Hittite was putting spaces between words in the 17th century BCE. And if we're just interested in Latin, it used the interpunct as a word divider hundreds of years before the use of the space as word divider happened. The use of scriptio continua despite knowledge of word dividers was a choice.

I wonder, how much was gatekeeping, keeping things hard on purpose, how much waas inertia, "that's just the way things are done", and how much was a kind of despairing "holy shit, it'd be so much work to have to go through and recopy everything in the new format, literally decades of effort, and there's other things we want to do with our lives".

The whole context of written words had so much implicit process and knowledge and institutional memory, compared to now when we have petabytes of throwaway logs and trivial scratchpads for software running on a "just in case I might need to figure something out" basis. I'd love to see a written word graph over time, starting ~4k BC to now. And the complexity and diversity of those automated words are going up like crazy since LLMs.