Cybernetics in the 3rd Millennium (C3M) Volume 12 Number 1, Jul. 2015
Alan B. Scrivener — www.well.com/user/abs — mailto:firstname.lastname@example.org
I spent most of 2014 consulting for a company called SynGlyphX, which uses
3D shapes called "glyphs" to display high-dimensional, complex spatial data
in an intuitive way. The above video explains the concepts. The work is based
on a public domain tool called ANTz.
( github.com/openantz/antz )
Also check out the SynGlyphX YouTube channel,
( www.youtube.com/user/SynGlyphX )
and my colleague Jeff Sale's site about ANTz at EdWorlds.
( www.edworlds.com/antz/toroids )
Nora Bateson's awesome film "An Ecology of Mind," which I have spoken
highly of before, is now on Vimeo.
( vimeo.com/ondemand/bateson )
I received an email from Anna Chekovsky (about a year ago) letting me know
she translated my 1993 article for the Medicine Meets Virtual Reality conference,
"The Impact of Visual Programming in Medical Research," into Swedish.
( www.teilestore.de/edu/?p=1384 )
Follow Up to "What Just Happened?" in C3M volume 11 number 1.
( www.well.com/user/abs/Cyb/archive/c3m_1101.html#sec_1 )
Shortly after I wrote the article, about Samuel L. Jackson and the Academy Awards
dissing digital artists at the 2013 Oscars, I attended the 2013 international
conference on computer graphics and interactive techniques, commonly called
SIGGRAPH, held in Los Angeles July 21-25.
( s2013.siggraph.org )
On the exhibits floor I found a booth for the labor union known as IATSE, the
InternationalAlliance of Theatrical Stage Employees, which as it turns out is
making a full court press to be the union of digital artists and animators for
the movie and television industries.
( iatse.net )
I spent some time talking to the representative in the booth, especially about
the events of the 2013 Oscars. He confirmed what I suspected, which is that
many actors, including Mr. Jackson in particular (according to a conversation
this gentleman had been a party to), feel that their craft is in danger of
becoming extinct, or at least marginalized, due to computer graphics technology.
He and I agreed that this is absurd. After all, for example, why does Pixar keep
hiring big stars to do voices, instead computer generating them? Because only
a human can give the kind of emotional performances audiences want to experience.
Those of us inside the CGI industry understand this, but unfortunately some
actors have believed the unrealistic prognostications that have been around
since the 1980s, predicting complete replacement of actors by VActors (see the
article at the end of this 'zine for more detail). If you ask me, the real
tragedy is that the digital artists and animators, who are also human beings,
are not getting the credit they deserve. The are still considered "below the
line" employees, meaning they are not believed to contribute creatively, but
only provide necessary grunt work, like a grip does. This is a modern scandal.
While I don't generally see the benefits of unionization in the tech world, I
am coming around to the viewpoint that in union-dominated Hollywood the
Johnny-come-latelies in the computer industry need union protection to get the
respect they deserve, as well as screen credits, benefits like overtime and
pensions, and often even just to get paid. More on this as I become aware of it.
"The right of the people to be secure in their persons, houses, papers,
and effects, against unreasonable searches and seizures, shall not be
violated, and no Warrants shall issue, but upon probable cause, supported
by Oath or affirmation, and particularly describing the place to be
searched, and the persons or things to be seized."
— Fourth Amendment to the United States Constitution, 1792
( en.wikipedia.org/wiki/Fourth_Amendment_to_the_United_States_Constitution )
Recent events have highlighted the intrusion of government into
the internet, and the relative insecurity of electronic communications.
The common perception is that all codes can be cracked given powerful
enough computers, and so the only use for encryption is discourage casual
snooping, like a lock discourages people from trying to open a door.
As far as preventing the government from reading your email, most people
think the goal is impossible to achieve.
However, the truth is that there is a technique for creating encoded messages
that cannot be "cracked," that cannot be read without the key no matter how
powerful a computer is used. This technique is well over a century old,
and completely in the public domain, so it cannot be patented and therefore
no vendor has a motive to sell it. It also has logistical challenges that
are greater than any technical challenge, but over time we believe these
associated hassles have gotten less severe. The time is ripe for adoption
of uncrackable encryption on the internet.
Google "one-time pad encryption" or ask any competent programmer with a
mathematical background for more information.
You've seen the news about the U.S. National Security Agency (NSA) copying all
the emails it can get its hands on, bad guys breaking into servers at SONY
Corporation and spilling corporate secrets, and most recently the Office of
Personnel Management of the U.S. government admitting they've had records of
tens of millions of federal employees "hacked" — all events which have
triggered a renewed interest in internet security. Today I want to tell you
about a technique that can be used to solve the first problem, but not the
other two. Uncrackable encryption can ensure that no matter how long the NSA
keeps your email, and how advanced computers become in the future, they will
never "crack" your encrypted data. As far as keeping bad guys from
compromising your computers, that's a discussion for another day. (One
solution to that involves custom hardware which can only run apps off a CD
or other non-rewritable memory, also known as "breaking the Von Neumann Model.")
So, to be clear, the problem statement is: in electronic communication
between two parties who trust each other, how can they send data files between
them that are encrypted in such a way that no one else can decrypt them?
A Menagerie of Analogies
Asking around, I've discovered that a lot of people find encryption to be very
mysterious, so I'm going to proceed by a series of simple analogies.
Analogy #1: The lock box. Most people understand the lock and
key, an amazingly digital technology from the analog era. If I give you a lock
box and send you to see Alice, who has the key, you can't see what's in the box
unless you pick the lock or break the box. This is the promise of encryption,
that someone — or some mechanism — can carry an encrypted message,
but no one who possesses it will be able to read it without the encryption key
that allows it to be decrypted back to its "plaintext" form.
the routes of Paul Revere and associates (thehistoricpresent.wordpress.com/2011/08/15
/what-does-one-if-by-land-two-if-by-sea-mean/)Analogy #2: "One if by land, two if by sea." Most every
schoolchild in America knows the story of Paul Revere's ride, warning the
Minutemen that the British were coming, and how a signal in the tower of the
Old North Church revealed their route -- "One if by land, two if by sea." Though
anyone in Boston could see the lights in the tower (did you remember that there
were two?) only those who knew the code could interpret them.
