Flashcards for your terminal
newspekidemo.2.mp4
It will probably be flagged as dangerous by windows
Package is simply named "speki" in the aur. Can be found here.
if you're using yay:
yay -S spekiClick to download binary You'll have to add it to your PATH manually for easy access. I'll look into how to to package it for various package managers, if you have experience with that then I'd appreciate help.
For something that will work on any distro as long as you have cargo installed, try this:
cargo install spekiNote that this won't install all the necessary dependencies, for example look at the note on ubuntu right below. You can get cargo by installing rust
If you install with Cargo, you will also need to install the following package: libasound2-dev
If you install with Cargo, you will also need to install the following package: alsa-lib-devel
--> sudo dnf install alsa-lib-devel
It's not just about having flashcards in your terminal
Speki has integrated the shared decks from anki, meaning you can jump straight in to learning different things, all without even leaving the app. It also allows you to customize the templates before importing. Soon I will add the possibility to add your local anki-files so that you'll get your anki-cards together with their learning-history to Speki.
Have a question but you're not able or don't have the time to find the answer to it? You can add unfinished cards in Speki, and be periodically asked if you're able to find the answer to them. No more needing a separate note-taking system for things you want to turn into flashcards.
You have the answer to your question, but you don't understand the answer? You can add the card as finished, but add dependencies to it, which would be other cards you need to understand for the current card to make sense.
For example, if you read this and you want to memorize it:
Alpha particles consist of two protons and two neutrons
Let's pretend you don't know what protons or neutrons are. If you were to memorize this sentence in that case, the knowledge would be completely useless to you. In speki however, you can add the two following cards as dependencies:
What is a proton?
What is a neutron?
If you add them as dependencies, the original card will be marked as unresolved, meaning it has unfinished dependencies. When it is unresolved, you will not be asked to answer it, but you'll be asked to learn what a proton and neutron is, when you've figured out their meanings, the alpha particles card will become resolved and you can review it as normal. This system ensures that no matter how complicated things you add, you'll never end up memorizing something you don't understand, as long as you mark your dependencies correctly.
It also works recursively, meaning in the previous example, if you found the answer to "what is a proton" but didn't understand it, that card would itself have dependencies. The dependency-graph could go on indefinitively, but if you keep on the dependencies would eventually all resolve and you'd finally learn the top-most card. This closely mirrors how your brain works, as everything you know is dependent on other pieces of knowledge. It also eliminates the need for making mind maps, as this is basically a more accurate type of mind maps.
Incremental reading is a concept originating in SuperMemo. It allows you to incrementally work your way through countless longform-texts in parallel, where you'll learn and memorize everything you want from it. This is done by by being presented with the texts within speki, and every time you see an interesting piece, you take extracts from it. As if you were going through a real book and using a highlighter. Later you'll be presented with just the extract, and here you can choose to make further extracts, or directly making cloze deletions by marking the things you want to remember. When you read a non-fiction book, not long after you won't remember much from it. If you do it through incremental reading, you'll remember every single thing you choose to remember, albeit with some extra efforts.
If you want to read more about incremental reading, you can check out this blog article, as well as many other articles from the same site.
Got a key to OpenAI? Then you can autogenerate the answers to questions
openaispekigithub.mp4
Simply put your key in the config file, which for linux users is in ~/.config/speki/config.toml, and for windows users in C:\Users\speki\config.toml
switch between tabs tab/shift+tab
switch between widgets: Alt+(h|j|k|l)
quit: Alt+q
toggle help menu: F1
generate answer (if gpt3 key in config) Alt+g
skip unfinished card Alt+s
mark unfinished card as finished Alt+f
suspend card Alt+i
mark incremental text as done Alt+d
skip incremental text Alt+s
rate recall grade (worst..best) 1..4
add new card as dependency Alt+Y
add existing card as dependency Alt+y
add new card as dependent Alt+T
add existing card as dependent Alt+t
generate answer (if gpt3 key in config) Alt+g
add card as finished Alt+f
add card as unfinished Alt+u
add wikipedia page (from sources or extracts list) Alt+w
download deck that you've revealed description of Enter
switch front and back template Alt+s
change which card to preview (on preview button) h/l
insert-mode -> normal-mode Ctrl+c
normal-mode -> insert-mode i
normal-mode -> visual-mode v
visual-mode -> normal-mode Ctrl+c
page-up Ctrl+u
page-down Ctrl+d
extract (from visual mode) Alt+x
cloze-deletion (from visual mode) Alt+z
delete right of line (from command mode) D
move cursor up/down k/j
move topic up/down K/J
delete topic Del
edit topic name e
add new child topic a
move topic up the hiarchy h
move topic below topic under l
Pending card Card has not yet been activated
Unfinished card Card does not have an answer
Finished card Card has a proper answer
Unresolved card One or more dependencies of the card is either unfinished or unresolved
Incremental reading A way of converting longform texts into flashcards
Suspended card Card won't show up in the learning process
- Add tab where you can browse and filter all cards and perform actions on them
- Add image suport for terminals that enable this
- Machine-Learning algorithm for spaced repetition
- Visualize dependency graph
I'm very open to any feedback or PRs, feel free to open an issue if there's anything you're wondering, and i'll make sure to get back to you!