Re:Liblithium: A lightweight and portable cryptography library
by lvh on Wednesday November 16, 2022 @05:17PM
(#33631655)
This seems extremely similar to https://github.com/jedisct1/libhydrogen -- X25519 and Gimli. (well, and the name...). It also appears to pull from its test suite and aim for compatibility in some ways.
[ Parent
| on HN ]
Re:Liblithium: A lightweight and portable cryptography library
by Grimburger on Wednesday November 16, 2022 @08:35PM
(#33633756)
> and the name
Frank Denis also wrote libsodium, so guess it is a pun by him? Don't think there was any lib{chemical element} popularised before that, and there's a few now, but could be wrong.
Re:Liblithium: A lightweight and portable cryptography library
by beagle3 on Wednesday November 16, 2022 @10:41PM
(#33634657)
libsodium itself is a fork of NaCl[0], an acronym for Networking And Cryptography Library, but also the chemical composition of table salt (Sodium Chloride; the latin name for Sodium is Natrium hence Na)
Re:Liblithium: A lightweight and portable cryptography library
by highwaylights on Thursday November 17, 2022 @04:53AM
(#33636828)
Which was also the name used by Google for the precursor to WASM.
Acronyms all the way down.
Re:Liblithium: A lightweight and portable cryptography library
by jedisct1 on Thursday November 17, 2022 @03:57AM
(#33636490)
Yes, libhydrogen is a pun, as it is lighter than libsodium.
The first version was a rewrite of libsodium, that was small and contained in a single C file. Also introduced xchacha20 for the first time. Then, the Gimli paper was published, and libhydrogen was rewritten to take advantage of it.
Smaller than libhydrogen, there's now charm: https://github.com/jedisct1/charm , also a pun. The plan was to eventually add bottom (platform abstraction layer), strange (asymmetric cryptography), and top (high-level APIs) to build a modular, component-based library.