Fabio Manganiello on Nostr: npub1carpg…yg4j0 one of the thorny points about the implementation of the DMA is ...
npub1carpg73ekh6qlv80pwvwxjmxsxfdw7n37wuns8acsalxcnazqa7sryg4j0 (npub1car…g4j0) one of the thorny points about the implementation of the DMA is indeed whether gatekeepers on the mobile OS side (namely Google and Apple) also have to support alternative Web engines to run Web apps.
Reading Apple’s official statement, it seems that their “concern” (which led them to drop support for PWAs altogether) is that rendering PWAs on other engines means a loss of security - they basically can’t control what code a PWA runs if the underlying Web engine isn’t tightly coupled to the OS.
Whether this is a valid argument or not (and whether the Web engine <-> OS APIs should be open and inter-operable rather than baked into the OS by the OS developer itself), the fact that Apple has backtracked sounds like this constraint has now been removed from their DMA compliance expectations - i.e. PWAs on iOS will always be rendered only within WebKit.
So it sounds like their malicious compliance stunt may have actually worked in getting a concession out of the EU.
Reading Apple’s official statement, it seems that their “concern” (which led them to drop support for PWAs altogether) is that rendering PWAs on other engines means a loss of security - they basically can’t control what code a PWA runs if the underlying Web engine isn’t tightly coupled to the OS.
Whether this is a valid argument or not (and whether the Web engine <-> OS APIs should be open and inter-operable rather than baked into the OS by the OS developer itself), the fact that Apple has backtracked sounds like this constraint has now been removed from their DMA compliance expectations - i.e. PWAs on iOS will always be rendered only within WebKit.
So it sounds like their malicious compliance stunt may have actually worked in getting a concession out of the EU.