References inside of the PDF will look much better if they are processed by LaTeX. While there are existing hacks to integrate bibtex into Markdown, none of these would work directly with Github Pages. So we need to find another way to bridge the gap.
Right now our standard is just to write things like [1] in the article and imply that this signifies a reference in a numbered list at the end of the section.
One idea would be it to instead use a format like [Corneli2020a] and then use these same tokens as the keys in our bibtex bibliography. We could put them into a bullet list at the end rather than a numbered list.
After conversion to LaTeX we would want to remove the References section so that the references can be regenerated natively.
We would alsol need a little bit of post-processing to scan for strings like {[}Corneli2020a{]} (which is what pandoc will produce) and convert them to \cite{Corneli2020a}.
References inside of the PDF will look much better if they are processed by LaTeX. While there are existing hacks to integrate bibtex into Markdown, none of these would work directly with Github Pages. So we need to find another way to bridge the gap.
Right now our standard is just to write things like [1] in the article and imply that this signifies a reference in a numbered list at the end of the section.
One idea would be it to instead use a format like [Corneli2020a] and then use these same tokens as the keys in our bibtex bibliography. We could put them into a bullet list at the end rather than a numbered list.
After conversion to LaTeX we would want to remove the References section so that the references can be regenerated natively.
We would alsol need a little bit of post-processing to scan for strings like
{[}Corneli2020a{]}(which is what pandoc will produce) and convert them to\cite{Corneli2020a}.