How many academic citations can we replace with import statements?
If we take reproducible research in the sciences seriously, the paper is a human-readable adjunct to the code and data.
The main conclusions of many scientific papers should be expressible as equations, API calls, or entries in a probabilistic knowledge graph (“A binds to B”).
In the same way that good code can be often used without reading the manual, a good scientific paper can often be boiled down to a few key assertions.
CS is the most obvious place to start. Each journal a code repository, each paper a library?
from nips import smith.2017
Academics spend a lot of time on citations. What if all these reference managers like @mendeley_com and @citeulike could be turned into software libraries?
The next version of BibTex might treat citations as code, not text.
The big thing about any idea like this is making the authoring flow as easy as it currently is, and ideally easier.
For example, you could take something like JANE (jane.biosemantics.org) and run it over all the text in your buffer to auto-suggest references. Or libraries.
Share this Scrolly Tale with your friends.
A Scrolly Tale is a new way to read Twitter threads with a more visually immersive experience.
Discover more beautiful Scrolly Tales like this.