Exporting search results

[bont007]

Is there a way in TriliumNext to perform logical searches (AND, OR, NOT) and save the results so that not only their links but also the texts of the notes can be exported to md or txt in a single text file? This would be useful for large databases if the terms to be searched for, such as “apple” and “pear”, occur frequently, are widely scattered and are not thematically limited to one section of the tree. The aim is to be able to capture and cache variable contexts of these terms ("apple pie", "pear tree" etc.) across the database.

[maphew]

Hmmm, a bit like a Book Note (renamed Collections in next release), but using a dynamic root query instead of a static note. I don't think this exists, it does sound useful. There is an [Action] for "execute script" on advanced search page header that could point to a note with an Export note function perhaps.

[bont007]

Many thanks for this hint. Basically, I had in mind a function that would allow you to search for a selection of keywords, the ramifications of which only become visible when you place the notes close to each other to examine them. This would mimic an old slip box, where you pull out a selection of indexed slips of paper and place them on your desk to directly compare multiple parent-child relationships and glean additional, unexpected information. If necessary, these adjacent slips of paper can then be copied and assembled into a text without much effort. This idea has been tried to be implemented in the zkn3 software, where "Zettel" (slips of paper) can be spread out on the "Schreibtisch" (desk) for this purpose (see e.g. https://www.hgwdavie.com/blog/2021/7/12/zettelkasten-zkn3-english-language-manual).

[maphew]

Making knowledge proximity/similarity naturally discoverable from content itself, as distinct from, say, a curator defining and establishing relationships ahead of time, is a very intriguing field. It's a part of machine learning I frequently think I would enjoy spending more time on; computer says "hey, did you know these three documents all talk about the same subject?". If only I had fewer interests!

Narrowing this down to Trilium today - perhaps chaining the search + action idea above with Relation Map would be fruitful.