Category Archives: Augmented Reality

SophiaBot: What I’ve been working on for the past month…

I have been busy in my basement constructing a robot with which I can have conversations and play games.  Except that the robot is more of a program, and I didn’t build the whole thing up from scratch, but instead cobbled together pieces that other people have created.  I took an Eliza-style interpreter written by Nicholas H.Tollervey (this is the conversation part) along with some scripted dialogs by Dr. Richard S. Wallace and threw it together with a Z-machine program written by Jason Follas, which allows my bot to play old Infocom games like Zork and The Hitchhiker’s Guide to the Galaxy.  I then wrapped these up in a simple workflow and added some new Vista\.NET 3.0 speech recognition and speech synthesis code so the robot can understand me.

I wrote an article about it for CodeProject, a very nice resource that allows developers from around the world to share their code and network.  The site requires registration to download code however, so if you want to play with the demo or look at the source code, you can also download them from this site.

Mr. Tollervey has a succint article about the relationship between chatterboxes and John Searle’s Chinese Box problem, which obviates me from responsibility for discussing the same.

Instead, I’ll just add some quick instructions:

 

The application is made up of a text output screen, a text entry field, and a default enter button. The initial look and feel is that of an IBX XT theme (the first computer I ever played on). This can be changed using voice commands, which I will cover later. There are three menus initially available. The File menu allows the user to save a log of the conversation as a text file. The Select Voice menu allows the user to select from any of the synthetic voices installed on her machine. Vista initially comes with “Anna”. Windows XP comes with “Sam”. Other XP voices are available depending on which versions of Office have been installed over the lifetime of that particular instance of the OS. If the user is running Vista, then the Speech menu will allow him to toggle speech synthesis, dictation, and the context-free grammars. By doing so, the user will have the ability to speak to the application, as well as have the application speak back to him. If the user is running XP, then only speech synthesis is available, since some of the features provided by .NET 3.0 and consumed by this application do not work on XP.

The appearance menu will let you change the look and feel of the text screen.  I’ve also added some pre-made themes at the bottom of the appearnce menu.  If, after chatting with SophiaBot for a while, you want to play a game, just type or say “Play game.”  SophiaBot will present you with a list of the games available (you can add more, actually, simply by dropping additional game files you find on the internet into the Program Files\Imaginative Universal\SophiaBot\Game Data\DATA folder (Jason’s Z-Machine implementation plays games that use version 3 and below of the game engine.  I’m looking (rather lazily) into how to support later versions.  You can go here to download more Zork-type games.  During a game, type or say “Quit” to end your session. “Save” and “Restore” keep track of your current position in the game, so you can come back later and pick up where you left off.

Speech recognition in Vista has two modes: dictation and context-free recognition. Dictation uses context, that is, an analysis of preceding words and words following a given target of speech recognition, in order to determine what word was intended by the speaker. Context-free speech recognition, by way of contrast, uses exact matches and some simple patterns in order to determine if certain words or phrases have been uttered. This makes context-free recognition particularly suited to command and control scenarios, while dictation is particularly suited to situations where we are simply attempting to translate the user’s utterances into text.

You should begin by trying to start up a conversation with Sophia using the textbox, just to see how it works, as well as her limitations as a conversationalist. Sophia uses certain tricks to appear more lifelike. She throws out random typos, for one thing. She also is a bit slower than a computer should really be. This is because one of the things that distinguish computers from people is the way they process information — computers do it quickly, and people do it at a more leisurely pace. By typing slowly, Sophia helps the user maintain his suspension of disbelief. Finally, if a text-to-speech engine is installed on your computer, Sophia reads along as she types out her responses. I’m not certain why this is effective, but it is how computer terminals are shown to communicate in the movies, and it seems to work well here, also. I will go over how this illusion is created below.

In Command\AIML\Game Lexicon mode, the application generates several grammar rules that help direct speech recognition toward certain expected results. Be forewarned: initially loading the AIML grammars takes about two minutes, and occurs in the background. You can continue to touch type conversations with Sophia until the speech recognition engine has finished loading the grammars and speech recognition is available. Using the command grammar, the user can make the computer do the following things: LIST COLORS, LIST GAMES, LIST FONTS, CHANGE FONT TO…, CHANGE FONT COLOR TO…, CHANGE BACKGROUND COLOR TO…. Besides the IBM XT color scheme, a black papyrus font on a linen background also looks very nice. To see a complete list of keywords used by the text-adventure game you have chosen, say “LIST GAME KEYWORDS.” When the game is initially selected, a new set of rules is created based on different two word combinations of the keywords recognized by the game, in order to help speech recognition by narrowing down the total number of phrases it must look for.

