A répondu dans un fil de discussion
Recherches récentes
Options de recherche
#orgmode
16 messages15 participants2 messages aujourd’hui
So... I strongly disagree with LLMs (mostly with the marketing and the training data issue), but I found a use-case for myself that they may actually be 'alright' at.
I organize my life in #Emacs #orgmode It's great.
But over the years, my notes and journals and everything have become so large, that I don't really have a grasp of all the bits and pieces that I have logged.
So I started using org-ql recently, which works great for a lot of cases, but not all.
Naturally, I wanted more consolidation between the results, and better filtering, as well as a more general, broad view of the topics I wanted to look up in my notes.
So I started writing some tooling for #gptel, to allow LLMs to call tools within Emacs, and leverage existing packages to do just that.
It's in its inception, and works only 20-25% of the time (because the LLM needs to write the queries in the first place), but it works reasonably well even with smaller models (Mistral Small 24B seems to do alright with 16k context, using llama.cpp).
In general:
- It kinda works, when it wants to.
- The main failure point at the moment is that the LLM isn't able to consistently produce proper syntax for org-ql queries.
- The context window sucks, because I have years of journals and some queries unexpectedly explode, leading to the model going stupid.
So far it's been able to:
- retrieve journal entries
- summarize them
- provide insights on habits (e.g. exercise, sleep quality, eating times)
- track specific people across my journal and summarize interactions, sentiment, important events
It doesn't sound like a lot, but these are things which would take me more time to do in the next year than I already spent on setting this up.
And I don't need to do anything to my existing notes. It just reads from them as they are, no RAG, no preprocessing, no fuss.
At the same time, this is only part of my plan. Next:
1. Add proper org-agenda searches (such that the LLM can access information about tasks done/ planned)
2. Add e-mail access (via mu4e, so it can find all my emails from people/ businesses and add them as context to my questions)
3. Add org-roam searches (to add more specific context to questions - currently I'm basing this entire project around my journal, which isn't ideal)
4. Build tooling for updating information about people in my people.org file (currently I do this manually and while there's a bunch of stuff, I would love if it was more up to date with my journal, as an additional context resource)
For now, this is neat, and I think there's potential in this as a private, self-hosted personal assistant. Not ideal, not smart by any means (god it's really really not smart), but with sufficient guardrails, it can speed some of my daily/ weekly tasks up. Considerably.
So yeah. I'm actually pretty happy with this, so far.
PS. #orgql because org-ql doesn't show as an existing tag.
Who wants to add #Journelly to the tool page of #orgdown?
https://xenodium.com/journelly-open-for-beta
https://gitlab.com/publicvoit/orgdown/-/blob/master/doc/Tool-Support.org
Or even assess its od support?
xenodium.comJournelly open for betaI've reignited Journelly, my note-taking/journaling project. The iOS app is coming along nicely.
I've been using Journelly daily. The best I can desc...
New post about #orgmode #orgbabel #emacs and #literateprogramming
Is there another way to do this?
Thanks in advance 
www.cnoceda.comA little trick in my literary programmingOver the years, my mind doesn’t work as well as it used to, and I need reminders for everything—even to know whether I wrote a program using literate programming or not.
Überzeugte Benutzer:innen von #Emacs #orgmode können bei dem Video von @CT3003COOL drunter einen Kommentar hinterlassen, sich eventuell mal in einem anderen Video mit Org zu probieren. (Ich habe kein Account, um YouTube-Kommentare zu hinterlassen.)
Bei diesem Zielpublikum eine interessante Herausforderung, denn sowas wie lock-in Effekte, Migrationsaufwand, Möglichkeiten, an den Rohdaten selbst mitzufrickeln (extern generieren/auswerten, ...) kam gar nicht vor.
Wichtiger scheinen da optische Gimmicks wie Titelbilder, Emoticons und Gifs. Wer das nachvollziehen kann, darf sich bitte bei mir zwecks ein paar Fragen melden.
