I was just watching PBS Eons about a housecat sized beaver like animal all the way back in the Mesozoic. The thought crossed my mind that it would be interesting if there was a Wikipedia type project for the Universe, Galaxy, Solar system, and Earth. I should be able to see how anything fits into the big picture timeline at a glance.
We should have a culture of nerds that extend beyond the written words of a Wiki. Forget the prestige of some elitist overpriced rag journal that acts as an outdated class and learning barrier in the present, only functioning as a makeshift poor quality reputation filter in the present. I want to see visual timelines where filling in some detail is an expected accomplishment within academia.
As I’m writing this I realized, this connects to how I perceive video based audio/visual information as the primary form of human communication. I think the next evolutionary step in teaching and studies is to forego the lecturer in favor of recorded media of empirically meritorious achievement compiled and tailored to each individual’s abilities. Along this line of thought, a scientific discovery should include an expectation to not only add to a public visual timeline archive, but to link the information in a way that teaches and connects the information to the world at large.
I guess I’m saying Science needs a cartographic like department/element where the figurative tree of branching knowledge is strengthened instead of independent niches on an ever expanding fractal edge of exploration. Publishing on the tree should be the driver of meritorious achievement instead of a redundant paper media subscription rag like the present. Reputation shouldn’t be limited to a few peers but instead showcased on a world stage that is as messy as reality but beautiful in aggregate. Publishing should involve an obligation to educate effectively where those that are limited in this skill are incentivised to add coauthors that are far more capable of effective communication. Persons with curiosity and time should be capable of freely navigating from what they know in the present to the messy edge of fringe science without any financial or circumstantial limitations in our digital age.
Does anything like this exist yet?
This should be quite easy to do (in principle). Every scientific paper has a doi, and any citation of said paper will create a directed backlink to that paper.
You can use this to build a connected graph of dois (nodes) bound by references (edges), and then use that as a basis for clustering (e.g. DBSCAN) which would naturally group papers by their topics.
To represent this in a 2D space you could do fancy embedding using some kind of distance metric between each doi, but you actually don’t need that if you know that one of your 2-axes is going to be time.
For less fancy embeddings, you can just feed the entire graph into graphviz and it will handle the rest.
What you’re describing is called a citation or reference tree and they are used to visualize complex set of references. Web of Science has a nice one, Scopus also has a tree view, I believe. Google Scholar has the information to make one, but doesn’t.