In dictation mode, the underlying speech engine simply converts your speech into words and has the core SophiaBot code process it in the same manner that it processes text that is typed in. Dictation mode is sometimes better than context-free mode for non-game speech recognition, depending on how well the speech recognition engine installed on your OS has been trained to understand your speech patterns. Context-free mode is typically better for game mode. Command and control only works in context-free mode.

48% of Americans Reject Darwinian Evolution

 

A new Newsweek poll reveals frightening data about the curious disjunct between faith and science among Americans.  Pundits have attributed these results to anything from poor science education in pre-K programs to global warming.  According to the poll, while 51% percent of Americans still ascribe to Darwin’s theory of gradual evolution through adaptation, an amazing 42% continue to cleave to Lamarkianism, while only 6% believe in Punctuated Equilibrium. 1% remain uncommitted and are waiting to hear more before they come to a final decision.

This has led me to wonder what else Americans believe:

The 2002 Roper Poll found that 48% of americans believe in UFO’s, while 37% believe that there has been first hand contact between aliens and humans.  25% of Americans believe in alien abductions, while approximately 33% believe that humans are the only intelligent life in the universe, and that all the UFO stuff is bunk.

The 33% of people who ascribe to the anthropocentric view of the universe corresponds numerically with the 33% of Americans who opposed the recent deadline for troop withdrawal from Iraq (PEW Research center poll).   According to the Gallup poll, in 1996 33% of Americans thought they would become rich someday.  By 2003, this number had dropped to 31%.  According to a Scripps Howard/Ohio University poll, 33% of the American public suspects that federal officials assisted in the 9/11 terrorist attacks or took no action to stop them so the United States could go to war in the Middle East.  A Harris poll discovered that in 2004, 33% of adult Americans considered themselves Democrats.

PEW says that as of 2004, 33 million American internet users had reviewed or rated something as part of an online rating system.  33 million Americans were living in povery in 2001, according to the U.S. Census Bureau.  According to PEW, in 2006 33 million Americans had heard of VOIP.  Each year, 33 million Americans use mental health services or services to treat their problems and illnesses resluting from alcohol, inappropirate use of prescription medications, or illegal drugs.  The New York Times says that out of 33 countries, Americans are least likely to believe in evolution.  Researchers estimate that 33% of Americans born in 2000 will develop diabetes.  In the same year, 33 million Americans lost their jobs.

CBS pollsters discovered that 22% of Americans have seen or felt the presence of a ghost.  48% believe in ghosts.  ICR says 48% of Americans oppose embryonic stem-cell research.  CBS finds that 61% support embryonic stem-cell research.  There is no poll data available on whether they believe that embryos used for stem-cell research will one day become ghosts themselves.

82% of Americans believe that global warming is occuring, according to Fox News/Opinion Dynamics.  79% believe people’s behavior has contributed to global warming.  89% do not believe the U.S. government staged or faked the Apollo moon landing, according to Gallup.  Gallup also found that 41% of Americans believe in ESP, 25% believe in Astrology, 20% believe in reincarnation, while only 9% believe in channeling.  A USA TODAY/ABC News/Stanford University Medical Center poll found that 5% of American adults have turned to acupuncture for pain relief.

According to Gallup, 44% of Americans go out of their way to see movies starring Tom Hanks.  34% go out of their way to avoid movies starring Tom Cruise.  Only 18% go out of their way to avoid Angelina Jolie.

Do Computers Read Electric Books?

In the comments section of a blog I like to frequent, I have been pointed to an article in the International Herald about Pierre Bayard’s new book,  How to Talk About Books You Haven’t Read.

Bayard recommends strategies such as abstractly praising the book, offering silent empathy regarding someone else’s love for the book, discussing other books related to the book in question, and finally simply talking about oneself.  Additionally, one can usually glean enough information from reviews, book jackets and gossip to sustain the discussion for quite a while.

