Frank, thank you 
I think DT (the ecosystem) was built as a sort of general infrastructure for finding, ingesting, grouping, tagging, classifying all kinds of bits of information. Because devil is in the details, it is the many little smart ways DT does this that is the secret sauce. For example, importing vs indexing; the flexibility of arranging your viewing/exploring space; the ability to work with all kinds of metadata (check also Tinderbox…); the use of Applescript for those who want it (and JS for others); smart rules; replicants; in-built richtext, markdown, etc preview/edit panels; and so much more I am certainly not doing it any justice.
On top of that, the amazing iPhone/iPad port, and the brilliant sync capabilities (I use webdav, bonjour). For such a small team, I can’t quite get it how they did it so well. And I have massive databases, very messy (all kinds of file types). I worked for database companies in the past (Oracle), ETL ones (informatica), Hadoop, you name it - always with products that dealt with massive amounts of data. But they had thousands of engineers. But maybe it’s because the team is small ?
I may be wrong, but I expect DT to evolve, I don’t think it’s going to be a revolution. And, IMO, it shouldn’t be, I suspect (and they won’t confirm or deny ha ha) it’s not their philosophy. These guys are not disruptors, they are magnificent craftsmen, like the Zeiss, the Leica, the Mercedes or the Range Rover of yesteryear. I believe they are focussed on fit and finish more than anything else. And they should be. There’s a great book by James Stewart, “How Buildings Learn” that explains why the old MIT buildings were so amazing. DT is the MIT buildings.
I don’t think DT is missing anything - I think it gets better and better, WITHOUT losing its readability as an ecosystem. My mental model of what DT is is clear. I basically GET IT what I can do with it - yet I am constantly getting surprised that I can do things a little better, or I can do stuff I didn’t know I could - and when I do it, I gasp at how they’ve already ANTICIPATED that I might want to do it.
So these guys know a whole lot about PATTERNS, architecture patterns and use patterns. And that is a MAJOR KEY - their users (us) use DT in many different ways and for many different purposes - but if you tried to listen to much to what your users_said_ they wanted (hat tip to Steve Jobs), you would get lost fast. I’ve seen that in most companies I worked with.
Patterns, especially MICRO-PATTERNS, is the key. Put together a platform that has a large set of very coherent micropatterns, that can be built in different workflows, and that’s it.
So, I just said, A PLATFORM.
I don’t like molecular food that much. Give me my Steak au Poivre, superbly executed, with fresh black pepper, fresh cream, grass-fed beef. Old-world like ingredient quality. DOP. DOP Micropatterns. That’s where I think it will continue going.
But, I could be wrong. I’ve been in the past. Many times.
Happy 2025!!