I experimented a bit. Maybe I’m misunderstanding something, but it seems to me that moving a file to another group is no more time-consuming than tagging it. In the end, both organizing methods are roughly equally efficient. Or am I wrong?
This might be true for a single group or tag
My records usually have multiple assigned tags
For example, a receipt; Type-Receipt, Vendor-aaaaa, Budget-aaaaa
I think the group process is move the record to a group, then create replicants for each group
.
I see, but these could also be groups. Then I could add an additional organizational structure with tags. Should I need that. In the end, the organization is so complex that I have to organize it
My approach is similar to that of @DTLow . In my primary database:
- A very large
Archive
group which contains almost everything, tagged with “object type” and generic topics. - A number of groups (
Projects
), each concerning a specific topic I’m interested in, such as gadgets and alternative theories of the universe. Currently I have 80+ such groups. - Inbox items are replicated into Project(s) when suitable, and then filed into the Archive group.
- Scripts and smart rules help reduce friction in organization.
- Search, pre-defined smart groups, and a custom Alfred workflow are my primary means of retrieval.
- Some smart group examples:
- I follow a number of new apps, which I want to check periodically. The smart group for them has the following search predicates:
tags:app kind:doc opened:!#90days created:!#90days label:5
. - Documents I want to read are given a specific label. They can be found in a smart group with
tags!=note kind:doc label:7
. When I’m in the mood to read something, I retrieve a random piece with a simple script, which basically does this:tell application id "DNtp" to (open tab for record (some child of this_smart_group))
.
- I follow a number of new apps, which I want to check periodically. The smart group for them has the following search predicates:
- Some smart group examples:
This system can be seen as a watered-down version of the PARA system. I do not manage my short-term responsibilities in DEVONthink, so I don’t need the first “A” and the “R” of PARA.
With regard to the original question of this thread: Although it should be possible to do all above with indexed files, I prefer importing because I don’t have to worry about missing files and other quirks. Peace of mind matters. Only files that are also utilized by another app are indexed.