Students, he noted from experience, are skilled at opining about books they have not read, building on elements he may have provided them in a lecture. This approach can also work in the more exposed arena of social gatherings: the book’s cover, reviews and other public reaction to it, gossip about the author and even the ongoing conversation can all provide food for sounding informed.

I’ve recently been looking through some AI experiments built on language scripts, based on the 1966 software program Eliza, which used a small script of canned questions to maintain a conversation with computer users.  You can play a web version of Eliza here, if you wish.  It should be pointed out that the principles behind Eliza are the same as those that underpin the famous Turing Test.  Turing proposed answering the question can machines think by staging an ongoing experiment to see if machines can imitate thinking.  The proposal was made in his 1950 paper Computing Machinery and Intelligence:

The new form of the problem can be described in terms of a game which we call the ‘imitation game.” It is played with three people, a man (A), a woman (B), and an interrogator (C) who may be of either sex. The interrogator stays in a room apart front the other two. The object of the game for the interrogator is to determine which of the other two is the man and which is the woman. He knows them by labels X and Y, and at the end of the game he says either “X is A and Y is B” or “X is B and Y is A.” The interrogator is allowed to put questions to A and B thus:

C: Will X please tell me the length of his or her hair?

Now suppose X is actually A, then A must answer. It is A’s object in the game to try and cause C to make the wrong identification. His answer might therefore be:

“My hair is shingled, and the longest strands are about nine inches long.”

In order that tones of voice may not help the interrogator the answers should be written, or better still, typewritten. The ideal arrangement is to have a teleprinter communicating between the two rooms. Alternatively the question and answers can be repeated by an intermediary. The object of the game for the third player (B) is to help the interrogator. The best strategy for her is probably to give truthful answers. She can add such things as “I am the woman, don’t listen to him!” to her answers, but it will avail nothing as the man can make similar remarks.

We now ask the question, “What will happen when a machine takes the part of A in this game?” Will the interrogator decide wrongly as often when the game is played like this as he does when the game is played between a man and a woman? These questions replace our original, “Can machines think?”

The standard form of the current Turing experiments is something called a chatterbox application.  Chatterboxes abstract the mechanism for generating dialog from the dialog scripts themselves by utilizing a set of rules written in a common format.  The most popular format happens to be an XML standard called AIML (Artificial Intelligence Markup Language).

What I’m interested in, at the moment, is not so much whether I can write a script that will fool people into thinking they are talking with a real person, but rather whether I can write a script that makes small talk by discussing the latest book.  If I can do this, it should validate Pierre Bayard’s proposal, if not Alan Turing’s.

Concerning Ladders

It is a commonplace that humor resists translation.  This was Pevear and Volokhonsky’s conceit when they came out with a new translation of The Brothers Karamazov in 1990, which they claimed finally brought across (successfully, I think) the deep humor of Dostoevsky’s masterpiece.  While accuracy is the goal in most translation efforts, to hold to accuracy when translating humor unavoidably leaves much untranslated.  Thus in translating Lewis Caroll into Russian, Vladimir Nabokov chose to replace English wordplay with completely different puns sensible only to the Russian speaker, all the better to capture the flavor of Caroll’s humor.


A former colleague at the infamous Turner Studios has posted the following joke to his blog, which I must admit I cannot decipher:



 


 Yet I know it is a joke, because he adds the following gloss to the image:


Let the hilarity ensue. Someone put up a site where you can build your own World of Warcraft talent trees. The priest one made me laugh, since it’s twoo, it’s twoo.


What is one required to know in order to decipher this particular joke?  Initially, of course, one must know that this is an artifact of the online game World of Warcraft, which is a complex virtual world people pay a monthly fee in order to gain access to.  Next, the artifact is a “talent tree”, which describes different abilities people can gain through accruing time in the virtual world.  The various talents form a tree in the sense that one must gain lower level talents before one may achieve the higher talents, and while the low level talents form a broad base, there are fewer high level talents to choose from when one gets to the top.  The choice of talents one chooses to acquire, in turn, determines what sort of person one is in the virtual World of Warcraft.


