Great points @Mark Walters . I too was thrown by the short fuse (see my earlier question to wbaer ). Bigger picture emerged with Mark-Kashman 's announcement. Having just put hands on a couple weeks ago in a dev tenant, hadn't drilled down on governance/permissions implications, nor surfaced the subtleties identified in Mark's post.
I agree, an incomplete story bridging old features to new -- especially deprecated as quickly as this one -- risks placing many in a box. My views a bit myopic as company-wide news hasn't been particularly important for customers. Based on limited testing, however, we're convinced the new News architecture will encourage greater end-user engagement, and therefore more highly valued.
Stepping back -- at the risk of sounding like a shill -- have a long history on the Microsoft stack -- for better and for worse. I prefer the 'new Microsoft's' aggressive roadmaps and rapid iterations. While a firehose hard to keep up with, offers the community much more say in shaping new features (e.g., User Voice); fixing older ones; and hands-on earlier in dev cycles. Doesn't mean we roll 'em out to our users/customers on those timeframes. Far from it. For example, will still be a while before moving many classic sites to modern given the incomplete story around custom content types and metadata. Heck, we just upgraded our internal Win 10 workstations to 1709 (last fall's creators update) :-)