Open leventov opened 3 years ago
I'm not really sure what this is asking for. Could you explain a bit more about what this is and how it's useful/how it would work for us?
More about Orbit: https://withorbit.com/
Basically, this is "modern, web-based Anki for everyone". Integrated with paper pages, it can help people to retain what they have read, rather than forget soon after reading.
I'm not sure what you mean by paper pages here. Do you mean scientific papers? That's not really the kind of content our network is geared towards - only one site has made an attempt at things like that. It sounds like other topics might not fit this quite as well.
Two communities, I think -- EE has papers and I think Physics was talking about them. I wouldn't be surprised if other science communities someday go down that path too.
If I understand correctly, the request here is for an affordance that allows people to use Orbit to annotate content for themselves. Is that right? And these annotations are strictly client-side, not shared, and what you need from Codidact is a hook? Or do you need a hosted Orbit server as well? That's a bigger request.
Anki is a flash card learning system based on a spaced repetition algorithm that basically controls the intervals where flash cards are shown and is known to be very time efficient. A flash card is basically a short question and a (relatively) short answer, so a bit similar to our Q&A but with some differences: for flash cards there is typically only a single answer and question and answer are much, much shorter (more like a summary here). A single good Q&A on Codidact can and should be divided up into many good flash cards.
And that is the big problem for all these systems (Orbit probably included). You need to do major work in converting the knowledge into very small pieces. This work would be needed to be done by somebody. One would need to go through articles, papers, Q&A and create short flashcard question and answers.
If I understand Orbit from their homepage correctly you can basically add such quick questions and answers to a text and then in between reading or even outside of it, Orbit will occasionally ask you these as kind of reinforced learning. One could see it as a spaced repetition memorization algorithm for arbitrary texts. I imagine a Codidact user one could possible select such annotated content of Codidact for "learning" and then have a Orbit section on the right side that keeps asking you short question and if you don't know the answers, redirects you to the original content with more extensive information. It would be a tool for revising the information stored in Q&A.
Not sure if this is really that useful. Typically I rely on search engines to bring me back to the knowledge sources, and I don't mind googling the solution to a problem multiple times.
But the general topic is, what can we do with the Q&A/content of Codidact after they have been created apart from just having a collection of content that is searchable. Specifically, how can one use that collection for learning.
Would be nice to add text like:
Into Orbit prompts. On the Codidact network, the Orbit server might be hosted by the Codidact network itself.