From 6d06059247e79e8d329526d4ee2c99894c88c7e0 Mon Sep 17 00:00:00 2001 From: Szymon Szukalski Date: Tue, 7 Apr 2026 11:54:35 +1000 Subject: Remove roster CSV seeding --- README.md | 15 +++++++++++---- 1 file changed, 11 insertions(+), 4 deletions(-) (limited to 'README.md') diff --git a/README.md b/README.md index 4225d93..5579d60 100644 --- a/README.md +++ b/README.md @@ -56,17 +56,20 @@ The minibuffer stack is intentionally small: - `marginalia` adds annotations. - `corfu` handles in-buffer completion popups for text and Org buffers. -Name entry uses two separate paths: +Name entry uses fixed abbrevs plus the roster: - `abbrev` provides deterministic one-shot shortcuts for fixed name expansions. -- a small CAPF feeds Corfu a fixed list of name variants from `name-dictionary.el`. -- `M-x ss/name-dictionary-add-name` and `M-x ss/name-dictionary-remove-name` update that file and refresh the current prose buffers. +- a CAPF feeds Corfu name variants from the legacy shortcut list and from the + structured people roster in `~/org/areas/people/roster.org`. +- `M-x ss/name-dictionary-add-name` and `M-x ss/name-dictionary-remove-name` update the legacy shortcut file and refresh the current prose buffers. - `M-x ss/name-dictionary-add-name-from-region` uses the active region as the name being added. +- `M-x ss/people-find` opens a roster entry. +- `M-x ss/people-insert-summary` inserts a compact roster summary at point. ### Persistent abbrevs Persistent abbrevs live in `abbrev_defs` at the repository root. The config loads that file on startup, enables abbrev mode only in text-like buffers, and saves learned abbrevs back to the same file silently when buffers are saved. -The name shortcut list lives in `name-dictionary.el` so it can be managed from Emacs with a single source of truth. +The legacy name shortcut list lives in `name-dictionary.el`, and the structured roster in `~/org/areas/people/roster.org` is the source of truth for people records, including name, role, employee time, engagement, team, abbrev, aliases, manager, email, and location. ### Babel tangle process @@ -118,10 +121,12 @@ This means the agenda reflects the current project and area files at runtime ins - daily notes - daily meetings - Denote-backed captures for generic notes, projects, areas, people, and resources +- structured roster captures for manager-facing people data Daily task capture writes under `Tasks`. Daily note capture and daily meeting capture both write under `Notes`, and the meeting template prefixes the heading with a timestamp and the word `meeting`. Denote captures still prompt for title, keywords, and subdirectory placement where appropriate, but folder placement does most of the classification work. The project capture template prepopulates the `project` keyword. Area, person, and resource captures do not inject structural keywords automatically, and there is no Denote-backed meeting capture template. +The people roster capture template writes to `~/org/areas/people/roster.org` and records name, abbrev trigger, aliases, role, engagement, team, manager, email, and location. `C-c n f` opens the roster lookup prompt, and `C-c n r` opens the roster file directly. `M-x ss/people-report-by-engagement`, `M-x ss/people-report-by-role`, and `M-x ss/people-report-by-manager` generate filtered roster views. ### Note creation and linking @@ -130,8 +135,10 @@ Denote handles long-lived notes. The main bindings are: - `C-c n n` to open or create a Denote note. - `C-c n l` to insert a Denote link. - `C-c n M` to open the central MOC note. +- `C-c n f` to search the people roster. - `C-c n m` to create a PARA subdirectory from the minibuffer before capturing into it. - `C-c n d` to open today's daily note. +- `C-c n r` to open the roster file. Keyword prompts and directory placement are part of the workflow, not an afterthought. The config is set up so structure is created first, then capture writes into it, with folder placement carrying most of the durable type information. -- cgit v1.2.3