I do agree with you that mind maps have their flaws and limitations. But I found most curated list and most content in general to be quite rigid in structure. These mind maps should let the users explore and be guided through arrows.

And the big problem that I find is that google and other search engines are a black box. You have to know what you want to ask for to get an answer. The poses a big problem because many don't know what to ask.

Wikipedia is great too but I found it to lack in the visualisation aspect. Everything is connected but many people don't know exactly how and where and with what it is connected. These kinds of visualisations are very popular on the internet (https://github.com/kamranahmedse/developer-roadmap is one example) because they do bring clarity as to what it is the person doesn't know. However this repository although great, it just shows an image.

With these mind maps I can replicate the same but make it interactive. Now users can not only see what they don't know, but if they do want to learn it, they don't have to waste time online finding resources, they just click on a link and see what the best way of learning the topic is. If any questions arise when they are learning the content, then the search engine of their choice will answer these questions.

I also try to solve the problem of dependencies of knowledge. For example if user wants to learn 'machine learning'. I show that machine learning requires knowledge of statistics and linear algebra (http://i.imgur.com/A75JMNt.png) and then give links to the mind maps of both of these topics so user can learn them. There should be no wandering around and wasting time finding resources and being lost.

A suggestion for your product: build these dependency graphs automatically, and let users mark subjects as already satisfied. That way you can omit digging into dependencies that are satisfied. If you really want to be righteous, recursively mark the dependencies of the satisfied dependency as also satisfied. That way a user can pick a topic and find everything that still remains to learn to get to that point.

And you can have the best of both worlds. Generating a linear progression of things to learn becomes a simple matter of printing the dependency tree in postorder.

On a side note, I would subdivide Philosophy into Epistemology, Ethics, Naturalism, and Phenomenology and reclassify/reassociate accordingly.

If I'm learning about something or just debugging or designing then I'll use the mind map like a text file. I just type a paragraph and hit return. Each paragraph is a node but initially it's structurally no different than a text file - a list of paragraphs. The big advantage is that later I can take some of those nodes and put them under a sub-node. Its a great filtering tool, (filtering in the sense of signal to noise reduction). Maybe it's just because I am naturally so disorganised that I like it. Stuff comes out of my mind unstructured. I need a refactoring tool and I want one that's as lazy/easy as possible.

Some kind of an org-mode tool can work ok but I found it to be too slow to create links and embed images. Wikis are just too painful. At least in a mind map what you see is what you get and in fact the structure is right in your face (a bit like lisp).

This, however, requires that I draw the map myself by hand. Drawing it by hand forces me to carefully look at every pixel of the map. Adding some fancy fonts and pictures like in your example can help in the sense that you spend more time looking at the pixels.

And this is the secret. It doesn't much matter if you use tables or bullet lists or mind maps or whatever makes sense to you for the particular domain, it's the fact that you are organizing the info in your own brain that matters.

From a teacher-student perspective, mind maps seem to work best if the teacher presents the student with a partially-filled mind map and lets the students fill it in.

I find them to be great at brainstorming, for example. Dumping several related aspects of a project into the mind map feels very intuitive to me, no overhead involved at all. And it eases the cognitive load of several concepts floating around your subconscious and taking up your brain CPU.

For learning, I plain disagree with you. I even see your point that information may be better 'first presented' to a learner in other ways (although I've had good experiences with mind maps being a 'discovery' tools as well). However, learning has other moments other than the discovery. Mind maps are great at teaching how concepts fit together, and also good as a studying tool in the recall moment (rather than linear notes).

You may also view them as an 'index' to information presented elsewhere. Once a book was posted here on HN which had a 'graph of chapter dependencies' inside the book, which people loved. A graph of chapter dependencies is at least very close to a mind map.

By the way, since you mentioned UDC and DDC, even library science and information science courses do teach mind maps as tools as well. Because they have their uses, even if you know UDC or DDC.

Given how rapidly x+y values for mind-maps increase alongside content complexity for any one 'abstraction level'; constraints like screen space and uni-dimensional view-points for hierarchies come into play. It would be nice to easily see how core skills connect to different disciplines, say. Then again, side-stepping this by using different kinds of connections/views for nodes across levels may have an exponential maintenance curve. It reminds me of that old/strange hyper-connected theory for Internet infrastructure I can't remember the name of.

I'm keen to think of new methods for map-building since I find them personally useful, but I wonder if I suffer from the hammer-nail bias.

But... I also agree that everyone learns and parses information differently.