Endlich Leben im Griff 
| #Anytype statt #Notion
https://www.youtube.com/watch?v=MKvDdsa6oyY
A répondu dans un fil de discussion
Zed's dead, baby, Zed's dead.
Anyway, I wonder if Zed can do all the things I do with #Emacs: It's my window manager, my terminal, my email client, my PIM system, sometimes also my web browser and my Mastodon client.
It could be my chopper, if I wanted!
Of course, if you do have long-time plans for your personal #knowledgemanagement, you should not use proprietary tools like #Obsidian, #OneNote, #Evernote and alike.
You also should think of switching to a much better (learnability, portability/compatibility, typobility, logic, featureset) markup syntax like #orgdown which is used in #Emacs #Orgmode (among other tools).
But if you start with Org-mode, you need to know the right way of doing it, as so many web pages recommend. Here's my take:
#UOMF: The Right Way to Use #OrgMode
https://karl-voit.at/2021/08/30/the-org-mode-way/
ad syntax/Markdown/Orgdown/#LML: https://karl-voit.at/2017/09/23/orgmode-as-markup-only/
A specific blog post about the #lockin effects of Obsidian and Co (despite using some tool-specific flavor of MD) is in the making. (Please don't start a religious discussion here.)
public voit - Web-page of Karl Voit · UOMF: The Right Way to Use Org ModeUOMF: The Right Way to Use Org Mode
Confused by the #Emacs key binding concepts? You might be interested in reading:
#UOMF: My Emacs Key Binding Strategy
https://karl-voit.at/2018/07/08/emacs-key-bindings/
public voit - Web-page of Karl Voit · UOMF: My Emacs Key Binding StrategyUOMF: My Emacs Key Binding Strategy
“Extending Org-Mode Export to Include Acronyms Section”
Background on my #OrgMode export ecosystem and changes I made in #Emacs to export acronyms and their names to #LaTeX (and PDF) documents. I also speak to some of the “why” of my personal glossary of acronyms.
https://takeonrules.com/2025/04/10/extending-org-mode-export-to-include-acronyms-section/
Take on Rules · Extending Org-Mode Export to Include Acronyms SectionIntroduction In Denote Emacs Configuration I wrote about adding abbr and abbr-plural link types. In short, when I exported these link types the process would render an abbreviation and its title, based on my glossary entries. At least that’s what they did for HTML 📖 and plain text.
But until today, I had deferred on handling the LaTeX export. There was a rudimentary handling of it, but I figured I could do more. So I did.
For folks who are interested, I've been working on a macOS utility (pre-release) that acts a proxy for Org Protocol URL requests. This utility is intended to complement another utility called Captee (http://yummymelon.com/captee/) which can send an Org Protocol request from the Share Menu. I'm currently looking for testers. If you're interested, please DM me your email address. Thanks!
#Emacs #OrgMode
yummymelon.comCaptee for macOSCaptee is an app that shares a link in either Markdown or Org syntax straight from the macOS Share menu.
From @jtr's lovely post:
"Journalley blows this mental challenge away. It invites me to take notes and just notes. It’s the opposite of a chore. “Go ahead,” it winks at me, “spill the beans. Snap some pictures. We’ll work on the details later.” Suddenly, I’m free to let my mind go."
#irreal and @bbatsov about #Emacs startup time: https://irreal.org/blog/?p=12903
I, too, am totally convinced that it really doesn't matter if it is one or sixty seconds.
However, if your bootup time is much longer than mine (15-30s once a week) then you most probably have room for improvement (if you want). Consistent use of #usepackage with dependencies and "defer" did help on my side.
My config has 7724 lines of #elisp (19851 including #orgmode comments): https://github.com/novoid/dot-emacs/blob/master/config.org
irreal.orgEmacs Startup Time | Irreal
A répondu dans un fil de discussion
@ctietze Yes, #Emacs has #org-babel thanks to #orgmode - I could see this being incredibly useful for non-Emacsians.
