I've run into this a bit when going through the url=value CS1 pages. So, this bot was just run on https://en.wikipedia.org/w/index.php?title=Tony_Martin_(British_singer). If you look at the comparison between 9 January 2024 and 2 March 2024 (04:13), you'll see that one of the changes made was to the shortened Billboard link used by previous editors. I'm fixing it with the long links, but it seems IABot wants to change the symbols used to shorten URLs on Wikipedia into the code used in URLs? I've been fixing these for a while, but they aren't the only issues I come across in the CS1 pages, so it's the first time I've noticed which bot is doing this particular function.
I though someone should know. OIM20 (talk) 09:51, 2 March 2024 (UTC)
Description
Description
Event Timeline
Comment Actions
To be honest, this should be disallowed:
| url={{BillboardURLbyName | artist=black sabbath | chart=all}} |
The URL field is for URLs that are encoded with percent-encoding. If percent-encoding is mixed with (or replaced by) wiki-encoding, it's not possible for automated tools to determine which encoding scheme to use. This is documented by an IETF RFC: URLs should only use percent-encoding and not mix encoding types. It's a fundamental breakage of Internet policy.
Comment Actions
This is even worse:
| archive-url=https://web.archive.org/web/20080603093537/{{BillboardURLbyName | artist=black sabbath | chart=all}} |
The output of the template can change when the template is modified, it will break the archive URL. It's an unstable and unreliable method for multiple reasons.