Awesome
"Cognate Language 2"
Inspiration: https://github.com/hchiam/cognateLanguage (keeping separate for reference). This repo cognateLanguage2 is to be a smaller project. Since the original project already covers Chinese, Spanish, Hindi, Arabic, and Russian, this newer project will use other languages: Malay, Korean, Swahili, and Japanese. (Or just these 3: German, Portuguese, and Polish)
DESCRIPTION in succinct, accurate jargon:
Mnemonic compression for easier simultaneous multi-language receptive vocabulary acquisition. Now with "order = source language", for an added mnemonic aid. (Word length can be overcome with proper mnemonic technique combinations.)
DESCRIPTION in plain English:
Code was used to create a list of words with special properties. These words can help you memorize words from a few different languages at the same time, but only if your goal is to recognize words and get the gist of sentences, not necessarily to speak with exact grammar/correctness, if that's your goal for now.
Also, there's an older version of this project, but this newer project adds an extra memory help by having words follow a built-in rule so you can more easily figure out which source word belongs to which language. This comes at the cost of making words longer, but you don't have to worry about that if you use a good combination of memory techniques (Got questions on that? Ask in the "Issues" tab above. Basically choose words that sound similar but are of more concrete things, and then make mini sentence stories out of them that relate to the meaning---often in funny ways.).
More info:
Major difference: A tradeoff. Don't try to compress word length (which can be overcome with proper story mnemonics using word homophones). Instead, build in consistency to make tracking source languages easier. How? Always make source languages appear in the derived words in the same order. Example: the derived word/mnemonic "castahokanbi" preserves the order of the source words "ca", "esta", "ho", "kana", and "bi". But in cognateLanguage2, all words should consistently be like this in the generated vocabulary/mnemonic list, for you to be able to make use of the order of source words.
Similar to the original project, adjacent source words with sounds that are "allophones" of each other can be "combined" to make words slightly shorter. Example: ben + mul + kit + mon -> bemulkitmon, not benmulkitmon.
Pronunciation: Same as in the original "cognateLanguage" project: all letters follow their IPA values/pronunciations, except for "c"=/ʃ/, "j"=/ʒ/, "y"=/j/, and "h"=/x/.
Learn words:
https://www.memrise.com/course/1589462/coglang2/
Related projects:
https://github.com/hchiam/cogLang-geneticAlgo