#opendirectoryproject — Public Fediverse posts
Live and recent posts from across the Fediverse tagged #opendirectoryproject, aggregated by home.social.
-
The Open Directory Project was a hierarchical tree of topics, which had in it websites that had been deemed credible by volunteers. There was sort of a Wikipedia model for how volunteers were managed and a webforum.
The site itself was made using the Unix file system as a database. Links to other parts of the tree were symbolic links. The links off to the end pages were little files.
This all ran via a set of small perl scripts. It was small and quick.
It was not open source, but it was open data. Once a week, the release engineer would run a script to create some rdf files, and then put these on a server.
The last of the rdf tarballs still exist. A lot of work would need to do in to updating them, but its not starting from scratch. In the early days of the project, we found that people prefer starting from something than starting from nothing.
I don't know how rebuild this project to work at the kind of scale that would be required if it relaunched. Nor is it obvious to me how it might be federated.
However, this might work against the AI slop machines. Doing a search engine is not a realistic project for a scrappy bunch of people who care about knowledge. But a directory is.
update: https://curlie.org omg
-
I was the release engineer for the #OpenDirectoryProject 20+ years ago and I cannot say how much I don't want to program on it now, but I also think it's the only scalable way to fight search engine heuristics.
I don't want to code, but I'd support the hell out of a pateron to resurrect this project.
-
Asking a crawler or any automatized bot to recognize decolonized knowledge is something impossible.
I think we should temporarily go back to a good old web directory of decolonized content. Highlighting such a content to make it easier for it to spread.