Documentation is easier. Tools for PDF, Markdown have increased in efficacy. Coding alone has lowered the barrier to bringing building blocks and some understanding to the masses.
I have seen none of these, in practice.
The documentation generated is no better than what a level 1 support rep creates, and needs to be heavily fixed before being relied on.
Pandoc still produces PDFs, Markdown, etc just as quickly as it always has.
The code produced still has the same issues as documentation: it’s shite, and not easily bug fixed due to a lack of understanding by anyone with what its actually doing. And, if you need someone who understand the code already to bugfix it, guess what? You didn’t save anyone anything.
And, all of this, only using terrawatts more electricity than before, with equivalent or worse outcomes.
I have seen none of these, in practice.
The documentation generated is no better than what a level 1 support rep creates, and needs to be heavily fixed before being relied on.
Pandoc still produces PDFs, Markdown, etc just as quickly as it always has.
The code produced still has the same issues as documentation: it’s shite, and not easily bug fixed due to a lack of understanding by anyone with what its actually doing. And, if you need someone who understand the code already to bugfix it, guess what? You didn’t save anyone anything.
And, all of this, only using terrawatts more electricity than before, with equivalent or worse outcomes.
OCR was more my thinking, not Pandoc. LLMs enable OCR to achieve greater accuracy through context enhancement for example.
That sounds like one of those rare appropriate use cases.