But what if the secret of the code got out? The British (or anyone else) could
know what the colonists knew. A more sophisticated code would work this way:
on the morning of the ride the signaler ( sexton Robert Newman ) could flip
a coin, and use this formula:
if heads, use "one if by land and two if by sea"
if tails, use "two if by land and one if by sea"
and then have a messenger deliver the coin to Revere, with a smudge of lampblack
on the side that came up. In order to intercept the message and know what the
colonists knew any third party would need to see the lights and the coin.
Note that there is no way to reliably guess the outcome of the coin flip.
People have been using coin flips to produce an unpredictable outcome for
millennia, and time has tested the technique pretty thoroughly. In my book,
unpredictable is a good working definition of random. So the above
encoding is an example of a cipher with a random key.
A Menagerie of Cyphers
"Once I had a secret code,
Where A was B, and B was G.
G was K, and K was J;
And J was M, and M was P.
V was X, and X was V,
And U was I, and I was U.
87 stood for Z,
And 2 for T, and T for 2.
O was 12, and Q was 17;
I still don't know what those numbers mean.
That is how we won the war;
My secret code's no secret anymore.
H was 9, and S was 33;
Oh, how that confused the enemy!
Then a recent survey showed
That I don't understand my secret code."
— "Secret Code" (parody of "Secret Love") by Allan Sherman, 1953
( www.youtube.com/watch?v=rdrN9lOw6po )
( www.amazon.com/exec/obidos/ASIN/B003SWFLYW/hip-20 )
So let's talk about ciphers for a minute. Any encoding technique which
substitutes one symbol for another is called a cipher. Fans of the movie
2001: A Space Odyssey may recall that clever viewers spotted the fact
that the name of the murderous computer, HAL, could be transposed to IBM just
by bumping each letter up to the next one in the alphabet (and from Z to A to
wrap around). This is a trivial cipher, and not worth using unless perhaps if
you're trying to confound children. This type of cipher is know as as a ROT1
since the letters are "rotated" by one.
( listverse.com/2012/03/13/10-codes-and-ciphers/ )
shift by 1 (ROT1)A -> B
B -> C
C -> D
Y -> Z
Z -> A
You can add a little complexity by bumping up by a different number than one (and
of course you must again wrap around from Z to A if you go beyond the end of the
alphabet.) This is essentially the encryption technique provided by the famous
Captain Midnight (above) and Little Orphan Annie decoder rings of yore. It is
called the Caesar cipher, named after Julius Caesar, who used it in his
( en.wikipedia.org/wiki/Caesar_cipher )
shift by 4 (ROT4)A -> D
B -> E
C -> F
Y -> C
Z -> D
The next step in increasing complexity is to have each letter encode as another
in some scrambled way, instead of having each "mapping" be by shifting all the
letters by the same amount. (Allan Sherman's song, "Secret Code," describes
such a cipher in the lyrics quoted above.)
scrambledA -> M
B -> K
C -> Y
Y -> D
Z -> N
It might seem like this would make a pretty good cipher, but actually it
wouldn't. The table above is considered the "key" to the cipher, and this
case the key is one letter long — it is applied to each letter the
same way. This means that every time a letter in the plaintext is repeated,
the encrypted letter will also always be the same. This makes it possible
to find patterns in the message. The historical novel "Cryptonomicon" (1999)
by Neal Stephenson has a passage in which such s cipher is quickly broken.
( www.amazon.com/exec/obidos/ASIN/0060512806/hip-20 )
In World War II a group of Army recruits who are being screened for possible
code-breaking talent are placed in a classroom with a chalk board, and an officer
begins writing without addressing them.
"Reading from a notebook, [Commander Schoen] writes out the following in block
Around the time that the fourth or fifth number is going up on the chalkboard,
Waterhouse feels the hairs standing up on the back of his neck. By the time the
third group of five numbers is written out, he has not failed to notice that
none of them is larger than 26 — that being the number of letters in the alphabet.
His heart is pounding more wildly than it did when Nipponese bombs were tracing
parabolic trajectories toward the deck of the grounded Nevada. He pulls a pencil
out of his pocket. Finding no paper handy, he writes down the numbers from 1 to
26 on the surface of his little writing desk."
( www.euskalnet.net/larraorma/crypto/slide7.html )
If you have the same techie impulses I do you might like to attempt to break the
code before you continue reading.
By doing a frequency analysis the character of Waterhouse quickly identifies
18 as standing for the letter E in the message, and goes on to solve it:
"ATTACK PEARL HARBOR DECEMBER SEVEN" — which immediately earns him a
place in the code-breaking team.
A more complicated case is called a Vigenére cipher. In this
case the shift is different for each character in the message, up to a
point where the key repeats. Often a phrase is used to represent the shifts.
The first time I ran across this technique was in a Hardy Boys book I read as a
kid. The boys found a piece of paper covered in apparent gibberish, with the
initials C.S.A. written across the page. They eventually figured, with the
paper dated from the Civil War and the initials stood for "Confederate States
of America" which was the encryption key. Using this key over and over they
decrypted the message.
In such an encoding each letter in the key represents a numeric shift in the
letters: A is 1, B is 2, etc. So the first letter in the message is shifted
3 letters (C), the second letter is shifted 15 letters (O), the third letter
is shifted 14 letters (N), etc. This kind of cipher is possible to crack using
the same letter frequency techniques (though a computer is probably required)
as long as the message is no shorter than the key. This brings us to the central
lesson here: if the key is as long as the message, and the key is random,
the cipher becomes uncrackable. Not hard to crack, not crackable only with
powerful computers and a long time, but theoretically uncrackable.
"Any one who considers arithmetical methods of producing random digits is,
of course, in a state of sin."
— John Von Neumann, 1951
( en.wikiquote.org/wiki/John_von_Neumann )
While current debates over encryption wrestle with "how strong is strong enough?"
