Changes between Initial Version and Version 1 of Ticket #64596, comment 10
- Timestamp:
- 02/13/2026 10:04:47 PM (2 months ago)
Legend:
- Unmodified
- Added
- Removed
- Modified
-
Ticket #64596, comment 10
initial v1 3 3 I very much share your general concern and agree about the need for both forward and backwards compatibility. 4 4 5 That's why we intentionally designed what we merged in around being about to make this decision in 7.0 in the most graceful possible way, since as outlined above and elsewhere the additive benefits to URL fragments are significant and justify the change.5 That's why we intentionally designed what we merged into 6.9 around being about to make this decision in 7.0 in the most graceful possible way, since as outlined above and elsewhere the additive benefits to URL fragments are significant and justify the change. 6 6 7 Ability authors who wish to target WordPress 6.9 can give ita non-nested name. Here's how that'd look as a one-line ternary:7 Ability authors who wish to adopt nested namespaces while supporting WordPress 6.9 can fallback to a non-nested name. Here's how that'd look as a one-line ternary: 8 8 9 9 {{{#!php … … 20 20 And if an author is targeting 6.9 but forgets to add backcompat? Well, that's why we made it so `WP_Abilities_Registry::register()` in WP6.9 doesn't throw an error that would break things for the user, and instead just logs a notice that it's unsupported and skips to the next. 21 21 22 And of course, Abilities registered without nested fragments continue to work no matter what version of WordPress they were added in. As with REST, URL fragments, etc there's nothing stopping developers from polluting their root ability namespace with as many flat-level abilities as they want.22 Ofc, Abilities registered without nested fragments continue to work no matter what version of WordPress they were added in. As with REST, URL fragments, etc there's nothing stopping developers from polluting their root ability namespace with as many flat-level abilities as they want. 23 23 24 24 To me that feels like we've more than passed the bar for bidirectional compat here. If there are some specific frictions you're seeing please share and I'm confident we can get them bugfixed in 6.9.x.