3 ms·
Yeah, URLs really don’t have much in the way of semantics. Path is clearly intended for hierarchical data and query for non-hierarchical data, and there are str
by chrismorgan 6mo ago
Yeah, URLs really don’t have much in the way of semantics. Path is clearly intended for hierarchical data and query for non-hierarchical data, and there are strong customs, some commonly supported or even enforced by libraries, but no actual rules. Ultimately, it’s just a string that the server can decide what to do with.
The really funny thing about this is that, when I was worrying about possible side effects if I responded 404, I somehow completely forgot how much of the web’s history the path has been useless for. Paths have won. No one really starts new things with URLs like /item?id=… any more. Yay!
- fpoling 6mo agoWikipedia web server treats anything after /wiki/ literally as the name of the article. So en.wikipedia.org/wiki/// is the article about C++ style comments
- chrismorgan 6mo agoOh, magnificent. Lovely high-profile example to add about empty path segments being meaningful.
- chii 6mo agoi wonder if it ought to be `/wiki/%2F%2F` instead...
- morganastra 5mo agoit looks like it goes to a disambiguation page for what "//" could refer to now (C++ style comments being the top entry), but that's delightful!