Currently, Ibis is branding itself as a decentralized competitor to Wikipedia, and honing in on questions about Wikipedia's moderation. The most common rationale for the federation I've seen is that on contentious topics you'll be able to have different articles with different perspectives on controversial topics.
But whereas federation makes sense to replace something like Twitter/YouTube, which are platforms and services, Wikipedia is a project. In something like microblogging, the service is separate from the content, as the users generate content and the experience is one of taking in different streams/creators in one feed. Federation works well there since social media is designed as a network.
An encyclopedia is more singular in how its used. The appeal of Wikipedia one place/article to act as a starting point for a topic, as opposed to having to cross-reference like ten articles each of which arguing a different thing.
However, Wikia (now named Fandom) is an entirely different story, as it is a platform. The local knowledge of various communities, fandoms, political groups, and technical tables is, despite the content entirely coming in-house, being hosted on proprietary platforms. Whether that be Google Docs, Reddit sidebars, or Wikia, this is where people are storing very vital information and links. Piracy megathreads, medical and scientific information for transgender individuals, political communities' sources list, obscure niches, etc, these are the sort of stuff which find themselves at the mercy of platforms.
The nature of this environment is one where there's a lot of room for competition, far less need for a massive network effect, and a lot of very disparate, smaller, communities which can move over with minimal hassle if we reach out to them.
Having a decentralized FOSS platform whereby people without much technical knowledge (which is the case for a lot of these people) can register on an instance and set up a wiki would do a lot of good and run into fewer logistical issues IMO. Gradually pick off and absorb these smaller wikis, rather than straight gunning to replace the everything-encyclopedia.
I do think the question of targeting though has an impact on what we design features in mind with and also what kind of usecases we pitch to people, which in turn will shape the sort of content and communities that show up. So there's value in figuring out this question.