Re: One vs many directories

[Jean Louis <bugs@gnu.support>]

* Dr. Arne Babenhauserheide <arne_bab@web.de> [2020-11-21 18:04]:
> 
> Jean Louis <bugs@gnu.support> writes:
> 
> > When there are more than 2000 people related notes, tasks,
> > calculations, questions arise if such better be kept in one Org file
> > or multiple Org files in one directory or multiple directories for
> > multiple Org files?!
> 
> This came up multiple times in discussions. I think that it is wrong,
> and the reason comes down to efficiency. If you want to work
> efficiently, then sub-second delays matter, and having 4 files with
> 500 entries means that to search in them you need to open 4 files.

Hallo Dr. Arne,

It maybe wrong and it depends of the approach. My approach is that I
think with people and subjects, not with notes only.

Subject can be special plan like ABC.org and I do not need to search
notes related to that plan outside of ABC, because I do not mix
things. I am searching within one file only.

Things TODO are per subject or per person.

Files pertaining to any person are filed in the person's folder.

Somebody else deals only with personal notes and they maybe put such
in various files and of course they need to search.

I am thinking of "Joe Doe" and here is the flow:

- press s-c (for contacts)

- enter Joe Doe or Joe Doe, Berlin, etc.

- among many Joe Doe, I may narrow down to right one

- click F4 there is Org file for Joe Doe, enter Tasks, Transactions
  and whatever else, send Tasks, Notes to Joe Doe, collaborate or make
  agreements. I never construct or open file for a person, function is
  doing that. It makes the file
  ~/Work/People/By-ID/320431/320431.org

  If I need to search, I search inside of the file. 

- click F5 and find all other files for Joe Doe. For example contracts
  and similar. If I need to search there then I use find and grep and
  similar tools. No need for indexing. Files are mostly sorted by data
  how they come.

There is same flow if I think of a group of people with the difference
that if I need a person I still need to find the person in the list of
people. 

So in general I never need to use some general search through Org
files or any other files as my way of thinking begins with People or
Groups and that narrows what has to be searched.

> If you have 100 files with 20 notes each, you have to open 100
> files.

You maybe mean opening automatically files and searching through
such. I do not find Org system comfortable for that. I see it tries to
remember files, IDs, and agenda among various files. Not that I find
it comfortable. My way of thinking is always People or Groups, and
from there various searches are performed and that narrows drastically
the subject that has to be searched.

> My current setup has around 1200 notes in 10 files (most of them in the
> two main files, some of the notes are several pages long, but most take
> up around half a page).

People are all over the world using Org in various manners and every
day I find different ways of using Org mode.

On my side I almost never put notes in Org files. As by definition
from Wordnet, note is "brief written record". If it is brief written
record I do record it in the database under Notes related to person,
or group or opportunity or some case, or it can be related to anything
else. Then again I think of person and I can get all notes for the
person.

Org files I am using mostly for planning and project
administration. There are almost no notes, just instructions on how to
execute specific steps and there are headings with articles or
instructions that do not need execution. There are no records that are
saved for later or that do not need any execution or learning.

Org files on my side thus offer:

- hierarchical knowledge database that may be shared with other
  people, and is almost always directed to sharing with other people

- plan and project administration with tasks, whereby such subtrees
  can be shared with other people and for multiple times executed

If those are called notes by other people, alright fine. On my side
those are not just notes.

Notes I relate to objects like People, Groups, Opportunities, Cases,
so I put some notes there. But general dynamical knowledge repository
is better, that is where I mention HyperScope.

It is like database of hyperlinks that hyperlink to anything, it is
more abstract and I find that approach also versatile. No need to
define specific database for notes, all I do is defining hyperdocument
type to be "Note" and I can link it to anything else.

Semantic Synchrony
https://github.com/synchrony/smsn

Semantic Synchrony is using maybe better type of a database I do not
know, I am using SQL, SMSN uses graph database.

> Using org-rifle (https://github.com/alphapapa/org-rifle) I can
> full-text-search them with barely perceptible delay on a system
> clocked down to 1 GHz.

That is great tool for many.

Org files are for me to write complex documents like 850 kb something
like a organizational knowledge, training for each staff member,
plans, projects, tasks, etc. Majority of that stuff can remain in Org
files.

Maybe that stuff related to execution and collaboration I will move to
the database approach. All the unique ID stuff drops down forever as
database unique IDs are handling themselves without me thinking about
it, and is hard to make a mistake. It is basically putting data into
meta level. When I need a project made out of that meta data, I can
mark it all like in Dired or just mark set of such and export into Org
file and send to the person for execution.

Overall from this discussion I hope that people find some useful ways
of using Org, like org-rifle, semantic organization of stuff and
similar.