(a 512-byte key is often defined as "strong" encryption), the technique of using
a random key as long as the message has been known since 1882, and has been
in the public domain since the 1930s, when a 1919 patent expired.
( en.wikipedia.org/wiki/One-time_pad )
Of course, creating random numbers is a tricky business. The best definition
of "random" is "unpredictable." But it is only possible to prove that a given
sequence of numbers isn't random, not that it is. Still, human history
provides a time-tested set of techniques, mostly associated with gambling. Coin
flips, roulette wheels, dice, lottery balls and thorough card shuffles have been
used for a long time without problems (in the absence of cheating) to create fair
games of chance. The above-mentioned "Cryptonomicon" describes secretaries
at the legendary Bletchley Park in England in World War II drawing pieces of
paper to create random keys for one-time-pad codes used by allied military
codes. The big problem is that these techniques are labor-intensive. Recently,
hardware random number devices have become popular, using various types of
electronic, thermal or quantum noise to generate random numbers, which at last
allows the process to be automated.
( en.wikipedia.org/wiki/Hardware_random_number_generator )
It must be emphasized, as Dr. Von Neumann warned us, that using an algorithm
or formula — which is a repeatable process — cannot produce random
numbers; that is why these techniques are called pseudo-random.
Recent advances in the mathematical field called chaos theory have yielded
a better theoretical understanding of how the traditional gambling techniques
actually work, but the best evidence of their effectiveness is that people keep
Both sides in World War II used this technique for some secure communications,
and when used properly these codes were never broken. In the 1960s when the
US and the USSR implemented the so-called "red phone," a teletype system
was combined with paper tapes of random numbers, identical at both ends,
to ensure secure transmission. (I was surprised to learn that the system
was never used for any "real" communication between the superpowers; only test
messages consisting of quotes form American and Russian literature were sent
back and forth.)
Believe me when I tell you that a message encoded with a truly random one-time-pad
can not be "cracked" by any cryptanalysis technique, now or in the future. If
the NSA is collecting citizen emails and archiving them against the day when
future techniques have evolved to allow them to be read, this approach will not
do any good, ever, with randomly encrypted communications.
How Do I Prove This To You?
Oops. It's really easy for me to say "believe me," but how can I make you
believe? I'm trying to get people to make decisions that affect the security of
their business dealings, and that takes some proof. I realize I have a
mathematical mind, and that makes me an outlier. You can prove things to me
with math, but I don't think that's the case for most people. Still, I'm
going to try.
Let's go back to the "one if by land" analogy. Forget the land and sea part;
imagine instead that someone in the Old North Church needs to communicate a
message to Paul Revere every night, about some tactical fact in the Revolutionary
War, a fact that can be represented by a yes/no message, what we call in the
cyber age a single "bit." If the British saw one or two lights in the church
tower every night, and they also could find out the message later, like whether
an army attacked, they might be able to figure out the code — but not if
a coin flip was being used to determine the encoding. Without seeing the
smudged coin they would have no way to infer what one or two lights meant.
But it gets better. If Paul Revere were arrested by the British and they
demanded to know the code, he could tell them either scenario: the one for
heads or the one for tails. In other words, he could force the message to
seem to be whatever outcome he wanted by manipulating the key.
Now, consider that a random one-time-pad code is just a series of one-bit
messages just like this strung together. Each random bit is independent, which
means if you figure out one of them it doesn't help you with the others. This
means that, like with Paul Revere, if someone demands the key you can force the
message to seem to be whatever outcome you want by manipulating the key.
This was the essence of Claude Shannon's proof in the 1940s that the code is
uncrackable. With any cipher that repeats the key there is an ability to guess
and then see if your guess is right. The key has to fit in a number of places,
and so a wrong guess is easy to eliminate. Since with a one-time-pad you can
pick a key to decrypt a code message into any plaintext (of the same
length), there is no way to verify or disprove any guess of the key.
(A little more detail: if you have an encoded message, usually called the
ciphertext, you can make it seem to decode to any fake plaintext of
the same length by performing an exclusive or operation on the
ciphertext with the fake plaintext, which will yield a fake key. Anyone
using this fake key on the ciphertext will get the fake plaintext. This
was the substance of Shannon's proof of uncrackability.)
Is this making sense?
The Trick Is Getting Everyone To Follow the Plan
"I usually go around speaking on the threat of the human element,
particularly on social engineering."
— Kevin Mitnick, famous cybercriminal
( www.inspirationalstories.com/quotes/t/kevin-mitnick )
Okay, I'm going to warn you up front that this plan, as I present it, is a pain
to implement. What you want is a system that is seamlessly integrated with
email and other infrastructure, but since you and I don't control that, instead
this is a bolt-on solution. And let's be clear what problem we're solving.
With the system I describe, you can be confident that anyone intercepting any
emails will be utterly unable to decrypt and read your messages (or look at your
files, such a Excel spreadsheets) if all they have to work with is the traffic
in and out of your network. This is not a plan for preventing cracking into
your network, and stealing data, installing keyloggers and rootkits (look them
up), or any kind of physical break-in resulting in theft of hardware and the data
on them. But we need to solve one problem at a time.
A reasonable use-case would be a company with a bunch of field offices, with the
need for each office to email confidential company information, such as financial
data, to headquarters every fiscal quarter so accounts can roll it all up into a
quarterly report. To protect these files, which are infrequently sent between
trusted parties, you would need this set of equipment, software and procedures:
key generation — Somewhere at headquarters is a special computer used
only for key generation. Ideally it has no internal hard drive and runs
programs off Read Only Memory (ROM), and is not connected to the internet.
A cage in the warehouse would be a good place, so it is easily observed. The
computer could be a small Linux system like a Raspberry Pi,
( en.wikipedia.org/wiki/Raspberry_Pi )
or a custom computer made with Arduino technology.
( en.wikipedia.org/wiki/Arduino )
Its unmodifiable code would only do one thing: copy random bits from the
hardware random number source and write them into two identical files on two
USB drives, along with a key id number (a serial number for keys). A human
operator would switch out the USB drives when they were full, and label them
with both with the same serial number.
key distribution — One copy of each key USB drive stays at
headquarters while the other is delivered to a branch office. This is the most
onerous part of the plan. They probably shouldn't be mailed, and using a
commercial shipper would be iffy. A bonded courier is the best solution.
