We had the great idea of translating, or better said encoding our long string of meaningless characters into meaningful words. But how’s that possible?
Let’s break it down step by step. The original key was something like this:
0011100000111000110001011110001001110001001001111111110000000000101101000000010101111110001101101101101001001111110000010100111100011001000000010000101101000101111100110101101101101001100001111101110011101000111101001101111111111000101111101111101001111101111010111111000100111101001000001001001000111111111111111000001001000000
Why is it pure zeros and ones? Well it’s the language that computers actually understand. Binary. things are either on or off. Ones or zeros. But notice how long that binary key is. It’s 256 numbers long! That’s what happens when you can only use 2 different characters, a 1 or a 0. You have to use a lot of them to write down a moderate piece of information.
To make it easier for us humans to check and read those keys, we could make it shorter for a start. To compress that key quite a bit we could use more characters than just 2. We could use 58 for example:
123456789ABCDEFGHJKLMNPQRSTUVWXYZabcdefghijkmnopqrstuvwxyz
That’s called Base58. That’s a set of 58 different characters that are hopefully distinct enough between them to reduce mistakes. For example, depending on the font you’re using, a 1 and a l could be almost identical. So, let’s convert our binary key into Base58:
378bc5e22fc9fc02b41af8dd943e794f70576634b6223f3eba7ea76efe2c4400
This is pretty good, way shorter. It’s a good thing we humans can read and write more characters than computers do. But still not super practical for human handwritten backup copies, or eyeballing differences. They’re still 64 meaningless characters. What if we could compress it even more, so that we only have 25, plus they made sense to us? What if we could use an alphabet we’re very good at using daily, so we can write it down, read back, and spot any mistakes with ease? Good idea. But unless we’re using Chinese, we have no more characters. To solve the compression problem we can use an alphabet of more than 1200 different characters, and to solve the readability problem we can use words instead of single characters.
We need to agree on a shared dictionary, a list of the words we can use, and then try to compress our key again:
nouns older lids rodent pumpkins railway wiggle mailed hoax waking evenings nocturnal oozed buckets germs gearbox gauze neon update axis macro syndrome madness magically rodent
Oh yeah. Now we’re talking. It may look longer at first sight, but now instead of 64 different “items” that could be right or wrong, we just have 25. And errors are pretty easy to spot as well. Even if the writing is pretty bad, we can guess it says evenings and not evonings.
That’s the reason we use mnemonic seeds. Mnemonic refers to something being easier to remember. Now we have easier to remember seeds that can be automatically translated into computer keys to prove ownership of our beautiful digital money.
From them we can create addresses for people to receive and send money from and to each other.