Thinking ahead: Proposing random strings as help for the Crypto.SE design

I think we're all well aware that we have formally graduated and that there will be a site-specific design in the future.

As Crypto involves many random and pseudo-random strings I many designs need such strings. However a core principle of Crypto is transparancy, so arbitrarily chosen random strings are not acceptable and may lead to unnecessary data analysis questions.

So I'm asking in this question for proposal for (pseudo) random strings with historical importance in cryptology:

What are (pseudo) random strings (with background) that reflect the history of cryptology?

Please note: This isn't official and intended as a help for the design team and hobby designer wanting to create a crypto-specific design for some products or web pages.

This is my set of random looking strings:

1. The first four MD5 collisions as per this question: 9603161f a30f9dbf 9f65ffbc f41fc7ef; 8d5e7019 61804e08 715d6b58 6324c015; 9603161f f41fc7ef 9f65ffbc a30f9dbf; 8d5e7019 6324c015 715d6b58 61804e08
2. The (in)famous seed for the most widely used elliptic curve NIST P-256. It's origin is dubious which has brought great cryptologic attention to this particaular "random" value: c49d360886e704936a6678e1139d26b7819f7e90

Perhaps it would be better to come up with a design that reflects the current state of cryptography rather than the history of the subject.

As you might have seen, I've been toying with a design that shows an encryption chain with a wave of random-looking ones and zeroes coming out from the right side. It occurred to me that the design would be a bit more meaningful (and probably more symmetrical) if it also showed some sort of information going in on the left side. So I came up with the following:

HTTP/1.1 200 OK\r\nDate: Fri, 13 Feb 2009 23:31:30 GMT\r\nContent-Type: text/html;charset=utf-8\r\nContent-Length: 46708\r\nSet-Cookie: SESSION_ID=SGVsbG8gV29ybGQK; path=/; domain=example.com; secure;
\x89PNG\r\n\x1a\x00\x00\x00\rIHDR\x00\x00\x00\x10\x00\x00\x00\x10\x08\x02\x00\x00\x00\x90\x91h6\x00\x00\x00MIDATx\x9cb\xf8\x8f\x04\x18p\x00\x145\xc8\xaa\xc1\,\x08Y\x0f\x03A\xd5hz\x18\x88Q\x8d\xac\x87\
Content-Type: multipart/signed;\r\n boundary="--3243f6a8885a308d-313198a2";\r\n protocol="application/pgp-signature";\r\n micalg=pgp-sha256\r\nMime-Version: 1.0\r\nSubject: See attached document\r\nFr
-----BEGIN CERTIFICATE-----\nMIIFTTCCBDWgAwIBAgISESFabet2Ut7cK/VvlXe2kDzoMA0GCSqGSIb3DQEBBQUA\nMF0xCzAJBgNVBAYTAkJFMRkwFwYDVQQKExBHbG9iYWxTaWduIG52LXNhMTMwMQYD\nVQQDEypHbG9iYWxTaWduIE9yZ2FuaXphdGlvbiB
my @top_passwords = qw(123456 password 12345 12345678 qwerty 123456789 1234 baseball dragon football 1234567 monkey letmein abc123 111111 mustang access shadow master michael superman 696969 123123 ba
CREATE TABLE IF NOT EXISTS users (\n user_id INT NOT NULL AUTO_INCREMENT PRIMARY KEY,\n username TEXT NOT NULL,\n password CHAR(60) BINARY NOT NULL,\n email TEXT NOT NULL) ENGINE=MyISAM DEFA
ECParameters ::= SEQUENCE {\n  version INTEGER { ecpVer1(1) } (ecpVer1),\n  fieldID FieldID {{FieldTypes}},\n  curve Curve,\n  base ECPoint,\n  order INTEGER,\n  cofactor INTEGER OPTIONAL,\n   ... \n}


(Note: I wouldn't intend for all of this text to be visible. I made the strings quite long so they can be faded out at both ends. Just a short chunk from the middle will probably be sufficient.)

These strings (all truncated to 200 chars) represent:

• Headers of a typical HTTP request
• This PNG image:
• The current HTTP certificate for stackexchange.com (retrieved via openssl s_client -showcerts -connect stackexchange.com:443)
• A snippet of Perl containing the most popular passwords of 2014
• A MySQL statement to create a table of users and passwords
• Form data containing a credit card number (invalid, of course)
• ECParameters struct from the Internet X.509 Public Key Infrastructure

Since some people might be worried about nefarious information being sneaked into the site header, I decided to use the digits of π (in bases 10 and 16) for the random-looking parts of this data, except for the timestamp in the first line (equal to 1234567890 in Unix time.

I also made a bunch of binary digits for the data coming out the other side (π again: 11.0010010000111...)

11001001000011111101101010100010001000010110100011000010001101001100010011000110011000101000101110000000110111000001110011010001001010010000001001001110000010001000101001100111110011000111010000000010
00001011101111101010011000111011000100111001101100100010010100010100101000001000011110011000111000110100000001001101110111101111100101010001100110110011110011010011101001000011000110110011000000101011
00001010011011011111001001011111000101000011011101001111111000010011010101101101011011010101000111000010010001011110010010000101101101010111011001100010010111100111111011000110111101000100110001000010
11101001101001100011011111101101011010110000101111111111010111001011011011110100000001101011011111101101111011100011100001101011111110110101101010001001100111111010010110101110100111110010010000010001
01111100010010110001111111100110010010010010100001100110010100011110110011100100010110110011110111000010000000000111110010111000101000010110001110111111000001011001100011011010010010000011011000011100
01010101110100111001101001101001000101100011111110101000111111010010010011001111010111111000001101100101010111010010001111011100101000111010110110010110000111000110001011110011010101100010000010000101
01010010101110111001111011010101001010010000011101110000100101101001011001101101011001110000110000110101010011100100101010111100100110000000010011110001011101000110110000001000110010100001100000100001
01111100001100101001000001011110010001100010111000110110110011100011101111100011100111100111011100101100000110000000111010000110000000111001101100100111100000111010001011101100000001111010001010001111