(I'm imagining motorcycle couriers in black leather riding suits, but hey,
that's just me.) This may sound like a big deal, but companies already must
solve similar problems: how to deliver cash, confidential medical records,
biohazard waste, etc. This could piggy-back off those efforts in some cases.
message encryption — When an employee at a branch office wants
to send an encrypted message, they plug in a key with enough random bits
available and run a program to encrypt the file. This may be the same program
as the decryption program listed below, or a separate program. The bits from
the USB drive are used to encrypt the plaintext file (or any type of file) into
an encrypted file. A straightforward approach is to use an exclusive or
function. I have tested this and it works, both to encrypt and decrypt.
( en.wikipedia.org/wiki/Exclusive_or )
A simple prototype program in the C language is on my web site.
( www.well.com/user/abs/Crypto/xor.ALL )
The program, for example, could take a Microsoft Word file, foo.docx,
and create an encrypted version called foo.rotp (the rotp stands
for Random One-Time-Pad). Then the program deletes the bits in the key file
it just used from the USB drive. This is so they won't be re-used by mistake,
and for security against crackers or subpoenas.
The program also adds "garbage" bits to the end of the file to mask its true
The sender then emails the encrypted file to a designated recipient as
a standard attachment.
file format notes — As a side note, the .ropt file will need
to contain the following fields in addition to the encrypted bits:
the key serial number
the start address in the key,
the actual length of the input file,
the file suffix (in this case docx) of the input file,
a hash of the encrypted bits (explained below).
All but the first two fields should end up encrypted with the data.
message decryption — The designated receiver gets the email
with the attachment, saves it, and then runs the decryption program to re-create
the original file. Ideally the rotp suffix is registered with the operating
system so that the file can be simply "double-clicked" to perform this operation.
The decryption program must:
locate the key file by serial number on a USB drive, or ask for it if it is not yet mounted,
exclusive-or the key bits and encrypted bits to restore the plaintext file,
delete the key bits used from the USB drive,
check that the hash is correct,
strip the garbage bits, and
write out the plaintext file and restore the original file suffix.
A little subtle point: the messages may not arrive in the order sent, so
deleting the key bits on the receiver side will involve maintaining a
map file of which bits have been used. (The sender program can just
lop bits off the end of the key file as they are used.)
I have been working with these ideas for nigh onto 16 years, and I've had time
to think about the implications. I've also wandered into a few on-line forums
on the topic, and know what kind of half-baked, as well as fully-baked,
objections people have raised. I'd like to address a few here. It seems the
format of a Frequently Asked Questions (FAQ) list is the right way to proceed.
Q: Nothing Is Perfect
That's not a question, but you're right. Still, every objection I've read
to this system applies to all other encryption systems, and as far as I can see
none is better.
Q: How Can This Fail?
Mathematically the approach is quite airtight. Operationally, there are things
users can do to damage the security. There have been numerous documented cases
of a one-time-pad being "cracked" because users ignored directives and re-used
the keys. Having the software delete keys as they are utilized reduces this
risk, but like the man said, the problem with making things foolproof is that
fools can be so ingenious. There is no realistic way to keep people from
copying the USB drives with the keys on them and using them again, which
defeats the whole point.
It is also possible that the hardware random number generator can fail to produce
random bits. More study of this is needed.
And, of course, while the keys are being delivered bad guys could intercept them
and copy them, making the whole exercise futile. This problem has to be solved
on a case-by-case basis. Sealing the USB drives in envelopes with wax seals
might help, as would having "tiger teams" attempt to compromise security. Kevin
Mitnick, who does security testing now, says he has never failed to find a hole
in security. It might be as simple as bribing or blackmailing a courier.
Q: What If Something Goes Wrong and the Receiver Can't Read a Message?
In this case the solution is to re-send the message with a new key. Any attempt
at message repair or error recovery would probably compromise the system.
Q: What If You Can't Generate the Keys Fast Enough?
Add more hardware in parallel.
Q: What If You Can Guess Some of the Source File Bits and Use This To Change
This is a legitimate, known flaw of the method. Most file formats have a
well-known structure. For example, a docx file is a zipped XML file, and a Zip
archive has well-defined bits at the beginning, including:
Version needed to extract (minimum)
File last modification time and date
If you can make a reliable guess what these values are, you can reverse-engineer
the key bits for those bits. (It won't help you figure out the other bits
in the key if they are truly random.) This would allow you to change the bits
you know to something else, garbling the message or even changing its meaning.
This is why the hash is added. It is extremely difficult (but not impossible)
to change a file without changing the value of its hash, which would signal that
tampering had occurred.
( en.wikipedia.org/wiki/Hash_function )
Q: If You're Going To All the Trouble To Hand-Deliver the Key Files On
USB Drives, Why Not Just Hand-Deliver the Messages the Same Way Instead?
Good point, and that's not a bad idea, but you should be able to deliver the
keys when it's convenient (as well as less frequently) and email the messages
Q: Why Can't We Email the Keys?
Don't you dare.
Q: What If Ninjas Sneak Into Your Headquarters and Copy the Keys?
Then they could just steal your computers and get all of your unencrypted data.
(Quite a few of the objections to one time pads, rebutting the claim that the
codes are unbreakable, seem to involve ninjas.)
Q: How Does This Protect Against Hacks Like the Sony Breakin?
It wasn't designed to.
Q: What About Quantum Computing?
Q: What If Ninjas Modify Your key-Generating System?
Enough with the ninjas!
Q: Do You Have Anything Else To Add?
In an earlier, more naive version of this idea, I proposed using bit streams
snatched from on-line sources like Jpeg files, which would certainly be strong
encryption, but not uncrackable.
( www.well.com/user/abs/Crypto )
Notice that a offered a $20 prize for decrypting a message when I provided a hint
to where to find the key, and a $2000 prize when there was no hint.
