More info – it is implemented in Photon, our network-transparent and reactive dialect of Clojure/Script (implemented as a macro like core.async). The explorer abstraction binds to server-side datafy/nav so you can attach it to any server side data structure, for example the https://gist.github.com/dustingetz/dd67a35d818e3a1bf6733147cf5cdea7

To bring this to parity with the https://docs.datomic.com/on-prem/other-tools/console.html (onprem) or https://github.com/tatut/xtdb-inspector or https://docs.datomic.com/cloud/other-tools/REBL.html is basically a tutorial exercise at this point. With added benefit of being web-first and maximally extensible

Wow! Perf is incredible, considering the scroller implies pagination that coordinates backend-frontend-DB nav on each scroll action, right? No promises, but here are our team's first thoughts in response: 1. We would use it to explore the heck out of our DB schemas and data. 2. We would use it not only to show, but also sometimes to edit things to debug our app when we transacted the wrong stuff and need to correct it. 3. We would use it to see new data trickle into the DB while we develop new schemas and features (and this would help us do 2. above less often). 4. We might add a https://babashka.org/scittle/codemirror.html to run arbitrary queries https://github.com/bsless/datalog-parser.spec and explore the results in this rich UI. 5. We might hook 4. above up to a filesystem storage of past queries, so that we might reuse the same queries over different DB replicas representing different environments (dev, staging, prod). 6. We might contribute back the general improvements we code.

Indeed the scroll is efficient, server streamed etc thanks for pointing that out