3 ms·
My understanding is that originally [and perhaps currently] Hacker News uses the file system for storage and that the id number is the name of a file on disk. A
by brudgers 11y ago
My understanding is that originally [and perhaps currently] Hacker News uses the file system for storage and that the id number is the name of a file on disk. A few years ago, I recall a discussion about a reorganization of the files from a single [or few?] directories into a more broadly branched tree. The basis for doing IIRC so was to improve performance.
My impression is that this would be a simple way to produce a RESTful interface. When the resource is a file on disk, how complex does application layer routing have to be?
Anyway, my guess is that general SEO is not particularly high on the list of features to implement. On the other hand, if adding Algolia and the API then it's another story. That was also a substantial improvement to the feature set. It might even turn up the aforementioned discussion about the file system hierarchy [I think PG wrote the post].
Good luck.