I only distributed the message to a few friends, but someone reposted the
challenge part of the file to an encryption newsgroup. I got some flack
because it was such a short message, and these kinds of stunts are usually
regarded as pointless. Nevertheless, I never had to pay the $2000 (or try to
Unfortunately I forgot what the plaintext said, and I lost the key, so I suppose
at this point I'll never know, unless I can recover the memory through hypnosis.
The Political Component
James Madison, author of the
Fourth Amendment in the Bill of Rights,
as pictured on the $5000 bill (Lunatic Blog)
"I think people watch TV and think the bureau can do lots of things.
We cannot break strong encryption."
— FBI Director James Comey testifying to the US Senate Intelligence
Committee on the need for law-enforcement "backdoors" to encryption,
July 8, 2015
( www.npr.org/sections/thetwo-way/2015/07/08/421251662/fbi-director-says-agents-need-access-to-encrypted-data-to-preserve-public-safety )
First of all let me say that, sad as it makes me to admit this, I know that my
ideas are controversial in this day and age. For example, I have a near, dear
relation who is a flight attendant, and I'm pretty sure she wants terrorists
stopped even if the Fourth Amendment gets shredded in the process. But I can't
go for that.
I believe that our founding fathers were quite wise (even if they didn't always
practice what they preached), and there are good reasons to maintain and defend
our constitutional rights. When James Madison wrote "The right of the people
to be secure in their persons, houses, papers, and effects, against unreasonable
searches and seizures, shall not be violated," I think that was meant to include
email, texts, private web forums, and other electronic communications. Our
government claims otherwise (when they bother to claim anything — much of the
time they just snatch what they want secretly). I want to roll this back. It
is still legal to encrypt your communications, though some in law enforcement
keep trying to make a run at this right. During the early days of the Clinton
administration a proposal called the "Clipper chip" would have mandated back
doors for government spying in all electronic devices. Luckily it was shot
down, but it keeps popping up with new names. For a while there they tried
to shut down just talking about encryption, calling it "exporting munitions"
if keys were 512 bytes or more. This has been eased, but you still have to
tell the Bureau of Industry and Security, located about 1000 feet from the
White House in Washington, if you want to release open source encryption software.
( en.wikipedia.org/wiki/Export_of_cryptography_from_the_United_States )
I believe one way to hold back the tide of encroachment is to get technology out
there into citizens' hands while it's still easy. Recall the cautionary tale
of dual key encryption, which barely missed being classified and buried because
the inventors had the good sense to publish their work in Canada just in time.
If you think it's too dangerous to give citizens superencryption technology
(which, as I've said, is already in the public domain), here's something to
consider. If your goal is fighting terrorists "at any cost" you can use that
argument to justify all of these steps:
opening everyone's mail and reading it
tapping all phone calls without a warrant
mounting government spy cameras in every room of every home
randomly arresting and interrogating citizens
torturing all suspects
threatening to kill family members of uncooperative witnesses
and so on. Where do you draw the line?
Look For Us On Kickstarter
With recent events I have decided the time is right to go into production on this
design. My current plan is to do a "Kickstarter" project to open source the
hardware, and sell the software (in source code form) along with pre-built
hardware for convenience. I'll publicize it here if you want to participate.
One big part of the project will be to buy and comparison test hardware random
( en.wikipedia.org/wiki/Comparison_of_hardware_random_number_generators )
We also plan to offer a cash prize for cracking the uncrackable, and hire some
lawyers to try and stay out of trouble. Wish us luck.
"If It's Just a Virtual Actor,
Then Why Am I Feeling Real Emotions?" (Part Six)
(If you haven't read parts one through five, see the archives, listed at the end.)
1993, second year in Anaheim. This is the year they had the virtual sex panel
which, uh, everybody had a good time with. 1993 was also the year that had,
what I think is probably the most amazing SIGGRAPH party that's ever happened,
which was the party at the Richard Nixon Museum. I have this image of watching
Timothy Leary ranting and raving on stage at the Nixon Museum, twenty feet from
the grave of Pat Nixon, and it's... just really cosmic, I don't know.
— Jim Blinn, "SIGGRAPH '98 Keynote"
Building a Better Dinosaur — 1993 was really Silicon Graphics Inc's
year. About half a decade previously, at the same Anaheim Convention Center,
I heard computer graphics conference keynote speeches by Robert Abel
( en.wikipedia.org/wiki/Robert_Abel_%28animator%29 )
and Roy E. Disney,
( en.wikipedia.org/wiki/Roy_E._Disney )
telling us that "someday" CG would be used to do live-action movie effects.
And now, SGI's computers had been used to do the first major photo-realistic
computer animations in movie history, for Steven Spielberg's blockbuster hit
( www.amazon.com/exec/obidos/ASIN/B0087ZG7HK/hip-20 )
To celebrate they rented four full-sized trade show booths on the SIGGRAPH
exhibit floor, and installed a theme park ride, complete with Jurassic Park
jeeps. I didn't get to ride it since demand far outstripped supply, and I
wasn't in the market for their hardware, so I didn't make the VIP list.
(Although, not long afterwards, the company I worked for — Advanced Visual
Systems (AVS) — bought an SGI Indigo 2 workstation and put it on my desk.)
SGI also launched a marketing campaign around the phrase "serious fun," and
printed a series of what looked like baseball cards showcasing the uses of
their computers in showbiz.
Van Dam Rex — About week before SIGGRAPH I had dropped in on
an academic conference at nearby UC Irvine on the topic of visualization in
education. In attendance was computer graphics pioneer Dr. Andres Van Dam,
co-author of the textbook on the subject I'd used in college. I had met him
when he was on the Technical Advisory Board of a previous job of mine, Stellar
Computer. (When I visited the headquarters in Newton, Massachusetts, they let
me use his office since he was never there, off teaching at Brown University
in Providence, Rhode Island.) At the Irvine conference, which I only attended
on the last day, I contradicted Dr. Van Dam during a feedback session. I said
I wanted to offer a view from industry, and explained that we no longer needed
a lot of graduates who knew how to write renderers; we needed experts in
computer graphics applications: chemistry, material science, oil and gas
exploration, 3D medical imaging, geographical displays, animation, kineseology,
etc. He got visibly irate and said "I completely disagree!" It was a learning
experience for me. I realized upon reflection that, especially for academic
conferences, it is important to be there for the whole thing, to build community
and relationships, and also that when offering a contradictory point of view,
it is best to put a positive spin on it.