This is the formal aspect of the talent tree.  In order to understand the hilarity of this particular talent tree, however, one must further understand the pictorial vocabulary used to represent talents in this tree, a task requiring a Rosetta stone of sorts.  The use of pictures to tell stories is old, and certainly predates any written languages, with exemplars such as the cave drawings at Lascaux.  Long after the advent of written languages, images continued to exert a central role in the telling of stories and the transmission of culture in societies where the majority of people were illiterate.  It was even the main way that Christianity promulgated its teachings to the masses, and the eventual eclipse of the central role of images in religious life by way of the Protestant Reformation can be seen as a direct result of the  emphasis placed on reading the Bible for oneself, and hence the importance of literacy.


Beyond pratfalls and scatology, I’m not sure that pictures without words are a particularly effective means of transmitting humor.  The talent tree for the priest represented above has less to do with cave paintings at Lascaux than with the Renaissance emblem book tradition, which does attempt to treat images as language, and reached its height of artistic expression with the HYPNEROTOMACHIA POLIPHILI.  The traditional emblem book was made up of a series of 100 or so images that were explicated with poems and allegories.  What sets them apart from instructive religious images is that they require a high level of literacy in order to read and enjoy, whereas religious images during the same period were particularly useful for the illiterate.  In some cases, due to the expense of printing woodcuts, emblem books would even forgo actual images and instead would include mere descriptions of the emblems being explicated.  Implicit in all of this, however, was the understanding that whatever could be said about the emblems was originally and overabundantly expressed in the images themselves, and that the accompanying text merely offered a glimpse into their hidden meanings.


Athanasius Kircher, the 17th century polymath, pursued a similar approach toward deciphering Egyptian hieroglyphs.  An interesting website dedicated to and to some extent influenced by his work can be found here.  Following the work of 19th century Egyptologists like Jean-Fran├žois Champollion, we know today that Egyptian hieroglyphs alternately represent either phonetic elements or words, depending on how they are used.  In the case of cartouches, the series of symbols often found on monuments and usually placed in an oval  in order to set them apart, hieroglyphs were exclusively a phonetic alphabet used to spell out the personal names of Egyptian dignitaries.  For Kircher, however, they represented a language of images which, if not actually magical, were at least possessed of superabundant and secret meaning.  Kircher sought transcendence in his efforts to cull meaning from cartouches.  How far he fell short can be gathered from this gloss by Umberto Eco in The Search For The Perfect Language:


>Out of this passion for the occult came those attempts at decipherment which now amuse Egyptologists.  On page 557 of his Obeliscus Pamphylius, figures 20-4 reproduce the images of a cartouche to which Kircher gives the following reading: ‘the originator of all fecundity and vegetation is Osiris whose generative power bears from heaven to his kingdom the Sacred Mophtha.’  This same image was deciphered by Champollion (Lettre a Dacier, 29), who used Kircher’s own reporductions, as ‘AOTKRTA (Autocrat or Emperor) sun of the son and sovereign of the crown, Caesar Domitian Augustus)’.  The difference is, to say the least, notable, especially as regards the mysterious Mophtha, figured as a lion, over which Kircher expended pages and pages of mystic exegesis listing its numerous properties, while for Champollion the lion simply stands for the Greek letter lambda.


 


 Whereas Kircher’s search for transcendence requires great learning, the icon to the right, of the Ladder of Divine Ascent, is accessible to the unliterate.  Most icons in the Eastern Orthodox tradition are of saints, and are used in prayer to the saints.  Icons that depict stories, such as this icon, are somewhat rare, though there is evidence that this was in fact the prior tradition, and the earliest Christian images, found in the catacombs of Rome, typically depict stories from the Bible.  The icon of the Ladder of Divine Ascent is based on the ladder described by John Climacus in the 7th century book of the same name, and the Orthodox saint can in fact be found at the lower left corner of the image.  Climacus, in turn, borrowed his ladder from the image of a ladder that Jacob dreamed about, a ladder extending from earth to heaven.  In this icon, Christ stands at the top of the ladder, welcoming anyone who can make the full ascent.  At the bottom are monks lining up to attempt the climb, while in between we see ascetics being diverted, distracted, and pulled off of the ladder by demonic beings.  The message is fairly straight forward.  Transcendence and salvation are possible, but very difficult.  The ladder represents the journey, but also the mediation required to ascend from the cthonic to the celestial.


