The Computers of Star Trek

by Lois H. Gresh

  Encryption, like other methods of computer security, can also open systems to abuse. If you think that you’re transmitting a message that’s totally encrypted, you might send extremely sensitive data across a network. Suppose someone intercepts your encrypted message and hacks the key you used to turn it into ciphertext. Your sensitive data is at the mercy of the wrong people. Think about transactions that typically occur today. Lots of people do online banking. Many people purchase items on the Internet. Many people trade stocks online. A very small number of these transactions are encrypted as they course the phone lines and travel from computer server to computer server along the global net.

  With all the talk about encryption, it’s worthwhile to point out that very few people use it. You may have PGP keysh, but nobody you know wants to learn PGP and obtain their own keys. One guy doesn’t have time to study the manual, which admittedly, takes a good amount of effort. Another is afraid that his wife will accuse him of sending and receiving adulterous emails if she finds encrypted letters on his computer! There are probably dozens of legitimate reasons why people don’t bother with encryption.

  Personally, we favor strong encryption to protect privacy as much as we can. But this points to the general debate that’s been raging for years about encryption. Some people, like us, think it’s critical to our future security. Other people, like governments, think that encryption will allow bad guys to transmit secret messages about bank heists, murders, and government revolutions.2

  At the present time, almost any encryption method can be hacked by brute force. This means that a programmer tries all possible key values until he finds the correct one.

  If a key is eight bits long, there are 28 or 256 possible keys. Using a programming technique that halves the possibilities and searches only the appropriate branches of a tree for a match, we guess that someone could crack the key after approximately 128 attempts.

  But if an 8-bit key has 28 possible keys, then a 64-bit key has 264 possible keys and a 128-bit key has 2128 possible keys. Bruce Schneier, the king of cryptography, says that it would take a supercomputer 585,000 years to find a correct key among 264 possibilities and 1025 years to find it among 2128 possibilities. 3He also points out that the universe is 1010 years old. On the flip side, Mr. Schneier says that most large companies and criminal organizations have the resources to crack a 56-bit key, and that most military budgets suffice to crack a 64-bit key. He predicts that within thirty years, it’ll be possible to break 80-bit keys.

  Within a hundred years, our current technology will be dust. Hardware will change dramatically into DNA, optical, holographic, and/or quantum forms. And software will change to fit its new hosts. Methods of cryptography will change along with the hardware and software. Who knows how long it’ll take a DNA computer, for example, to crack a 128-bit key coded in flesh rather than metal registers? It might be a quick job using a quantum-level computer.

  In the time of Star Trek, nanotech implants in our bodies will dictate entirely new methods of encryption. Possibly a chemical method based on our neurotransmissions. Or an algorithm based on our blood chemistry. Or on our genetic makeup.

  However, the basic cracking technique will remain the same: infiltrate and break the code. Imagine having a computer function infiltrate your body and attack your implanted body network identity chip. No doubt you’ll have a mechanism to fight the disease of infiltration, much as our blood fights infections today. Tiny nanotech-manufactured devices will scour your bloodstream, find all attacking cracker code, and eat it.

  In summary, today’s encryption methods are not terribly relevant to the world of Star Trek. The use of “fractal encryption algorithms” by Data is absurd. Just more technobabble to make the show sound futuristic and serious.

  For the curious, here’s a brief summary of the state of current encryption technology. For details, we suggest that you study not only Applied Cryptography by Bruce Schneier, but also recent articles in magazines such as Dr. Dobb’s.

  Symmetric encryption means that both the sender and receiver of information use the same secret key. The Data Encryption Standard (DES) is the most famous example of symmetric encryption. DES uses a 56-bit key applied to 64-bit blocks of data. DES is still in widespread use in the banking community. In July of 1998, the Electronic Frontier Foundation at http://www.eff:org reported that it had created a $220,000 computer that could break a DES key in four and a half days. For people who believe in Moore’s Law, this means that the DES-cracking machine will cost $110,000 in five years, and it might crack the key in two days. Further, in the landmark The Electronic Privacy Papers, coauthored by Mr. Schneier and David Banisar, it’s noted that “Within a few years, experts anticipate that DES will no longer be secure from even low-level attacks.”4

  Another symmetric technique is triple DES, which inputs three 56-bit keys to an array of three DES chips. Triple-DES is supposedly too slow for various applications. Then there’s IDEA, which uses a 128-bit key on 64-bit blocks of data.

