I long to see the day when laws and other civic documents are compiled and evolved in some source control like git or something domain specific for such documents.
It seemed to work fine to record incremental changes, tag specific revisions as being checkpoints for various stages of review and approvals. And anyone could search current and historical versions.
The overall legislation system needs flexibility to deal with complex and changing workflows, but storing documents (with change over time) and the links between them can be encoded just fine. As long as there are flexible ways to string them together, they can be represented in whatever structure makes sense: final law with all current amendments; original law, with amendments as annotations, separate pieces of legislation, linked by the overall law, etc. None of that has to be encoded in a super-strict heirarchy, but as long as it's all being stored in can be presented however one wants.
I don't see the fatal flaw. I see imperfections and compromises, but not the flaw so bad as to say "don't bother".
Do you mind sharing with us the company and the product you developed in this space?
Untrustworthy laws are rather more foolish than doing revision tracking on English text. I mean... FFS how hard is it to track changes to English text? Really?
It's not for software, it's for text. Is the law no longer written in text?
The procedure is similar to they way git is used for code though with a greater or lesser amount of automation - in the old days you had multiple documents and you physically cut and pasted them together to produce the final document