Key pairs and GPG¶
The basics are that all modern security relies on a simple feature of prime numbers - it’s really hard to find a factor of a very very large number. Hard as in hundreds of years via brute force.
Edward Snowden said it simply - trust the math
But the math is simple - how is it implemented
My goal is to look at the anatomy of a GPG key (openpgp) and show how the primary key self signs it’s user ids and subkeys - that way I can show a calculation in process and show how subkeys can be split up and still have link back to primary key pair
Using larger than 4096 size http://www.jroller.com/robertburrelldonkin/entry/gnupg_8192bit_rsa_keys
Basdically adjyst ‘keygen.c’
Biblio¶
https://davesteele.github.io/gpg/2014/09/20/anatomy-of-a-gpg-key/
https://alexcabal.com/creating-the-perfect-gpg-keypair/ https://wiki.debian.org/GnuPG/AirgappedMasterKey
https://wiki.debian.org/Subkeys?action=show&redirect=subkeys https://security.stackexchange.com/questions/31594/what-is-a-good-general-purpose-gnupg-key-setup https://security.stackexchange.com/questions/29851/how-many-openpgp-keys-should-i-make/29858#29858 https://futureboy.us/pgp.html
How does SSL/TLS verify a certificate https://security.stackexchange.com/questions/72077/validating-an-ssl-certificate-chain-according-to-rfc-5280-am-i-understanding-th
https://tools.ietf.org/html/rfc5280#section-6.1
https://tools.ietf.org/html/rfc5246
How does this all fit together and why does it matter?
Encryption underpins everything. But just thinking its multiplying primes and then ... secure is fallacious and just leaves that “Gotham city” feeling (lego batman)
I don’t like floating on nothing.