential analogy in the physical world. We’ve virtualized complex mechanical interfaces with knobs, dials, sliders, and various data readouts. The dashboard is the holy grail of business intelligence. Some have even proposed a real-time dashboard as the new center of our computing experience.

Consider for a moment how we’ve begun to dub our virtual space to synchronize it with the physical space of our environment. My iPhone uses a traditional telephone ringing sound to signal when a call is coming through. I selected this sound from a menu of possible sounds. Actual telephones that contain metal bells that ring on an incoming call event are pretty rare these days. Many younger people have only experienced the virtual sound of the old telephone.

The link between sound and vision is arbitrary in the virtual world. Our cheap digital camera can sport a sound sample taken from the most expensive mechanical camera. What’s the sound of code executing? We extend the context from our mechanical physical universe into the virtual universe to give us a sense of which way is up, when something has started and when it’s finished. The sound track to the virtual is a matter of cultural practice, but it’s both variable and personalizible. However, as the mechanical recedes around us, our context also becomes fainter. Will the virtual always be a mirror world, or will some new practice emerge from the Network itself? Can a concept of natural sound be generated from a world where sound doesn’t naturally occur, but is rather always a matter of will?

Comments closed

Collections, Time, Distance: From Medium to Meta-Data

Published December 20, 2009 by cgerrish

These are interesting times for the collector. Collections of books, records, DVDs— these all used to matter. What does his bookshelf say about him? And did you get a look at his record collection? I never knew he collected DVDs of musicals with music by Cole Porter.

As the underlying media that holds these recordings moves toward the digital, the bookshelf and the record cabinet give way to the computer hard drive. The physical limitations of the bookshelf no longer trouble us. We can collect to our heart’s content.

Once we have every piece of music as a digital file on a hard drive, our relationship to the music is displaced from the recording medium (vinyl, tape, cd) to the meta-data about the file. We have no relationship with the bits stored on the disk. If asked to point to which bits represent which song, we would be unable to do so. So instead, we now relate to meta-data in an index. The index of titles assures us that we have indeed collected 20 versions of Beethoven’s 9th Symphony. We can push this button, or that one, and call up the file to be played on a connected sound system.

When my music collection was encoded on vinyl platters, I had a direct relationship with the medium. I was careful not to scratch the vinyl. The record was kept in a paper sleeve that fit into a cardboard album cover. For very special records, I’d keep the album in a protective plastic covering so the artwork on the album cover wouldn’t get worn. I used a fairly complex system to clean the records with a special liquid and a brush. I have no such specific relationship with the bits that now hold much of the music that I ‘own.’

In fact, it’s largely a matter of faith that the bits I think I own are physically located on my hard drive. Frankly, the bits could be anywhere. In this relationship, all I care about is the latency between when I locate the song in the index and push the button that connects it to the sound system, and when the music comes out of the speakers. Increments of time displace the qualities of physical extension.

David Gelernter’s manifesto, The Second Coming gets to these changes very directly:

Today’s operating systems and browsers are obsolete because people no longer want to be connected to computers — near ones OR remote ones. (They probably never did). They want to be connected to information. In the future, people are connected to cyberbodies; cyberbodies drift in the computational cosmos — also known as the Swarm, the Cybersphere.

Where a song encoded as bits is doesn’t really matter. I’m only interested in what action creates the connection between the meta-data in the index, the stream of data from the file, and a system that can decode that stream into audible sound. At this moment in history, physical proximity along with wires and plugs seem to be the best guarantee of delivery of the stream with a minimum of latency. Once that service level agreement can be met via the Network, the local and the remote become displaced by the service contract. Apple’s interest in LaLa.com’s approach to collections of music reflects a recognition of how this relationship is changing as a matter of practice.

Real collectors, the completists, often don’t open the package, don’t interact with the collected item in any way that would damage its potential value. While actual contact is minimal, physical delivery of the items is important. A collector of digital bits collects meta-data in an index; however the emotion and the ritual of collection doesn’t really transfer to the digital realm.

As we make these transitions to the digital, we need to renew our understanding the metaphors we use to navigate this space. We take them granted, we assume desktops, two-dimensional screens, files and folders. Even the idea of name spaces could use rethinking.

Once again, Gelernter on how we create models from metaphors, and how those models are going to change by incorporating time (tangible time = the stream):

34. In the beginning, computers dealt mainly in numbers and words. Today they deal mainly with pictures. In a new period now emerging, they will deal mainly with tangible time — time made visible and concrete. Chronologies and timelines tend to be awkward in the off-computer world of paper, but they are natural online.

35. Computers make alphabetical order obsolete.

36. File cabinets and human minds are information-storage systems. We could model computerized information-storage on the mind instead of the file cabinet if we wanted to.

37. Elements stored in a mind do not have names and are not organized into folders; are retrieved not by name or folder but by contents. (Hear a voice, think of a face: you’ve retrieved a memory that contains the voice as one component.) You can see everything in your memory from the standpoint of past, present and future. Using a file cabinet, you classify information when you put it in; minds classify information when it is taken out. (Yesterday afternoon at four you stood with Natasha on Fifth Avenue in the rain — as you might recall when you are thinking about “Fifth Avenue,” “rain,” “Natasha” or many other things. But you attached no such labels to the memory when you acquired it. The classification happened retrospectively.)

4 Comments