Web permanence and URL rot: still no good solution
A thread about a 2011 Long Bets prediction whose original URL is expiring sparked a conversation about the fundamental problem of URL permanence. Commenters noted that the solution to link rot is not technical discipline but economic incentives, and nobody has cracked it. One commenter brought up Freenet's approach from the early 2000s as a philosophical alternative. Another noted that the Disqus comments surviving was more impressive than the page itself.
The pattern here is that the web's memory problem is structural. URLs are controlled by whoever pays for the hosting. When that person or organization disappears or loses interest, the link dies. Archive.org is the de facto solution and it is a nonprofit running on donations.
This matters practically for anyone building products that link to external content, write documentation, or depend on third-party resources staying live.
So what?
If your product depends on external URLs staying alive (documentation, citations, legal references, training data), build in redundancy now. Archiving, local caching, or content-addressed storage are not overengineering in this context. The bet that a URL will still work in 10 years is a bad bet.