Finally finished up a bot for posting an article from Suck.com everyday through Mastodon and BlueSky (I would have done Threads too, but I can't seem to get Threads to give me a working API access token).
You are probably too young to know Suck.com. It was a wonderful, and incredibly influential, early web magazine/blog/newsletter that started publishing in 1995 and closed in June 2001. It had wonderful writers and illustrators, an often funny but always intelligent point of view, and did a great job skewering the earlyish Internet business world that I was a part of.
Andy Baio wrote a good requiem for Suck that includes this list of other summaries:
"For retrospectives on Suck’s role in early web history, these links do a good job:
- Josh Quittner’s 1996 Wired article about the site’s origins.
- Keep Going’s “The Big Fish,” a 2005 retrospective about the impact of “the first great website”
- Engadget’s Nicole Lee interviewed the Suck crew after their appearance at XOXO 2015.
- The Internet History Podcast with Suck founders Carl Steadman and Joey Anuff, and the followup episode.
- In 2016, The Atlantic’s Anna Wiener wrote about “the best magazine on the early web.”
- In 2017, The History of the Web covered “The Web After Suck” in its second post."
I miss it and wanted more excuses to take a look at it. I also generally wanted to see if there was a good way of surfacing old content related to today's news because I generally think there is a TON of great content out there that no longer gets seen. I would use the Suck.com archive to try out matching an article per day from that archive with today's news.
To build it, first I crawled the Internet Archive's collection of old Suck articles. Then I used Claude Code to write a tool to select one archived Suck.com article each day based on its relevance to current news. It works by converting both the archived articles and today's news headlines into numerical representations (vectors) that capture their meaning. Articles about similar topics end up with similar vectors, so the system can find which old Suck.com piece best matches what's happening in the news today. This approach finds conceptual connections. For example, a Suck article on media hype from 2000 might match a news story about viral misinformation, even if they share no exact words.
The system also tracks which articles have been shown and won't repeat any until the entire archive has been cycled through.
The accounts are: @barrelofsuck@mastodon.social and @barrelofsuck.bsky.social
More about my general Claude Code approach is here. This is yet another example of Claude Code giving me the ability to do a little something with very limited audience that I've always wanted.
If you know of a good way to make the Threads api work and are willing to talk me through it, lmk.
No comments:
Post a Comment