You seem to be looking for a research assistant, phd student, etc

The key thing is not discipline. It's to make it yours. It's easy to make a software that would link your thoughts or weblinks together one way or another, but it's impossible to make it just like you would.

There is no software that would piece different ideas together to come up with a new one. That is called AI and we are not there yet.

What you want is get the results without doing any actual work yourself. This is not a problem that any software could fix.

Concur, to rephrase it, it's not the software that must make connections while you sleep, but rather you must make connections and then transfer them to software. From there you can do spaced repetition or whatever to map those connections to a mental model. There are two friction points here: realizing what connections are made and then externally documenting the connections.

Why do you need to document the connections externally? Isn't it enough to reference one node from another and provide some context?

> dump interesting thoughts (or links to articles, videos, whatever) into a "knowledge base" and it finds connections and labels things for me while I sleep. ¶ Who's building this?

Mek from archive.org has recently decided to give it a shot.

https://www.youtube.com/watch?v=Q3W7rU_xKKc

> Even the simplest organizational schemes require a great deal of _discipline_ to be successful with

Which I think means: all that's needed is discipline. Which organizational scheme you use (if any) is almost irrelevant.

The useful question is then: which methods help building discipline?

Perhaps, but you could say the same thing about something like an `if` statement in the era of branch assembly instructions. One requires more discipline (and knowledge) but that doesn't mean the problem is solely a lack of willpower/discipline.

I'd love something like nvAlt that has zero hierarchy, but instead offloads to blazing-fast search. The problem with nvAlt is discovery so bolting on some smart filters like LRU, MRU that you can access through commands in the search bar plus a roam-style graph would help you move through the note space.

Then when you open a note, a side panel with back-links opens side by side with the note. Press a key command to flick between the search bar and the note, and update the side-panel whenever a new note is opened. This would be the perfect app for me.

you should check out my (open source) project https://mmap.io. It's a zero hierarchy, short cut driven note tool. Would love to hear your thoughts!

404 FYI :(

I guess they meant https://mmap.it which is open sourced at https://github.com/pdepip/mmap.it (matches their username).

This has always been the premise/promise of DevonThink, but I never found it quite hit the spot for me – I end up imposing organisation on it, and then we're back at your original complaint. Some users do absolutely swear by it, though.

You and me both. My thoughts on this subject are that all knowledge is related in some manner, so unless you want your knowledge points to form an all to all connection (useless), there needs to be some probabilistic connection.

This means that the user blanket accepts the probability that language is similar enough in two separate thoughts to form a connection, or they receive a basket of suggestions from a program and approve/disapprove each. One inevitably leads to meaningless connections, and the other leads to work on the user's part to prune out all the trash.

What do you think?

> I'm looking forward to the day when I can dump interesting thoughts (or links to articles, videos, whatever) into a "knowledge base" and it finds connections and labels things for me while I sleep. > > Who's building this?

I do and so do others. But this is not magic. Computers can't read your mind, they can't create perfect connections without the users guidance. A certain level of discipline will always be necessary. At the end you are the only one who knows what you know and need at a specific moment, any tool can only assist you in this.

What you're describing is essentially a recommendation system with an understanding of natural language. There's a lot of research going into it, but it's still a very difficult problem to solve.

I'm working on an app to solve the taxonomy problem at least! it's currently in very early development in a private repo but I'll make it public at https://github.com/pxeger/tome at some point

Eastgate Tinderbox has been doing this for 20 years.

Mark Berenstein is one of the original Hypertext pioneers and has built a fantastic tool for notes.

http://www.eastgate.com/Tinderbox/

This and DevonThink are the only reasons I remain on a Mac.

> dump interesting thoughts (or links to articles, videos, whatever) into a "knowledge base" and it finds connections and labels things for me while I sleep.

I think you just described dreaming.

This.