  In January of 1997, hoping to replace DES, the National Institute of Standards and Technology (NIST) asked for a new Advanced Encryption Standard (AES). In June of 1998, fifteen encryption algorithms were submitted for review. At the time of the millenium—scheduled for January of 2000—the final AES will be chosen to replace DES. 5

  One of the really cool things about the AES contest is that three of the submissions have already been cracked.

  It’s amusing to note the names of the encryption techniques, as reported by Mr. Schneier (his comments are in quotes): 6 CAST—256. Slower than other AES submissions.

  LOCKI—97. Already cracked.

  Frog. Already cracked.

  Mars. “IBM gave the world DES, and Mars is its submission to AES ... the pedigree and impressive design document make this a strong candidate despite its ‘kitchen sink’ appearance.”

  Magenta. Already cracked.

  RC6. From Ron Rivest at RSA Data Security, Inc.

  Decorrelated Fast Cipher (DFC).

  Serpent.

  E2.

  Rijndael.

  DEAL. A variety of triple-DES.

  Hasty Pudding Cipher (HPC). “Take everything you can think of, throw it in a cipher, shake well, then add some attitude. ‘Bizarre’ is all I can say.”

  Crypton.

  Twofish.

  SAFER+.

  We mention RC6, so we should also mention its predecessors. All come from RSA, which is not only the name of Ron Rivest’s company but also the name of yet another encryption technique. RC, in the wonderful world of computer acronyms, stands for Ron’s Code. It also stands for Rivest Cipher. RCI, missing from the list, was a design that flopped. RC3, also missing, was cracked before it was released. RC2 uses a variable-length key on a 64-bit block of data. RC4 is the same as RC2, except the former is a stream cipher (operates on the plaintext one bit or one byte at a time) rather than a block cipher (operates on the plaintext in blocks of data). RC5 permits you to change the block size, key length, and the number of iterations used for encryption. The RC algorithms are all symmetric encryption techniques.

  RSA, on the other hand, is an asymmetric technique, also known as a public key approach. This means that the encryption key differs from the decryption key—often called the private key. RSA multiplies two huge prime numbers to obtain its decryption key. Factoring the key using today’s computers could require several billion years.

  PGP combines IDEA for encryption, RSA for key management and digital signatures, and MD5 for hashing functions.

  So what is MD5? There are more encryption techniques than fleas on a dog. Before MD5, we had MD2 and MD4. All were created by Ron Rivest of RSA, Ron Rivest’s company. And even if we told you about MD5 and hashing, you still wouldn’t know about Blowfish and Twofish. Or Panama.

  In the not-so-distant future, it’s hypothetically possible that a digitally encrypted transmission could be further encrypted with the fingerprint of the receiver. Thus, only the specific person b
eing sent the message would be able to read it. But if this technique becomes common, hackers will quickly develop methods to duplicate fingerprints—

  Hey, haven’t criminals done that already?

  Only when biometrics reach the level of nanotech will we see real biometric encryption—in three or four hundred years.

  In Star Trek, messages between starships and their bases, and between crewmembers and their ships, are encrypted before transmission. The coding technique is never mentioned. The encryption algorithm is no doubt far more advanced than today’s methods. Still, the Trek encryption codes are clearly imperfect. The self-destruct code on two versions of the Enterprise rely on voice recognition, surely not a very secure encryption technique (as demonstrated by Data in several instances). The Borg, as evidenced by Seven-of-Nine, have little trouble breaking Voyager’s encryption codes (“Scorpion” and all episodes that follow, VGR). While the Borg are a unique race, there’s no reason to believe that their skill at encryption is unmatched. Garak, on Deep Space Nine, often breaks Cardassian encrypted messages for the Federation. Arturis, a member of Species 116 (“Hope and Fear,” VGR) is shown as adept or even more talented at breaking codes (including those of the Federation) than the Borg.

