Vague statement. I'd like to see this reformulated as something more disprovable; a more tangible claim.

I know Eminem would agree. I'm not sure Jung would agree.

Hi from Hypothes.is!

I'd love to see a more interactive web, where we can collaborate on knowledge through public web pages & comments. Hypothes.is facilitates this.

• idea: technical subsystem with good naming & versioning. Example: clojure code published as maven artifact. group+project+version+namespace+name == piece you can pull out.
  • We could layer a system like Roam on top of a technical knowledge base. We just need foreign linking to work.
  • ... but I accept the stronger statement of "backlinks are useful from the inside of code bases too". I just think it's really expensive to fracture in tooling. Literate programming can be amazing, but it's hard to get PRs for a repo where all your code is written as literate org-mode tangled into clojure, compiled into an artifact, with home-grown tooling.

‘It is a huge misunderstanding that the only alternative to planning is aimless messing around.“

(from https://roamresearch.com/#/app/sritchie/page/Dnlb2ShjX)

I agree so much with this statement. I've written similar things:

https://www.teodorheggelund.com/posts/static-dynamic-interaction.html#planning-making-stable-agile-systems-by-setting-static-dynamic-boundaries

• YES. This. I want this. I want the feeling of Roam's freedom working with programming systems.
• I think dependencies and updates is the major challenge. Roam works great in a system where we can always depend on latest. The concepts are stable. In distributed systems, this becomes harder. I refer to something you have written. Then you change what you mean. What happens then?
• I believe the reason why there's no "official" multiplayer [[Roam Research]] yet. They haven't figured out "stable dependencies / immutable references", and that needs to be in place before it really makes sense to connect your Roam database to mine.

• If functions are namespaced by their page, and the page has a namespace, does that "double-layer" namespaces? Is there a difference betweeen "page without namespace", "page with namespace", function with namespace given by page"?
• does this open for collisions between pages and functions? ref "lisp 1 vs lisp 2". does the page with namespace ns (page/ns) conflict with the function on page ns (page/ns)?

That's right.

typo? https://en.wiktionary.org/wiki/dessicated

Woah, that's a broad scope! Essentially, tackle all the hard problems at the same time. [[Rich Hickey]] did some work one "granular function dependencies", I recall. That was about providing stable references, though. Also related to the scope of [[Unison (programming language)]].

Remember to properly reload your browser. I first tried a normal F5 refresh. My browser didn't reload all the javascript files, which didn't give me the right websocket-port. Refreshing with Ctrl+Shift+R fixed this for me (Vivaldi 1.13).

Why is this surprising? It is a definition. A force is what causes an acceleration.