Not sure if this has been requested before, but I couldn’t find it with a quick search of a few things I thought would bring it up. One thing that I frequently find myself wishing that Logseq had is some sort of “context” feature (perhaps in the right sidebar) that turns up blocks/pages that are related to what I’m currently working on.
I recently saw the announcement of “Mem X” from the folks at mem.ai, and it looks extremely promising. IMO Logseq + a language model like the one powering Mem X would be a pretty unstoppable PKM app.
I agree… We currently use Python to generate external AI auto links and “near match” linking analytics. This works well - but deals directly with the markdown text files. I think ALL “note taking” / PKM applications must implement this “forward looking”, real-time AI processing natively within their app if they want to be competitive in the PKM space in the near future. Integrated GPT-3 is also going to be a requirement - both from a “text generation” standpoint - as well as using NLP to generate well formatted, complex query results. Again you can easily do NLP to Datomic queries externally - but this really needs to be included in the PKM in order to make data inquiry simple for non-technical users.