I appreciate that the article mentioned that Threads isn't federated and these claims often never materialize, because I feel like this is the most important fact, and it's barely discussed.
Threads is not federated. There's nothing to defederate from, and it's unclear if there ever will be.
All the benefits of embracing activitypub are provided by announcing adoption, and all the downsides only materialize by actually doing the hard part. Based on experience and logic I think federation will happen far, far in the future if ever.
Yeah, that's exactly what I think is the strategy.
If I were the VP or whatever in charge of this -- which means that I'm assuming I'm a dishonest corporate shill -- I'd create a roadmap to integrating activitypub that includes the very most basic functionality as the first deliverable, then any usability or UX features as round two deliverables. Then I'd tell the project manager to plan for an open-ended "assessment" period between completing round one and starting round two, and tell them to move on to other tasks until further notice,
IF they ever finish, you just announce that it's technically possible for someone on Mastadon to query the Threads server and vice-versa, but, like there's no way to actually find people on Mastadon servers in Threads or follow them, and vice-versa, and then leave it that way. And that assumes the project doesn't just get suspended at some point and left that way.