Anyhoo, a week later I was standing outside of the Anaheim Convention Center
at the beginning of SIGGRAPH, when I was approached by a group of Dr. Van Dam's
graduate students, a few of whom I knew from an AVS connection. One of them
said that he'd heard I'd angered Dr. Van Dam at the education conference,
and gave me a high five. Then he pointed to a large button he was wearing,
which had a T-rex, altered to have Dr. Van Dam's head, and the caption:
"VAN DAM REX." Clearly there was some dynamic at work here that I hadn't quite
(I'm proud to say that I saw Dr. Van Dam two years ago at a SIGGRAPH Pioneers
banquet, and he claims to remember me positively, so either he's placed the
incident behind him or, more likely, forgotten all about it.)
Birds of a Feather Flock Together — Shortly before SIGGRAPH
I was contacted by an academic named Benjamin H. Bratton, who I hadn't heard of
( en.wikipedia.org/wiki/Benjamin_H._Bratton )
I have no idea how he found me, though I was developing a reputation as someone
who liked to speak at conferences, and showed up on time with something to say,
amusing if not enlightening. He invited me to speak at a panel he'd organized,
called "Electronic Image and Popular Discourse." Being a fan of McLuhan's
distinction of literary vs. oral cultures, I prepared some remarks I entitled
"Hypertext or Game Boy: We Are At a Fork In the Road," looking at the important
role going forward of good old alphabetical text in computer interactions. It
was fun and educational, and I met some nice, smart folks, but an important
take-home lesson was this: at the time (but no longer) any SIGGRAPH attendee
could organize a "Birds of a Feather" (BoF) session and be assigned a room and
listed in the program. The people who attended did not have to show a
SIGGRAPH badge (which was always fairly expensive to get). This is what
Benjamin had done for his "panel." Instead of submitting a proposal and
awaiting the decision of a jury, he was able to create this sub-event by fiat.
I learned from his example, and used this lesson in subsequent years.
Next Year in Orlando — One of my traditions at the SIGGRAPH
conference is to visit the booth for next year's conference, as early as possible
during the show, and get a pin and a poster for the following year. (They
usually ran out quickly.) That year the pin and poster showed a stylized
nautilus shell, and advertised the conference a year away in Orlando, Florida
— the first to be held there. SIGGRAPH was staking a claim for the
invasion of theme park entertainment by computer graphics technology, and many
of us were salivating at the chance to participate.
Subsurface Scattering — They say it takes more than a day to
understand the events of a day, and this was true of the handful of days of
the SIGGRAPH 1993 conference. The conference proceedings ended up having
impacts moving forward for quite some time.
( www.amazon.com/exec/obidos/ASIN/0201588897/hip-20 )
I encountered several examples of this, but I'll share one that I have better
recall of, because I blogged about it before. At SIGGRAPH 2003, ten years
later, there was a presentation on the skin rendering in the movie "Shrek"
( www.amazon.com/exec/obidos/ASIN/B0039N74CW/hip-20 )
which was one of the first feature films to do human skin well. The presenter
said the tricky thing was that skin had multiple layers, and a good amount of
water, and simulating how they reflected light was non-trivial. (This is why
movie makeup was invented, because under the bright lights cameras used to need,
deep layers of skin would become more visible and look strange.) He explained
that the mathematics was worked out in the 1930s, but the first use in computer
graphics was described in a paper published in the proceedings of SIGGRAPH 1993.
( www.creativeplanetnetwork.com/news/news-articles/skin-deep/383131 )
I found that many threads similarly connected ongoing developments to work first
published that year.
SUCH HILARITY AND MIRTH
The Bonzo Dog Band on
"Do Not Adjust Your Set" (BBC)
Late in the summer of 1993 I decided I wanted to do something to stay in touch
with Anne Herring. One of the shows she'd performed at the Adventurers Club
was called the "Maid's Singalong," done in character as Ginger Vitus the maid.
She worked the room alone (except for her invisible accompanist, Fingers the
ghost of the keyboard player). She sang some songs, lead the crowd in some
singing, and even encouraged a few brave guests to stand up and perform songs
they knew. (Going through various evolutions, this show continued to be
performed until the bitter end of the club.)
( www.youtube.com/watch?v=T3Z5msSkDEw )
In the spirit of this show, I made a short mix tape — of two songs — that
I thought would be great in the club, "Mad Dogs and Englishmen Go Out in the
Midday Sun" by Noel Coward (1932),
( www.youtube.com/watch?v=z2YvYiWtovM )
and "Hunting Tigers Out In Indiah" performed by the Bonzo Dog Band (1969),
a cover of a 1930 song by Hal Swain & His Band.
( www.youtube.com/watch?v=ZNmL1L3dF6g )
I was able to mail her the cassette, since we'd talked her out of her postal
address. Later she told us she was familiar with the Noel Coward song, but
not the Bonzo Dog Band one. (An aside: the above video of the Hunting Tigers
song is actually from a rare BBC program for children, "Do Not Adjust Your Set,"
which starred several comedians who later went on to fame in the troupe Monty
Python's Flying Circus.)
( en.wikipedia.org/wiki/Do_Not_Adjust_Your_Set )
WATCH US ROAR IN '94
"Bug, Sig and Karla were all a bit annoyed by how 'family-oriented'
and we yearned for traces of its proud history of sleaze and corruption.
I mean, if you can't get lost in Las Vegas then what's the point
of Las Vegas?"
