Co-creator of Metacademy here. Colorado and I are intending Metacademy to be a “package manager for knowledge,” where you can easily find the particular thing you want to learn about (e.g. deep neural nets) without having to track down all the prerequisites (e.g. gradient descent) yourself. Basically, we’ve annotated a dependency graph for the core concepts in the field, and it uses the graph to produce step-by-step learning plans geared directly towards your goals.

Sorry if you can't reach the site. We’re both doing this as volunteers, and I guess we weren’t prepared for the level of traffic we’re getting. In the meantime, you can find more details about our high-level goals here: http://hunch.net/?p=2714

If you don’t have a particular goal in mind and just want a general overview, check out the roadmaps, e.g. http://www.metacademy.org/roadmaps/rgrosse/bayesian_machine_...

IMHO, the dependency graph is pure gold! I think it should be the core focus of this project as it will be quite difficult to aggregate, clean and present all the needed information about a given topic (unless we're rebuilding wikipedia). It will be easier to link to credible resources available across the web through the dependency graph. Maintenance wise it would be much more interesting not to mention easier.

Nice work! I've been working on something very similar. My project still needs a lot of work (and talent, I started working on this site when I first learned html/css/js 2 years ago) but you can check it out at: http://subjectflow.com , or for a full tree: http://www.subjectflow.com/viewer.htm?=mewvzl2i6bt97gnjf9i4 (click on the white dots to expand the tree).

I really hope to see this type of idea take off. I would love to see this sort of thing combined with simplified git related operations to make it easy to improve education content. Additionally I think it would be great to add in a comment system where you can leave a question connected to the exact text that is confusing you.

Been clicking around for 20 minutes. This is like TV Tropes for knowledge.

How is "x hours to learn" measured? If I search for logistic regression, I get this page: http://www.metacademy.org/graphs/concepts/logistic_regressio... stating 1.7 hours - is that the estimated time to work through "Core resources", the "read" time + video length or something else?

Really exciting project, and very useful to me personally.

This is what wikiveristy should've been.

Good job guys.

this seem like an interesting idea, but it lacks some "high level structure", or possibly a better search.

For example, I am sure there are general topics for "classification", "recommendation", or "similarity", but I wouldn't know where to start.

Is there a way to view the DAG of concepts itself?

NB: Metacademy is open source: https://github.com/metacademy/metacademy-application

The site is not accessible (not responding to http requests)

I like the approach! I feel that search could use an improvement. I'm using k-means++ at the moment for something and searched for "k-means" which yield no results [1]. I clicked on the link to look at the full-list and found k-means mentioned 3 times, to my surprise.

[1] http://www.metacademy.org/search?q=k-means

Nice! This is the first time I've seen the dependency graph implemented in e-learning! Actually I started a programming site in spanish a while ago with the same idea (there is a dependency graph underlying the content). If you are curious: http://www.minidosis.org.

I think this is fantastic. Obviously stuff to be ironed out, tidied up and better organised but genuinely think this is a great idea, how does one go about developing something like the dependency graph? Would love to implement it in a site I am building while learning Rails.

This deserves to exist and flourish. We built something similar (skilldom.org) but it hasn't yet deserved the love it should have... :) partly due to me not being disciplined enough to dedicate time to it.

The DAG of concepts is brilliant. I've manually made my own DAG for interesting branches of mathematics over the years and used it as a guide for self study. This could be big for education.

Metacademy looks like a great source for learning, one which preserves most of the associativity benefit of learning through e.g. wikipedia, but which allows for much more focus.

Thank you so much for doing this!

love this. i've wanted something like this for a long time. Kahn Academy mind maps are awesome but limited in terms of subject matter. having an open database of this is amazing; enables self learning while eliminating the problem of not knowing what you don't know.

I've been using this. It's extremely useful. I wish they did this for every subject.

This is awesome. I can see this, if it gets larger, being a ton of help throughout college.

Nice way to organize major sources for learning Machine learning topics.

This is a dream, thank you.

This is really cool. I wish I could give it more upvotes than the one I gave it.

Thanks for building something cool and useful. I love the idea and hope you develop it more.

HFT?