My experience has been very different so far.
I want to see if I’ve hit diminishing returns or if original insights are still available with this method.
I am not sure what exactly you try to achieve by cutting down on the number of nodes. Are you dissatisfied with the number of connections you make between your notes or do you feel that the method is losing its novelty for you?
For me, coming up with connections is easy. Noticing connections between different ideas feels really rewarding, and this is the main reason it has stuck with me so far. Cutting down becomes the hard part. Principles/heuristics for note-taking, like “in what context do I want to see this again?”, have helped me with that. But it might also be that this effect is mostly due to me using this method for only 3 months, and the novelty will wear off eventually.
I cut down the number of nodes because I felt like the project would be too tedious at scale, and having a handful of very fruitful nodes would make it harder to show if the rest of them weren’t doing anything.
I’m not sure I would say the method’s lost its novelty for me, since it’s more of an afterthought to note-taking usually, but I’ve found it unrewarding to look at this web of concepts swimming together and not get any eurekas out of it. It’s possible that cutting the chaff out might produce a tighter web that makes more meaningful connections, but this seems like a very daunting housekeeping project if I can really do it at all.
Drawing connections between Zettelkasten-style atomic ideas is better to me than full-throated complicated ones, and that’s where I’d apply the virtue of narrowness—if you try to smash whole fields together you get new fields less often than smashing ideas together generates new ideas.