— Douglas Coupland, 1995
( www.amazon.com/exec/obidos/ASIN/0061624268/hip-20 )
I've been a fan of Preview Centers for a long time. I think my first one
was the Walt Disney World Preview Center, which our aunt, uncle and
cousins took us to when our family visited Florida in the summer of 1971,
just a few months too soon to visit the Magic Kingdom. Later I also
greatly enjoyed the Disney's California Adventure Preview Center, located
near the Disneyland ticket booths, which in some ways was better than the
finished theme park. But the world record for preview centers must be held by
Las Vegas in the 1990s. Amid well-videotaped casino implosions (most of which
can be found on YouTube),
( www.youtube.com/watch?v=5igVwiXk-gM )
Sin City was working hard to reinvent itself at that time, and I remember
visiting trailers with concept art, models, and advance souvenirs for sale for
at least three new casino/resorts: the New York New York, the Aladdin, and the
MGM Grand. In each case I bought a t-shirt, and for a while I considered it
to be a techie status symbol (with time travel overtones) to wear a shirt for a
casino that was in the future. (I got a copy of the poster above at the
MGM Grand Preview Center in 1993.)
During this heady era the SIGGRAPH conference came to Las Vegas its one
and only time, in the summer of 1991, and several computer graphics vendors
took advantage of the opportunity to do a hard court press with casino
managers, pitching ideas for Virtual Reality (VR) and other eye candy
to the leaders of the gambling empire. I can't show cause and effect,
but a great deal of cutting edge "new media" style spectacle —
what they used to call "location-based entertainment" — popped up in
Vegas. There was the Pharaoh illusion in front of the Luxor, the
audio-animatronic dragon that emerged from the Excalibur, only to be
vanquished by Merlin the Wizard, and eventually an authentic Virtual Actor
in the form of a computer-generated pirate which had live conversations
with folks entering the Treasure Island casino from the parking garage and
riding down an escalator. (The latter was not very effective; the phenomenon
lasted such a short time on the escalator ride that very few guests realized
it was a live, interactive character.)
Along with this new tech revolution there was also a general trend of Las Vegas
attempting to become more family friendly. Along with a Wizard of Oz theme,
there were lions in lobby of the MGM Grand. The Mirage had the white tigers
from the Siegfried and Roy show on display in their lobby, and a dolphin garden
out back. The afore-mentioned Treasure Island had a live pirate ship battle
with a British man o' war every hour in the false bay at the front of the
casino. The Luxor had a whole level in its pyramid of family-friendly
movie/rides and a world class Sega arcade. The Excalibur also had a whole
level of carnival-style games, puppet shows, medieval fantasy crafts, and not
a gambling machine in sight. In the basement was a jousting dinner show, much
like Medieval Times near Disneyland in California and Walt Disney World in
Florida. Circus Circus and the MGM Grand added full-up amusement parks out
back. New York New York had a roller coaster running right through the
Manhattan skyline facade. Also right on "the strip" Coca Cola and M'n'Ms
built museums, and the DreamWorks-inspired GameWorks opened an arcade.
(My all-time favorite Vegas attraction was Star Trek: the Experience,
which I can't say enough good things about.)
( en.wikipedia.org/wiki/Star_Trek:_The_Experience )
In addition to these attractions at A-list locations, other lesser venues
spent big to attract families with kids, including casinos at State Line
on the California-Nevada border, which had the advantage of being the first
legal gambling drivers came to when arrive on Interstate 15 from the Los
Angeles area. Buffalo Bill's built a water ride, amusement area, and possibly
the scariest roller coaster I've ever ridden, while the Primadonna next door
had a ferris wheel out front and a merry-go-round and arcade in a basement
This dream came crashing down in 1997 when a seven year old girl, left
unattended by her gambling father, was murdered near that merry-go-round.
( articles.latimes.com/1997-05-28/news/mn-63101_1_casino-surveillance )
The casino management was so mortified they changed the name to the Primm
to escape the bad publicity, and ripped out the rides and attractions.
The businesses of Las Vegas did some soul searching, admitted that
gambling-obsessed parents leaving their kids unattended was in fact a
big problem, and backed away from the family-friendly image. A subsequent
national TV ad campaign was "What Happens in Vegas Stays in Vegas," and the
town went back to being Sin City.
D: You just bring that right on, right on up then, uhh?
L: Alright, where do you want me to deliver it?
D: Up at my house
L: Where do you live?
D: Up on the north side
L: On the north side
L: Whereabouts on the north side?
D: Up there by the Japanese amusement park
L: The Japanese amusement park
D: Bambi's deer place there
L: Is that right?
D: Where Bambi goes, nothin' grows
— Hudson & Landry "Ajax Liquor Store" (audio comedy)
( www.amazon.com/exec/obidos/ASIN/B000069Z0U/hip-20 )
Despite the fact that I was involved with two previous computer game projects
which ended badly, in late 1993 I again was bitten by the game design bug,
and decided I wanted to create a new game design. (I wasn't so keen to
implement, so my goal was a design document.)
I mentioned this inclination to Steve Tice, my old boss from Rockwell
who had gone on to lead the company, Simgraphics, that invented the Virtual
Actor and Performance Animation, and who at that point was managing computer
game developers. He said he wanted to encourage my efforts. By way of
inspiration he took me to the International Association of Amusement Parks and
Attractions (IAAPA) conference at the Los Angeles Convention Center in
November of 1993.
( www.iaapa.org )
The L.A. Times reported that over 25,000 people attended that year.
( articles.latimes.com/1993-12-03/local/me-63347_1_convention-attendance-los-angeles )
Steve especially wanted me to see the "VR aisle" that year — in addition
to the traditional vendors selling carnival-type rides, water park equipment,
game prizes, liability insurance, popcorn machines, etc., this new category
featured "motion-base" rides, innovative interactive games, and other "new
tech" solutions to the age-old problem of amusing the Rubes.
Well, I was impressed with the technology and the business opportunities,
and Steve and I agreed to begin having a series of informal discussions.
