The US Library of Congress operates the THOMAS system for legislative bills. How could THOMAS be improved; in order to support evidence-based policy; in order to improve democracy in democratic republics like the US? https://en.wikipedia.org/wiki/THOMAS
How do state systems for legislative bills document workflows and beyond compare to THOMAS, the US federal Congress system? Why do we need 50+1 independent [open source?] software applications; could states work together on such essential systems?
IIRC, LOC invested in improving THOMAS with: markup language to typographically-readable HTML support?
It may be helpful to send an email to your state with the regex regular expression pattern necessary to make online statute section and subsection references a href links instead of non-clickable string. TIL the section sign ("§") is older than the hyperlink; https://en.wikipedia.org/wiki/Section_sign
Parliamentary informatics: https://en.wikipedia.org/wiki/Parliamentary_informatics
FWIU, diffing documents is a solved problem.
How could pull request workflows be improved in order to support legislative and post-legislative workflows (like controlling for whether a funded plan actually has the proscribed impact)?
High School and College Policy debate (CX; Cross Examination debate) enjoy equal rights to 'fiat'. IRL, we must control for whether the plan and funding have the predicted impact.
E.g. GitHub and GitLab have various features that could support legislative workflows: code owners, multiple approvals required to merge, emoji reactions on Issues and Pull Requests and comments therein such that you don't have to say why you voted a particular way, GPG-signatures required for commit and merge, 2FA.
FWIU, the Aragorn / district0x projects have some of the first DLT-based (Blockchain) systems for democracy; all of the rest of us trust one party to share root access and multi-site backup responsibilities and have insufficient DDOS protections.
Any such legislative solution may involve meeting the users where they are, first. Understanding the whole lifecycle, so to speak.