  How best to manage these security problems? Present-day hackers have proven again and again that no program, however sophisticated, is invulnerable to attack, that no code is unbreakable. It’s difficult to believe that this situation will change over the next few centuries. We suspect three hundred years from now the methods might change somewhat, but that the problems of security will remain the same.

  4

  Navigation and Battle

  When Captain Picard, on the bridge of the Enterprise, receives an emergency message from the Federation science station on Ventax II (in “Devil’s Due,” TNG), he immediately instructs Wesley Crusher to lay in a course for that planet at maximum warp. Wesley pushes a few buttons, and Picard tells him to “engage.” Off they go.

  What exactly is Wesley doing? The ship’s computer handles all of the calculations involved in plotting the course, laying in the coordinates, and setting the starship into motion. At best, Wesley just punches in the name of their destination, like a counter attendant at Burger King, and the computer does the rest. There’s no actual reason for the captain to do anything more than speak, other than that it may satisfy Starfleet regulations. The helmsman and pilot on Federation starships serve no purpose aside from tradition.

  Any meaningful discussion of space navigation requires acknowledging two facts: 1. Light travels at approximately 300,000 kilometers/ second (around 186,000 miles/second).

  2. Space is BIG. For example, the nearest star to Earth, Alpha Centauri, is 4.3 light years away, with a light year being the distance light travels in a year (9.46 * 1012 kilometers). The sun, 149.6 million kilometers from Earth, is a little more than 8 light minutes away.

  Size and speed are the reasons the ship’s computer handles interstellar navigation. Because space is so vast and light travels so fast, we need pinpoint accuracy traveling through the void. Close isn’t good enough when we’re talking about warp drive shooting our ship at many times faster than 300,000 kilometers per second. The smallest mistake could plunge a starship into a nearby sun—or far more likely, send it hurtling billions of miles beyond our destination. Navigation in outer space requires precision impossible to achieve other than with computer-generated exactness.

  Yet ship captains have managed to steer their vessels across Earth’s oceans from one port to another with reasonable accuracy for hundreds of years—without the benefit of computers. Ocean of water, ocean of stars, what’s the difference?

  Obviously, the size of the ocean.

  Outer space is so incredibly vast, so immense, it’s difficult to express distance in a manner that makes sense.

  The Atlantic Ocean is approximately 4,025 kilometers wide. Let’s say we are flying in a straight path across the Earth’s surface (a curved line known as a geodesic, since our world is a globe) from point A in London to point B in New York City. If for every kilometer we travel, we deviate off our path by one millimeter (1/1,000,000 of a kilometer), we’d end up in New York City approximately 4 meters from our destination. That’s quite acceptable for such a trip.

  Now, let’s travel from point A in London to point C on an imagined planet in the Alpha Centauri star system, 4.3 light years away. Given that a light year is 9.46 X 1012 kilometers, 4.3 light years is approximately 40 X 1012 kilometers. If we deviate off course by that same millimeter for every kilometer we travel, we’d arrive approximately 40 million kilometers off target.i That’s about the distance from Earth to Venus. Or about fifty round trips between the Earth and the Moon.

  In interstellar terms, that’s not terrible. At normal impulse speed, it would take the Enterprise a little less than ten minutes to reach point C. Not bad. But we suspect Captain Picard would find the delay unacceptable. Especially when that serum needs to be delivered.

  Let’s consider a somewhat more typical trip: a routine mission to a Federation outpost that’s four weeks from Earth traveling at Warp 6—approximately 300 times the speed of light. Assuming the same error as before, 1 millimeter off for every kilometer traveled, the Enterprise would drop out of warp approximately a billion kilometers from the outpost. Almost the distance from Saturn to the sun: four hours’ travel at maximum impulse. Captain Picard surely would not be pleased.

