CONNECTICUT LIBRARIES
July/August 1997, Vol. 39, No. 7
CONNECTICUT LIBRARIES
July/August 1997, Vol. 39, No. 7
Hal’s Legacy: 2001’s Computer As Dream and Reality
Good afternoon, librarians. I am a HAL 9000 computer, production number 3. Most of you remember me from the 1968 film 2001: A Space Odyssey. Some of you never saw the film, but know of me anyway. I am an icon of popular culture! Recently, I heard a Jiffy Lube commercial featuring an automobile in need of an oil change. In a voice similar to mine, it summoned its owner, “Dave,” recalling my conversations with astronauts “Dave” Bowman and Frank Poole aboard the Discovery.
I have two birthdays. Director Stanley Kubrick chose January 12, 1992 for his engrossing, mind-blowing movie, while Arthur C. Clarke preferred January 12, 1997 for his pedestrian novel. Favoring Clarke, editor David Stork is celebrating my nativity with a collection of essays by computer experts who assess how close they are to actually building one of me.
The essays vary in length, style, and technical detail. Several, especially those about computer speech by Joseph P. Olive and Raymond Kurzweil, are lengthy and fairly technical. Others, notably that of Donald A. Norman on humans working with machines, are more concise. This latter essay nicely summarizes most of the key issues treated elsewhere.
Much of the dramatic tension in 2001 derives from the growing realization that I am malfunctioning. Discussing the problem out of earshot, the crew decides that they must shut me down. I eavesdrop by lip-reading, and hatch my own plan to rid myself of the troublesome humans. Unlike your 1990’s computers, I speak; I understand what you say; I think; I see; I lip read; I show emotion; I disobey; and I respond with violence when threatened.
My artificial intelligence stirs the imagination of everyone who is familiar with 2001. You may remember that I handily beat Frank Poole at chess. When IBM’s “Deep Blue” computer won the first game of its 1996 match with chess master Garry Kasparov you may have thought that computers had achieved a high level of intelligence. Don’t be fooled! Deep Blue uses “brute-force” processing power to search out 100-million positions per second! It will beat most humans. But it does not think. It does not reason. It lacks common sense. Deep Blue cannot play chess and pilot a spacecraft at the same time. Even with its “attention” totally focused on chess, it lost to Kasparov. Computer scientists, once very optimistic about creating human-like intelligence have a long, long way to go.
Thinking, planning, and problem-solving are tough without common sense, and today’s computers do not have it. They lack the “street smarts” of even a child who has experienced the real world for only five or six years. This deficiency is at the bottom of human inability to create another HAL in 1997. For much of my ability to speak and to respond to the spoken word depends on human-like intelligence, even emotion.
Understanding the spoken word often requires the listener to think in the context of what the speaker is saying, and to anticipate the boundaries of the conversation. For example, researchers have found that humans find it harder to understand words spoken at random than words spoken as part of a natural conversation. You have computers that recognize very carefully pronounced statements, but you would not want to depend on one to get you to Jupiter.
Humans, in everyday conversation, communicate a great deal that is unspoken. You frequently rely on visual cues, like watching the speakers lips and facial expressions, to help you understand what is being said. Fortunately, I can see, recognize a variety of objects and people, and even read lips. Your computers can “see” carefully lit, stationary, two-dimensional objects , but do not do too well with 3-D items that move and change shape (like faces). Perhaps by 2001, they will do better. As far as speechreading of lips and facial expressions goes, your computers are in the dark ages.
I could go on. Instead, I will close by congratulating you. In some ways your 1997 computers surpass me! Size matters. Think small! E pluribus unum. You pack more processing power into smaller boxes than my creators could have imagined. I am monstrous! When Dave “pulls the plug” on me, the memory modules are in a huge room. Today, you get small, cheap computers to do amazing things. When you duplicate my capabilities, it will not be in a giant mainframe, but in a distributed system of some sort. Your HAL will not look anything like me.
Speaking of looks, your interfaces are much handsomer than mine! My terminals display hard to read numbers and text, occasional charts, and very simple visuals. My terminals are surrounded by columns and rows of control buttons and dials. What a mess! In contrast, your computer screens are aglow with bright colors, realistic visuals, and graphs that bring information to life. And, instead of all of those buttons and knobs, you control almost everything with software menus!
For a change, I am impressed. But do not try to disconnect me, at least not until you have downloaded me to a laptop.
NOTE: On May 11, 1997 chess champion Gary Kasparov lost in a re-match against IBM's Deep Blue. The new and improved computer processed chess moves at twice the speed (200-million positions per second) of the earlier version that Kasparov had previously defeated. In June 1997, Dragon Systems of Newton, Massachusetts announced the release of "continuous-speech recognition" software for personal computers.
Return to Looking at Books Online