2 ms·

But the point here, I think, is that when I first submitted link A it was changed to link B, and then a commenter noted that link B took you to a login page. T

But the point here, I think, is that when I first submitted link A it was changed to link B, and then a commenter noted that link B took you to a login page. Then subsequently link C (which may or may not be the same as link B) works as expected. Too late to track down exactly what happened, I'll make more detailed notes next time.

pvg 4y ago
Right but I think your theory of the case is backwards - it's not something clever or complicated HN is doing, the too-clever-by-half and weird things seem to come from mastodon/pleroma. The original link leads to two separate redirects and then ends up on a very dynamic, js-heavy page which could have easily screwed up and not displayed fully for that other commenter. In fact, if you look at it in the Chrome debugger, the first warning it gives you is 'fix your code to not depend on navigator.userAgent', which it apparently does. The share/link thing on that page is the final link itself so the real oddity is the original link. Maybe you remember where you got it because it's a strange one - it's a double-redirect to a shorter and more direct link, sort of like using one of those jokey URL longeners (elongators?).