  Simply put, the tiniest navigation errors become magnified by the vast distances involved in traveling through the galaxy. Speed and distance are so great that normal degrees of measurement become meaningless.

  Yet only rarely do we see helm control explicitly turned over to the computer. Consider the Deep Space Nine episode “A Time to Stand.” Sisko and his crew guide a stolen Jem’Hadar starship to a Dominion storage facility. The plan is to blow up the base, disrupting Dominion war activities. Sisko and friends manage to plant a bomb on the base. That’s when Chief O’Brien warns that the explosion will destroy everything within a radius of 800 kilometers. Unfortunately, as the ship prepares to go to warp speed, the base raises its security shield, trapping them close to the asteroid surface.

  The only way for them to escape, the Captain realizes, is to be traveling directly at the security barrier when the bomb blows up the base and takes out the shield’s power generator. Using the computer, Dax calculates they must accelerate to impulse speed exactly 1.3 seconds before the blast. Sisko turns piloting controls of the ship over to the main computer to ensure perfect timing. It’s a rare admission by the Captain that sometimes human reflexes aren’t accurate enough for interstellar navigation.

  Too bad the details as stated by the crew don’t make any sense. The barrier around the base obviously is less than 800 kilometers from the asteroid, otherwise Sisko wouldn’t be worried about the blast. But impulse speed is 75,000 kilometers per second. Having the computer bring the ship to full impulse 1.3 seconds before the blast would instantly flatten the ship against the forcefield. Better to accelerate at 0.001 seconds before the blast. That’s not a problem for a computer that calculates in nanoseconds. But it’s quite impossible for human senses and reflexes.

  The history of Star Trek does offer a possible reason for this massive distrust of computerized space war. An early attempt to test a supercomputer named M-5 as commander of the Enterprise in a battle situation resulted in disaster (“The Ultimate Computer,” TOS). A hundred years later, when Data created the android, Lal, Starfleet command viewed his work with suspicion, citing the M-5 disaster. They wanted the work stopped.

  It’s difficult to believe that the Star Trek writers could have been so traumatized by the M-5 incident that they halted an entire field of scientific endeavor. Still, genetic engineering seems to have been outlawed for three hundred years by the Federation (“Doctor Bashir, I Presume,” DS9) because of the eugenics experiments of the late twentieth century that resulted in the creation of Khan Noonien Singh. People (like D
r. Bashir’s father) are sent to prison for violating these laws, even if done without malice. The universe of Star Trek does have its illogical aspects.

  With that thought firmly in mind, let’s consider the much more complex problem of a full-scale war in outer space. This has always been one of the mainstays of science fiction. Great battles between opposing star systems have filled the pages of science-fiction magazines for as long as there have been science-fiction magazines. Scholarly articles were even written on the subject, such as “Space War,” published in 1939, in which the noted rocket scientist and science writer Willy Ley discussed possible weapons for spaceships. He concluded that ordinary cannons using explosive shells would be quite effective because, to save weight, ships wouldn’t be heavily armored. Though these early studies were detailed and intelligently presented, none of the writers could guess the amazing advances that would occur in the physical sciences in the next half century. Nor did any science-fiction writer predict the astonishing revolution that would take place in cybernetics.

  Some of the best episodes of both Deep Space Nine and Voyager have dealt with battles in the interstellar void. As when Voyager encountered the Borg battling Species 8472 in “Scorpion.” Or even more interesting, the huge space battle with the Dominion shown in Deep Space Nine adventures “Favor the Bold” and “Sacrifice of Angels.”

  In the latter episode, Commander Sisko leads a fleet of six hundred Federation starships on a desperate last minute mission to free Deep Space Nine from the forces of the Dominion. However, an armada of more than twelve hundred enemy ships blocks Sisko’s path. The commander states that the only way to save Deep Space Nine, and thus the Alpha Quadrant, is to punch a hole through the Dominion fleet so that the Federation starships can get to the station. The collision of the two fleets and the ensuing conflict make for exciting television. But none of it makes much sense.