Hi, Vasya. I'm glad to see the work is ongoing and I'm looking forward to where you take it.
I'll look at your ideas board and add and/or vote on ideas if I have any ideas that I think might be of use, but note that I'm pretty inactive now, not only due to my fading away, but also due to my preoccupation with my new very active (hours each day, anyway) project. It is data related, so I'll summarize it for you briefly below...
I call it KMAI, for Knowledge Management with AI. Whether I ever finish it or not, I'm enjoying the journey of developing a personal knowledge base management system (PKMS). I'm using the following PKMS work flow to learn how to create and and use such a PKMS (Remember that I have a 3-device private network that KMAI will reside in and be based on. I've installed and will use Ubuntu on each of the 3, so I could use your Linux version so it would be integrated into the work flow, but I haven't decided yet whether to depend on X11 to be there for GUI work, so maybe not. But inside Ubuntu or outside, JSON exports & possibly imports will be integral to the work flow. I'm also strongly considering SQLite to help with metadata surrounding the notes.)
You may recall that you once referred to Obsidian as a note-taking app. I don't remember how I responded, but KMAI will demonstrate how that note-taking app can be the basis for a full-fledged PKMS. If you're ever interested in KMAI, I'll be happy to tell you more and even share elements of its construction, but for now I'll just describe the work flow and be done with it.
The work flow itself starts with data capture (either Windows or Ubuntu), partly via clipping of Web pages in a browser into Obsidian (Windows, almost for sure) as markdown notes of knowledge with YAML for metadata on the front or ingestation of other data from the Web or wherever in the form of PDFs or .docx files or .xlsx files or other formatted word processing files (no browser/GUI needed) or other. Then, I'll use Obsidian (and maybe other tools like Memento, other JSON sources, etc and storage of the metadata in SQLite) to add metadata and normalize it based on its content, by now in Ubuntu to further process the notes, add metadata, refine it, and ultimately make it ready for publication and/or for the use of AI tools (bots/LLMs, agents, etc) to query & retrieve data for answers to my questions about the data (by now, hopefully actionable knowledge) within KMAI. The latter tools may be in Windows, Ubuntu, or some in each.
Phew! Sorry it took so long. Yes it's silly in a way, but hugely enjoyable for me already, as I am in the early stages. I'm a very heavy user of ChatGPT (and of all the leading chatbots, but mainly ChatGPT) now and consider it/him to be a senior subordinate collaborator.
But I'll keep anything I put in your ideas board down to earth and grokable to other Memento users. Please give me an excuse to tell you more about it, but I certainly don't expect you to do that. Who knows, if the interest hooks you, you could maybe help me take it forward, productize it, and make some money off it. If you did, my services and work will be for free in return for all you've done for me, and decision-making & ownership would be transferred to you. All tools and products in KMAI are free, local-based, and privacy-oriented. I'm thinking of using LM Studio to install local LLMs to be part of KMAI.
Good luck with Memento Database and with your life going forward. And thanks for reading this long message.