Internet Is Fun 
The internet is supposed to be permanent. The reality is that an enormous amount of what was published online is simply gone, and more of it goes every year.
Lost media is the hobby, and increasingly the discipline, of finding it again.
What counts as lost
The term covers a wide range: a television episode that was broadcast once and never rerun, a video pulled from a platform, a game demo that shipped on a magazine disc, an advert everyone remembers and nobody can produce.
The common thread is that it demonstrably existed and no accessible copy is currently known. Some of it is genuinely destroyed. A lot is sitting on a hard drive in someone’s loft, unlabelled.
Why digital things vanish
People assume physical media is fragile and digital media is safe. It is closer to the reverse.
- Hosting is rented. When a company folds or a service is retired, the files usually go with it.
- Formats stop being readable. Flash is the obvious case: an entire era of web games and animation became unplayable almost overnight.
- Rights cause deletions. A music claim can remove a video permanently, along with the only copy.
- Nobody chose to keep it. Most of what disappears was simply never anyone’s job to preserve.
How things actually get found
Recoveries rarely start with the file. They start with evidence that it existed: a review, a screenshot, a forum post, a listing in an old schedule. From there it becomes detective work.
Communities like the Lost Media Wiki and the subreddit built around it run on exactly this: someone remembers, someone else finds a trace, and eventually a person who taped it in 1997 notices the thread. Most successful recoveries come from one individual who happened to keep something.
The Flash case, and what fixed it
When browser support for Flash ended, a very large body of games and animations stopped working. The reason a lot of it survives is that volunteers spent years downloading and cataloguing files before the deadline, and built the emulators that run them now.
It is the clearest example of the pattern: nothing was preserved automatically. People did it, unpaid, because they cared.
What you can actually do
- Save local copies of things that matter to you, rather than trusting a link
- Submit pages you care about to a web archive, which takes seconds
- Do not delete the old drive. Label it instead
- If you remember something clearly and cannot find it, write down what you remember somewhere public. That is often the first evidence a search needs
Our running lost media database collects the main communities and archives. If you want the adjacent territory, the Wikipedia iceberg and the rest of Rabbit Holes will take the afternoon.