We ended up having a series of very interesting conversations, often in places
that provided additional stimulation for our ideas. One place we visited
several times was Beach Boulevard in Buena Park, the street that Knott's Berry
Farm was on. I had read in a newspaper article that a two-block stretch of
this street had the largest concentration of separate amusements —
location-based entertainment venues if you will — found anywhere
Historically this street had been home to some almost legendary roadside
amusements, including the above-pictured Japanese Deer Park,
as well as a whole menagerie of other oddities, including famous and
infamous extinct attractions such as:
Movie World/Cars of Stars/Planes of Fame Museum
California Alligator Farm
Wild Bill's Wild West Dinner Extravaganza
Kingdom of Dancing Stallions (home of the world-famous Lipizzaner
Stallions, which were brought to America from Vienna, and later moved
into the Excalibur in Las Vegas)
Po' Folks Restaurant with a model train running through it
At Steve's encouraging I initially presented him with three different ideas
for games, and after some debate we settled one one of them, which I called
"The Big Heat," (inspired by a song of the same name by Stan Ridgeway),
which involved a noir detective playing cat-and-mouse with sinister forces
via time travel in Southern California over an interval of several centuries.
I worked on the design document intermittently for about four years, and
finally took some time off from my day job to work in it full time in late
1997, but it turned out the extra effort didn't lead to me completing it
because I was having a problem with convergence. Like Hercules and
the Lernaean Hydra, which grew two new heads for each one cut off, my
project seemed to sprout new tasks faster than I completed them. I was
also hampered by the growing realization that computer technology wasn't
really ready for this game at the time. I finally gave up and went back to
being a salaryman, and Steve moved on to other projects as well.
(I still have about 4 banker's boxes of material from this unfinished
project in a storage locker, which I may one day revisit. I am pleased
to report that I have gotten much better at getting projects to converge
since then; the best thing I ever did in that regard was to begin writing
this eZine, which taught me how to finish things better than anything I
had done before.)
SCREWING UP MY COURAGE
Kitt Car at Universal Studios Hollywood
I think it might have been New Years Eve 1993/94 that I met Richard Cray; if
not then shortly afterwards. We were introduced by my friend Greg Panos, who
as I mentioned previously is a supernode in the social network. Richard is an
actor, and had been performing in New York in "The Mystery of Edwin Drood," an
interactive production in which the audience helps write the ending. He had
come to Los Angeles because he wanted to become an operator of Virtual Actor
and Performance Cartoon technology. Richard sought out Greg because of his
fame in the VR community, and Greg offered him a place to stay as he chased
his dream on the Left Coast. (Twenty-one years later they are still roommates.)
As 1994 began I was looking ahead to the SIGGRAPH conference coming up in
Orlando. Many of the deadlines for papers, panel, courses, etc. are in
mid-January. I had been thinking about interactive media, Virtual Actors,
and their connection to the Adventurers Club. Though the club didn't use any
technology more advanced than puppets, the actors there had been racking up
close to five years of experience entertaining real audiences with interactive
entertainment. I wanted to find some kind of synergy between their skills
and experience and those of the computer graphics community.
In one of our early conversations I shared with Richard this idea, and he
was wildly enthusiastic. He had recently been doing performance animation gigs
for Simgraphics Corporation, and said he wanted to participate in whatever
event I organized. I was leaning towards a panel, but then realized I should
instead do a Birds of a Feather (BoF) session, as Benjamin Bratton had taught
me at SIGGRAPH 1993. This wold easier, and I would have more time to pull it
together, since the BoF deadline was much later. At some point I came up with
the title, "The VActor and the Human Factor."
Then I had a crisis of confidence. I didn't have as much faith then as I do
now in my ability to make things happen. I almost chickened out, despite
Richard's encouragement. When I shared the idea with Steve Tice he was also
enthusiastic, though he wasn't willing to commit to participate. He did
suggest I involve Brad de Graf, of De Graf-Wahrman fame, who had done
early experiments with Jim Henson with an interactive wire-frame Kermit the
Frog on an Evans and Sutherland vector-based graphics system, and offered to
make an introduction for me.
Still I dithered. I decided I needed to work on screwing up my courage.
I had an annual pass to Universal Studios Hollywood, and they had a small
attraction there which consisted of the "Kit Car" from the TV show "Knight
Rider" (1982-86), about a spy name Michael, played by David Hasselhoff,
who drove a superintelligent computer-controlled car named Kit.
( www.amazon.com/exec/obidos/ASIN/B00005JLG4/hip-20 )
Guests could sit in the car and talk to the computer. A voice actor (not of
course the original star William Daniels, of "A Thousand Clowns" fame) provided
the interactive voice of the car from an unseen location.
As I related in C3M volume 4 number 8:
"I once met a comic (teaching traffic school) who claimed that his first job in
Hollywood was the voice of Kitt; he said the guests all asked, 'Where's
Michael?' all day long."
( www.well.com/user/abs/Cyb/archive/c3m_0408.html )
(As it turned out, when I met him again he admitted he'd made up the story for
comic effect. Oh well.) Anyway, I used this attraction as a laboratory of
interactive entertainment. I would go to Universal and sit, sometimes for
hours, watching and listening as guests young and old interacted with the
car. And as I learned and observed, I worked on pumping myself up.
Later I got an annual pass to Knott's Berry Farm, which at the time had
a themed "land" called "Knott's Airfield." (It had originally been the "Gypsy
Camp," then the "Roaring 20s," and after the airfield theme it became the
"Boardwalk," which is now where most of the big iron thrill rides are located.)
I found the 1920s airfield theme to be quite charming. There was a
restaurant there called Captain Kelly's which I enjoyed, next to the Wacky
Soap Box Racers, which were taken out two years later. When I finally decided
to "go for it" and organize the BoF/panel for SIGGRAPH, I would go to Captain
Kelly's and sit enjoying boysenberry pie and coffee while working on the
paperwork: letters to Brad de Graf, the Adventurers Club actors, and imagineer
Joe Rhode, who helped design the club, and application forms to the 1994
SIGGRAPH committee. I enjoyed the ambience as my reward for sticking with the
task, which I still had some resistance to completing.
TO BE CONTINUED...
Privacy Promise: Your email address will never be sold or given to
others. You will receive only the e-Zine C3M from me, Alan Scrivener,
at most once per month. It may contain commercial offers from me.
To cancel the e-Zine send the subject line "unsubscribe" to me.
I receive a commission on everything you purchase from Amazon.com after
following one of my links, which helps to support my research.
Copyright 2015 by Alan B. Scrivener