It took some serious brainpower and patience, but I finally managed to scrape about 95% of my archived Forbes articles -- despite their dumb paywalls and viewing limits. I used a combination of nprofile1qy2hwumn8ghj7un9d3shjtnyd968gmewwp6kyqpq5c0uz3l55klsrq6l9zlvmtkvxwv2wy7q0rm7y249qkmzxpaj5gks7z60xv (nprofileโฆ60xv) and an open-source tool called Trafilatura (https://github.com/adbar/trafilatura).
So I was pretty thrilled about that! I'm gradually trying to build a personal library of EVERYTHING I've written or podcasted (personal and professional), and that was a big step in reaching that goal.