I posted this to Mastodon but thought it bore repeating here:
Im sure Purism intends to and likely will be privacy respecting. However we need to be building and using infrastructure where one entity isn’t well positioned to mine or observe all of our data.
We shouldn’t be depending on Purism to be honorable.
Our platform vendors should not be our service providers.
We should have networks of diverse service providers.
Collusion between all these silos of data should be difficult. Not easy.
I’d say so. I suppose I get the appeal, but even looking at the list of services they are providing, I’m not sure anyone actually needs those.
I think Purism is fighting that battle Mozilla lost, because it can not be won in any meaningful way. So how do we get people to stop relying on all encompassing platforms? 
this is such a depressing sentence 
Short term I think we need to explore techniques that brings similar convenience without the downsides of centralization. One such example: having a single login for everything even though the services themselves are not centralized.
Purism should be something akin to an OpenID provider and working with lots of other partners such that users can use that purism login with other vendors like say ProtonMail and other mail providers as one example. (In the case of email an alias that forwards from the librem one domain to the provider the user has chosen, might be apropriate.)
Long term, I think the future is peer to peer. Such that it doesn’t even make sense to centralize services.
I loved OpenID, but it always occurred to me it was a single point of failure. I mean, any service depending on a single login is bad news bears, ne?
I mention that, because that’s where my brain goes, and I think, hmmm, not sure what is better: load everyone up behind a single login account, or lose the masses to the services that make it oh so easy with a single login account.
The future is already here – it’s just not evenly distributed. - Willy G.
I’m still digesting your vision, and I’m down with turning this discussion into an action plan for it. 
Other thoughts: librem.one just makes me think of Ubuntu One, which is not great marketing.
I mention thats just one such possibility for nailing centralized convenienced in a non centralized system. But there are others.
I often think for example that the web/desktop browsers needs something akin to android’s intents. Such that these distributed systems can provide features on par with integrated systems. For example: when a user opens a document attached to their proton mail / they can seemlessly open it in their next cloud on an entirely different server.
Man that shit is high concept even for me; I wouldn’t know where to start with actually making it reality.
EDIT
Though it is highly inspired by my thinking the future is peer to peer.