Many are by bricolage
Raul Pacheco-Vega is a researcher who blogs about his process, which is a mix of things he calls synthetic notes (analogous to compressed individual grounded claims) and memoranda (which synthesize across claims; @pacheco-vegaWritingMemorandums2020) and Conceptual Excel Dumps (more elaborate compositions) (see, e.g., R: pacheco-vegaSynthesizingDifferentBodies2016 )
The practice of synthetic-note and memo-writing are inspired in part by the practice of memo-writing in qualitative research (see, e.g., Corbin and Strauss's grounded theory guide)
Money quote: "Writing memos should begin with the first analytic session and continue throughout the analytic process. Doing diagrams is more periodic but nevertheless very important. Doing memos and diagrams should never be viewed as chores, or as tasks to be agonized over. They are also not to be confused with finished papers ready for publication. Rather, memos and diagrams begin as rather rudimentary representations of thought and grow in complexity, density, clarity, and accuracy as the research progresses. One of the complaints we often hear from students is that writing memos and doing diagrams are just too time consuming. They say that they would rather make a few notes in the margins of their field notes. We puzzle over those remarks. Writing memos and doing diagrams are part of the analysis, part of doing qualitative work. They move the analysis forward and as such are just as important to the research process as data gathering itself."
Another money quote (about compression): "Though analysts can write on the actual interview or field notes, this is not practical, except perhaps in the earliest phases of open coding...it is difficult to retrieve information, in other words to combine or sort memos, if the margin of a field note or interview transcript is the only place where information has been stored"
It eschews folders and tags for a more innovative networked structure to allow for the notes to "talk back" to you)
Others apply the framework of qualitative analysis to literature reviewing, correctly understanding the importance of context (rich, flexible connections to sources and their context) alongside Atomicity and compositionality.
and networked notebooks are maturing significantly as a class of tools, an ideal environment for idea: discourse graphs, particularly to deal with the Z: Compression and contextualizability are in tension
Related to Z: Effective individual synthesis systems (seem to mostly) exist (for a select few), in that, often a good bit of "semantic publishing work" (at least in proto form) is intrinsically and immediately useful for the sensemaker.
another thing to consider is that "single-player' personal sys/Memex systems are kind of actually alive and well, albeit mostly living in hacker space (i.e., Z: Effective individual synthesis systems (seem to mostly) exist (for a select few))