I can reproduce this. My guess is that when you do it like this, org-agenda-files becomes a buffer-local variable, it is defined correctly only on that buffer. And when org agenda is doing its thing it uses many other buffers where that variable is not defined. When you add a new file to the agenda using C-c [ it uses org-store-new-agenda-file-list which uses customize-save-variable which saves the new value as default for all buffers, which is against the goal of using directory variables. Maybe it's a feature more people would be interested in. You might want to take it to the org mailing list, where a real discussion over org development is happening.
Off topic, you mention your life divides into 2: before and after using roam-like concepts and dropping the concept of hierarchies. Any references regarding that? I just can't seem to find a personal usage for the graph mental model for knowledge.
In markdown you need two kinds of brackets for a URL, in latex just \href, in markdown you must remember how many # you have to put for sectioning, in latex you just write \subsection if you want that level, in markdown you have to take care of numbering lists and you have to indent them, you don’t need to care about either in latex… overall latex is way simpler.
That's pretty close to perfect for me! A clock table with a scope of subtree and start at today would already work when I have all todos in one tree. If I could link a todo from file B as a subtask in file a and then summarize everything in a clock table that would be it! Maybe this works with transclusions. Thank you!
Won't that mean all my tasks would be tracked in that one clock table? I have a lot of projects in my notes, each with its own sub tasks. But you're right anyways, I can just include the project specific files. Thanks!
There is also a time frame to choose when you build a clock table, for example a day. If in a day you work on 10 different projects then you need to give some thought to how you organize your information in org, so you get better, more focused aggregations.
Overleaf, as a product, has the feature to track changes, make comments, real-time editing with multiple people, though some features cost money. Just spend some time looking at their website to see this.
Emacs has a steep learning curve, even the Doom distribution. You will definitely need to spend more than a couple of days to learn how to do things in Emacs. Everything is available through a simple web search: videos, tutorials, blogs, code snippets. But you need to spend the time. Up to you.
Scheduled is very simple in org mode: it's the point in time from which the item is going to appear in the agenda. As such, I don't see what using a range is going to represent. If it's in the agenda, it means you are working on it. You can think of the range [SCHEDULED-DEADLINE] as a range to represent what you are looking for.
What you are looking for is called RAG (retrieval augmented generation). You get you data indexed and then when you ask a question, the relevant information is retrieved and presented to the LLM. You need two things: index you data - it has nothing to do with LLMs; either a regular DB or a vector DB if you are more advanced. Secondly, you need a framework that is the glue; e.g. Llama-index. Nothing for Emacs afaik, yet.
So why don't you have the emacssql code before (above) the consult code? Why an entire thread for this? Put another way, after: is not doing what you think it's doing.
I'm waiting for my new Kaweco nib because "I like this pen so much, I wish it would write differently". Fine is a bit scratchy on the low quality notebooks I sometimes use. I hope medium is better.
Is this `code`?
Is this code?
I can reproduce this. My guess is that when you do it like this, org-agenda-files becomes a buffer-local variable, it is defined correctly only on that buffer. And when org agenda is doing its thing it uses many other buffers where that variable is not defined. When you add a new file to the agenda using C-c [ it uses org-store-new-agenda-file-list which uses customize-save-variable which saves the new value as default for all buffers, which is against the goal of using directory variables. Maybe it's a feature more people would be interested in. You might want to take it to the org mailing list, where a real discussion over org development is happening.
The variable org-agenda-files should be either a string or a list. You've given it a symbol.
There are the "maxlevel" and the "indent" options, did you try them? Did you have a look at the manual, at all?
That is pretty nice and colorful, good job! And enough already with the remarks about startup time, nobody cares how quick your Emacs is starting.
Off topic, you mention your life divides into 2: before and after using roam-like concepts and dropping the concept of hierarchies. Any references regarding that? I just can't seem to find a personal usage for the graph mental model for knowledge.
How are those simpler than latex?
Simpler syntax: a url is just a url, this is a # Headline. Lists are literal, you just write 1. 2. 3.
In markdown you need two kinds of brackets for a URL, in latex just \href, in markdown you must remember how many # you have to put for sectioning, in latex you just write \subsection if you want that level, in markdown you have to take care of numbering lists and you have to indent them, you don’t need to care about either in latex… overall latex is way simpler.
Sure buddy, latex is simpler than markdown.
That's pretty close to perfect for me! A clock table with a scope of subtree and start at today would already work when I have all todos in one tree. If I could link a todo from file B as a subtask in file a and then summarize everything in a clock table that would be it! Maybe this works with transclusions. Thank you!
The clock table has a scope argument, you can include all org files in your agenda, so you don't need to link anything.
Won't that mean all my tasks would be tracked in that one clock table? I have a lot of projects in my notes, each with its own sub tasks. But you're right anyways, I can just include the project specific files. Thanks!
There is also a time frame to choose when you build a clock table, for example a day. If in a day you work on 10 different projects then you need to give some thought to how you organize your information in org, so you get better, more focused aggregations.
Overleaf, as a product, has the feature to track changes, make comments, real-time editing with multiple people, though some features cost money. Just spend some time looking at their website to see this.
Emacs has a steep learning curve, even the Doom distribution. You will definitely need to spend more than a couple of days to learn how to do things in Emacs. Everything is available through a simple web search: videos, tutorials, blogs, code snippets. But you need to spend the time. Up to you.
yeah i guess i will check other editors..
No worries. In the future, you might get an itch to try again, please do, as the rewards for learning Emacs are immense.
I got a pen from a friend. It was used, but it was a going away present and it is still dear to me.
Is he the creator of Hyperbole? respect.
That looks great! I'm reevaluating the way I install stuff from time to time, I'll have another look at Elpaca soon!
Scheduled is very simple in org mode: it's the point in time from which the item is going to appear in the agenda. As such, I don't see what using a range is going to represent. If it's in the agenda, it means you are working on it. You can think of the range [SCHEDULED-DEADLINE] as a range to represent what you are looking for.
Yes.
"What is the most important thing for a company? Is it the cash flow? Is it the inventory? Nuh-uh. It's the people. The people."
More than a friend, a co-worker.
Because the actual work doesn't matter. Like in real life.
I gotta come clean. I've been using Emacs since 1992 , git since 2012, and I've tried magit at least 3 times, once only last week.
Down voted not because you don't like a tool but because you try it 3 times and then call it horrible.
You can try the "-s" option for stats, including a list of changed files and number of +/- lines.
What year is it?
The emperor's new clothes.
What you are looking for is called RAG (retrieval augmented generation). You get you data indexed and then when you ask a question, the relevant information is retrieved and presented to the LLM. You need two things: index you data - it has nothing to do with LLMs; either a regular DB or a vector DB if you are more advanced. Secondly, you need a framework that is the glue; e.g. Llama-index. Nothing for Emacs afaik, yet.
So why don't you have the emacssql code before (above) the consult code? Why an entire thread for this? Put another way, after: is not doing what you think it's doing.
The pen comes with one cartridge. After that, the sky is the limit.
Are you trading outside regular trading hours?
No. Just normal trading hours.
Maybe it's because it's one share. It took them a while to find someone who is selling a single share.
VSCode uses a proprietary, closed-source python extension called Pylance. It's very good. So you need to adjust your expectations.
Same brand, same model, different nib size...
I'm waiting for my new Kaweco nib because "I like this pen so much, I wish it would write differently". Fine is a bit scratchy on the low quality notebooks I sometimes use. I hope medium is better.