I find the metaphor of the ladder striking because 1) it is man-made, and 2) it is something that one steps off of when one reaches the top.  These two features explain why the talent tree depicted above could never be a talent ladder, even though both are things that one climbs.  The tree is something made of the same earthly material that it grows out of.  It reaches for the sky, but because it is not truly a mediator, it cannot allow one to step off of it, and in fact the higher one climbs, the less stable one’s purchase is.  Just as a pier is a disappointed bridge, as James Joyce indicated, a tree is a disappointed ladder.  It goes nowhere.


This, I take it, is the humor inherent in the talent tree above.  The talent tree provides a semblance of movement upwards, but ultimately disappoints.  It always provides more, but the more turns out to be more of the same.  For an interesting unpacking of this phenomenon, one could do worse than read this cautionary blog about the dangers of playing World of Warcraft:


>60 levels, 30+ epics, a few really good “real life” friends, a seat on the oldest and largest guild on our server’s council, 70+ days “/played,” and one “real” year later…

It took a huge personal toll on me. To illustrate the impact it had, let’s look at me one year later. When I started playing, I was working towards getting into the best shape of my life (and making good progress, too). Now a year later, I’m about 30 pounds heavier that I was back then, and it is not muscle. I had a lot of hobbies including DJing (which I was pretty accomplished at) and music as well as writing and martial arts. I haven’t touched a record or my guitar for over a year and I think if I tried any Kung Fu my gut would throw my back out. Finally, and most significantly, I had a very satisfying social life before.

These changes are miniscule, however, compared to what has happened in quite a few other people’s lives. Some background… Blizzard created a game that you simply can not win. Not only that, the only way to “get better” is to play more and more. In order to progress, you have to farm your little heart out in one way or another: either weeks at a time PvPing to make your rank or weeks at a time getting materials for and “conquering” raid instances, or dungeons where you get “epic loot” (pixilated things that increase your abilities, therefore making you “better”). And what do you do after these mighty dungeons fall before you and your friend’s wrath? Go back the next week (not sooner, Blizzard made sure you can only raid the best instances once a week) and do it again (imagine if Alexander the Great had to push across the Middle East every damn week).

 


The burden of Sisyphus is a perennial staple of humorists, and not a tragedy at all.  Consider the most famous Laurel and Hardy short, The Music Box, in which the conceit of the whole film is the two bunglers trying to move a piano to a house on top of a hill.  Perhaps the most iconic example of this sort of humor is Nigel’s amplifier from This Is Spinal Tap, which “goes to eleven”.  For Nigel, eleven is a transcendent level of amplification, while for the mock interviewer, it is just one more number.  Why not just re-calibrate the amplifier and make ten eleven?  Nigel believes that eleven transforms the amplifier into a ladder, whereas the audience recognizes that it is just a tree.


I am at a point in my life where I see trees and ladders everywhere.  For instance, the constant philosophical debates around the mind-body problem can be broken down into a simple question about whether consciousness is a tree or a ladder.  If consciousness is the complex accumulation of basically simple brain processes, then it is a tree.  If aggregating various physical processes never can achieve true consciousness, then consciousness is a ladder.  And then from these two basic theses, we can arrive at all the other combinations of mind-body solutions, for instance that it is a tree that thinks it is a ladder, or a ladder that thinks it is a tree, or that ladder and tree are simply two equivalent modes of describing the same phenomenon, depending possibly on whether one is in fact a tree or a ladder.


Science fiction plots, in turn, can be broken down into two types: those in which ladders pretend to be trees, and those in which trees pretend to be ladders.  Virtual worlds, finally, are the culmination of a historical weariness over these problems, and a consequent ambivalence about whether trees and ladders make any difference, anymore.  For those who have chosen to forgo the search for ladders, virtual worlds provide a world of trees, which simulate the experience of climbing ladders — virtual ladders, so to speak.


Having had several years of success, Blizzard, the makers of World of Warcraft, have recently released a new expansion to their online world called The Burning Crusade.  Whereas up to this point, players have been limited to a maximum level of 60, those who buy The Burning Crusade will have that ceiling lifted.  With The Burning Crusade, World of Warcraft goes to level 70.