Vocab Management System

Reflecting on Reviewing

Part of learning all of this really disparate vocabulary is that I have no larger organizing framework to put in, hence the risk of me not remembering some words increases. It's also true that I have an easier time producing signs that I learn in Vicars lectures (harder time with reception), but an easier time receiving signs that I learn in FB reels (harder time with production). I do think the FB Reels signs do tend to be a little harder in general though...

Perhaps I should take some time to organize the existing signs that I've learned so far into a TP-style framework.

akesi

Snake

alasa

Research, Experiment

ante

Conflicts

esun

Cost, Corporation, Change-out, Deserve, Bet

ike

Human Trafficking, Kidnap

ilo

Telescope, Printer, Lock, Metro, Binder, Backpack, Balloon, Belt, Degree, Gas-empty, Gas-full

jan

Pirate, Leprechaun, Human

ken

Chance, License

kili

Potato, Strawberry, Melon, Corn, Garlic, Lettuce, Cherry

ko

Smoothie, Jelly

kon

Tornado, Inspiration

lili

Blah-blah-blah

luka

Fingerprints

lukin

Attention, Narrowed-Attention, Visually-take-it-all-in

ma

Mexico, Australia, Europe, England, Boston, Ireland-International, Ireland-Iowa

mani

Welfare, Medicare

moku

Mashed-Potato

mun

Space, Galaxy

mute

Tablespoon, Pound, Cup-(Measurement)

namako

Vinegar

nasa

Mischevious, Alcohol

nasin

Tradition, Repulican, Planner, Prepare, Conservative, Democrat, Liberal, Leadership, Government

pakala

Trauma, Darn-it, Fake, Jammed, Lose-Fail, Dimwit

pan

Pie

pana

Spread,

pilin

Anxiety, Empathy, Concerned-about

pona

Kindness, Lucky, Official, Value

sama

Identical

sijelo

In-person

sona

Shared-Understanding

soweli

Donkey, Bacon, Sheep

suli

In-your-face, Boringly-prolong

taso

Flat

tawa

Twirl, Travel-Whirlwind, Transfer-(public-transport), Skating, Figure Skating, Ballet

telo

Sweat,

tenpo

Three-days-ago, St Patrick's day, Honeymoon

toki

Telepathy, Poetry, Exaggerate, Dwell-on

tomo

Bar

utala

Rebel, Resist

wile

Willing, Goal

I actually really like this. This is going to be really helpful for me.

New Systems?

I'm going to need to find a more robust system to keep all my vocab in order, and it's looking like a spreadsheet of some kind is likely going to be the best way to do so.

In a completely dreamlike spreadsheet, it would be automated pretty naturally from the entries I have here. It would have the date that I learned it, the source that I learned it from, a screenshot of the general sign, the toki pona word it can be most associated with, and a place for additional notes. And it has to be sustainable, something that I can keep up with not much additional work.

Yesterday I already spent some time making a script that goes and yoinks all of the vocab from all of the lists with a level 4 heading called 'Vocab.'

It's likely possible that I can write something that goes through and zoinks all the vocab and the vids they are associated with and shoves it in a spreadsheet??? But then I'm offshoring the enterprise. HMM.

HMMM.

HMMMMMMMM.

Potentially, we're now talking about-- HMMM.

You might be like, Chelsea, stop thinking about this and just learn the language already. But you have to understand- I have lost so so many signs that I've learned in the past. I really do feel that if I put some extra investment into creating a better system now, I will be generally more able to retain things throughout these next couple of years of my life.

Okay. I think I have the beginnings of a plan.

Brainstorm

Currently Eleventy treats vocab as an object with both a word and a date (the date being in the format of YYYY-MM) that is associated with a particular post. It generates this automatically based on the presence of a heading level 4 titled 'Vocab'.

It then uses this within Eleventy Collections to generate this index...

I could instead have it write directly to file...

I could also have it scan for notes for glosses that match the assigned gloss.

And then I could add any tp classes I want later, to the raw files. I could then photo capture every sign into one large folder and match by gloss name... HMMM.

But how do I get it to not do duplicate work? I only want it to do this kind of processing like... once. I also want it to be able to take in just straight up words, in case I learn stuff in the wild that isn't associated with a particular material.

A New System!

So I didn't change anything about the way I have my Eleventy set up. I did use Claude to make a python script that will extract vocab from a markdown file I give it along with various associated metadata and send it to a Google Sheet.

I then manually went back and screencapped the vocab and things. Obviously, the screencaps are not CC-BY.

So I can run the script and it will update my sheet, as I want to. Doing this whole thing was good review. I actually culled a few vocabulary items that were likely just misreadings, and fixed a broken link.

This is a very good and robust set up to start out with. In the future, this will help me to make an Anki deck, or something resemmbling an Anki deck.

Here's what I have to remember to do in the future:

To execute save to Google Sheets:

cd /home/lakuse/VSCodium/asl-sheet-vocab
source venv/bin/activate
python add_vocab.py post-X.md

To exit virtual machine:

deactivate

I also have to remember to NOT leave the table sorted by anything else by the default. I unfortunately didn't work in some kind of indexing system so if I want to preserve the order of signs in each vocab list, I have to maintain the order they are loaded in.

But while I am mid study session, I can totally go ahead and sort the table to accommodate different needs- alphabetical, by tp word, etc.

The downside is that I require the internet to access this sheet, which is why the Eleventy system needs to stay here as back-up.