I actually made https://hn-blogs.kronis.dev https://hn-blogs.kronis.dev a while ago, which was based on RSS/Atom feeds that people shared on HN, it's still runn

I actually made https://hn-blogs.kronis.dev https://hn-blogs.kronis.dev a while ago, which was based on RSS/Atom feeds that people shared on HN, it's still running in the background and pulls in the latest posts daily. There are links to the original HN post and a blog post that I made about it all on the site, but here's the post directly: https://blog.kronis.dev/articles/ever-wanted-to-read-thousands-of-tech-blogs-now-you-can https://blog.kronis.dev/articles/ever-wanted-to-read-thousan... It was certainly interesting to do and I love that the technology exists, however a lot of people handle integrating it differently. Sometimes the feed type is misreported, other times you can't quite get all of the meta data about the feed or items you'd like, whereas sometimes you either have the whole thing just throw a network error, or just have someone with control sequences in the text which breaks your attempts at parsing XML. I'm convinced that the larger the scale of your dataset, the more potential issues you're going to run into, at some point if any type of error can occur, you're going to have to deal with it.