Our opinions are pretty similar… great to have well structured re-useable components but I don’t want to use it to build my app.
The plan for famous-views is to add Famous Framework component support and target that middle ground… build your app in Meteor but seamlessly use Framework components with the usual Meteor magic. But FF is not ready yet for integration. You can see my post on the matter which links to a Framework issue that I created to track relevant progress. The issue makes it look like the Famous devs are really quiet but a lot of conversations have been happening on slack – from their side, they absolutely want to support these types of integrations, but (sensibly) need to prioritize more critical parts of Framework first (which is still alpha).
The alternative, that some people are doing (see #meteor on Famous slack), is to drop Meteor client-side and use something like asteroid to connect to a Meteor back-end from within a 100% Famous Framework front end app. I don’t personally like this approach, since I’m a big fan of Meteor as a full-stack framework with all the isomorphic advantages it brings (in Atmosphere too).
Yeah, Engine is pretty barren atm. Nevertheless, it’s still useful to build some cool stuff (sort by Views). Atm I’ve partially built a RenderController view, because it’s so useful for transitions. There are also some other proof-of-concept layouts from michaelobriena (from Famous Inc). Waiting to see at what level IjzerenHein will target the new famous-flex. Atm I’m quite tempted to write a few basic layouts, since it’s not that hard, and works so well with famous-views and routers / famousEach. But of course, concerned that I’ll be doing superfluous work where a lot of these things might be written more comprehensively in Framework in the future. But I think it’s the route I’m going to go for now, at least.
Would love to hear from others experimenting in this field.