Various experiments relating to information extraction.
Extract text formatted as markdown from webpages using headless chrome and the accessibility tree. This is far more accurate for generic parsing as all websites that have decent SEO scores should be able to be parsed this way.