A répondu dans un fil de discussion
@bastian_S Jetzt noch den Nerdteil dazu, daher ohne Applist-Hashtag aber mit #linux.
#logseq nutzt als backend entweder #markdown oder #orgmode . Letzteres kann man natürlich über #emacs auch direkt nutzen und damit sehr gut aus Notizen und Ideensammlungen sehr akkurat gesetzt über #LaTeX PDFs kreieren... Für Poster, Präsentationen und Fließtext schon getestet.
die komplette verschränlung mit Logseq ist noch nicht gelungen, werde berichten, wenn fertig.
Suite du fil
Meeting notes can be found at https://list.orgmode.org/87friitxvw.fsf@localhost/T/#u
TL;DR: latex fontification slowness, org-font-lock-element branch,
fast searching of Org buffer AST, evolving Org notes over years, org-merge-tool,
org-crypt, org-crypt-refactor branch, GPG on Android, note-taking
packages for Emacs, speed of opening many Org buffers, highlight-parentheses
Earlier notes on the meetup page:
https://orgmode.org/worg/orgmeetup.html
#OrgMeetup #orgmode #meetup #emacs
CC: @sacha
list.orgmode.org#16 [[bbb:OrgMeetup]] on Wed, Feb 12, 19:00 UTC+3
It's been a long time since I tried #SymPy, but yesterday I went back to it to do some simple symbolic linear algebra, and I have to say it is pretty good these days.
The same project was also the first time I used #orgmode source blocks to insert generated output (from SymPy i this case) into a document. It took me a few times returning to the docs, but it is pretty nice for sharing calculations with fellow researchers.
#Emacs #Orgmode: I just committed a new function "my-org-link-to-headline".
It's asking for an org file (from your default-directory). Then it asks for a query that should match a heading title. On confirmation, it inserts an id:-link to that heading:
[[id:this-example][This is an example]]
I use this to link entries from my contacts.org (replacing now unmaintained helm-org-contacts) but not just that.
Initial code:
https://github.com/novoid/dot-emacs/commit/19d2bd32a45a077f76c733ea6b08999ff2e4b32e
Bonus: the entries are cached which is neat: it only indexes a bit on first use of each file. And there is a cache invalidation function included as well.
HTH
GitHubmy-org-link-to-headline · novoid/dot-emacs@19d2bd3to replace unmaintained helm-org-contacts
En réponse à Karl Voit 
@publicvoit @ajlewis2 @ellane @feralthoughts @hbowie @reichenstein Sadly, LMLs with complex syntax (like Asciidoc, orgmode) only have full support in particular software (Asciidoctor, Emacs). While markdown which is widely supported is too simple, and its extensions are different across different software (like admonition, python's markdown library and pandoc have different syntax for it).
A répondu dans un fil de discussion
@ajlewis2 @ellane @feralthoughts @hbowie @reichenstein
Just for clarification: same holds true for any other markup supported by pandoc, not just #Markdown.
However, if you stick with a syntax language that doesn't come with this explosion of flavors, you have less issues converting your data - in some cases you don't even have to convert at all any more.
The issue with Markdown is that its original form defined a small minimum of elements and each tool defined its own potentially incompatible extensions. With other #LML, the "original" or its standard defines the maximum set of elements and therefore, there is no need for "flavors" and no data loss or conversion effort.
HTH
A répondu dans un fil de discussion
@jonathanmatthews
Why not #Emacs and #OrgMode? 
My (wannabe) digital garden (not a blog) https://marcoxbresciani.codeberg.page/ is written in OrgMode syntax and automagically translated into HTML at the touch of "one button" (that is `C-x C-e P p` sequence, in Emacs) and with a simple configuration file and a CSS you'll be happy.
Some techy details: https://marcoxbresciani.codeberg.page/digital-garden.html
marcoxbresciani.codeberg.pageKnowledge 枯山水 (/karesansui/)