Org-brain – mind-mapping for org-mode(github.com)
github.com
Org-brain – mind-mapping for org-mode
https://github.com/Kungsgeten/org-brain
12 comments
> Personally, I look forward to ontologies becoming more common in orgs, as they are immensely powerful for getting everyone aligned on a body of knowledge - assuming that's what one is aiming for.
The Big Hedge Fund Bridgewater uses decision-making assistants in their common tooling. I assume that ontologies come part and parcel with that package, considering that "alignment" and cohesion is one of their key motives.
The Big Hedge Fund Bridgewater uses decision-making assistants in their common tooling. I assume that ontologies come part and parcel with that package, considering that "alignment" and cohesion is one of their key motives.
I've always thought of mind maps as spatially confused trees, and Org already does trees very well. This package introduces the "friends" concept -- is that the fundamental difference between trees and mind maps?
Org-brain allows me to have more than one parent. For example, I have a list of books I've read and reviewed (for personal use). When I started using org-brain I added author entries as parents; for comics I can do author and artists. So in the tree it looks like
* Read Books
-- Book Title
-- Author Name
-- Author Name 2
* Artists
-- Artist Name
But in org-brain I can connect all of these in a logical way so I can jump through it quickly.
* Read Books
-- Book Title
Review
-- Author Name
-- Author Name 2
* Artists
-- Artist Name
But in org-brain I can connect all of these in a logical way so I can jump through it quickly.
Mind mapping is more of a process, rather than a data structure. The process helps us brainstorm, recall, and visualize our ideas.
You're right in that some of the data structure behind mind maps can be represented using a tree. However, you'd lose lots of the benefits. For example, trees might not be as memorable or as unique as a mind map[1].
[1]https://imindmap.com/blog/mind-mapping-for-memory-why-is-it-...
You're right in that some of the data structure behind mind maps can be represented using a tree. However, you'd lose lots of the benefits. For example, trees might not be as memorable or as unique as a mind map[1].
[1]https://imindmap.com/blog/mind-mapping-for-memory-why-is-it-...
I'd say mindmapping is an inferior term for describing org-brain even. Concept mapping is better. But that term is also already taken for something else. Org-brain and thebrain provide another organizational system. It's closer to graph databases (but more restricted with only three kinds of relations). The way it mimics our natural way of thinking about stuff is really helpful. Few tools help us easily capture those relations. Often we're stuck with pure hierarchical systems (such as mindmaps or folder structures). So I look forward to seeing this org-mode add-on evolve and truly hope that this way of organizing information gets more widespread. And covered in more tools and on more platforms.
Double thumbs up to Kungsgeten!
Double thumbs up to Kungsgeten!
Does it support more than one brain? I can't tell.
The ultimate that I've found so far is Flying Logic (flyinglogic.com).
- directed acyclic graph structure
- boolean logic (both AND and OR) and truth-value tracking
- multiple documents
- automatic layout (like graphviz/dot)
- python import/export api
Flying Logic is nice but it's expensive and non-collaborative.
I'm not sure what exists that is a next step up from that. Maybe owl/rdf stuff?
I generally think this entire field is basically sort of addled, and we're just waiting for software to catch up to brains. Our brains aren't hierarchical, they're graph-based. But since our physical world is hierarchical, there's this impedance mismatch and most of our software (todo lists, outliners, mindmaps) makes the wrong call by forcing us into tree-based structures.
Try representing a recipe's instructions accurately in something like OmniOutliner, OmniFocus, or org-mode. Meaning, where blocking steps accurately block their dependencies. You can't do it. But you can with a DAG.
As another example, you can't even represent a simple argument in anything less than a DAG. It's like a compiler.
1. Socrates is a man (axiom)
2. All men are mortal (axiom)
3. Socrates is mortal (1, 2)
4. All mortals will die (axiom)
5. All mortals must eat (axiom)
6. Socrates will die (3, 4)
7. Socrates must eat (3, 5)
You can't represent those dependencies in an outline, nor in most mind-mapping apps (since most of them don't allow a child to have multiple parents).
Gantt charts can represent those relationships, but if you only care about the graph structure and the dependency relationships, why be saddled with a huge project management application? And, most of them can't handle OR relationships anyway.
Graphviz/dot is decades old and isn't collaborative. But at least it's DAG with good layout.
Free-form "just draw any edge to any node!" apps aren't all that great either, not without automatic layout, not without respecting any logical relationships indicated by the edges.
The ultimate that I've found so far is Flying Logic (flyinglogic.com).
- directed acyclic graph structure
- boolean logic (both AND and OR) and truth-value tracking
- multiple documents
- automatic layout (like graphviz/dot)
- python import/export api
Flying Logic is nice but it's expensive and non-collaborative.
I'm not sure what exists that is a next step up from that. Maybe owl/rdf stuff?
I generally think this entire field is basically sort of addled, and we're just waiting for software to catch up to brains. Our brains aren't hierarchical, they're graph-based. But since our physical world is hierarchical, there's this impedance mismatch and most of our software (todo lists, outliners, mindmaps) makes the wrong call by forcing us into tree-based structures.
Try representing a recipe's instructions accurately in something like OmniOutliner, OmniFocus, or org-mode. Meaning, where blocking steps accurately block their dependencies. You can't do it. But you can with a DAG.
As another example, you can't even represent a simple argument in anything less than a DAG. It's like a compiler.
1. Socrates is a man (axiom)
2. All men are mortal (axiom)
3. Socrates is mortal (1, 2)
4. All mortals will die (axiom)
5. All mortals must eat (axiom)
6. Socrates will die (3, 4)
7. Socrates must eat (3, 5)
You can't represent those dependencies in an outline, nor in most mind-mapping apps (since most of them don't allow a child to have multiple parents).
Gantt charts can represent those relationships, but if you only care about the graph structure and the dependency relationships, why be saddled with a huge project management application? And, most of them can't handle OR relationships anyway.
Graphviz/dot is decades old and isn't collaborative. But at least it's DAG with good layout.
Free-form "just draw any edge to any node!" apps aren't all that great either, not without automatic layout, not without respecting any logical relationships indicated by the edges.
org-brain doesn't support more than one brain out of the box. In practice though you can have two folders (one for each brain) and change the path which org-brain uses, and thus activating one of your other brains.
Hi! I'm the author of org-brain. I'm happy to see there's an interest for it. There's a screencast of me explaining it, which might give a better overview than the documentation: https://www.youtube.com/watch?v=3EGOwfWok5s
FWIW this reminds me of my iOS app Mindscope which is more visual but has the same core concept of zoomable hierarchies. Kind of a mix of Workflowy and Scapple. http://www.mindscopeapp.com
That looks pretty good. I’m pretty happy with scrapple on my Mac, and have been looking for something similar on iOS. Thanks!
My desire would be the same but with writing via stylus (Apple pencil specifically) instead of typing.
I take very free form notes during meetings and brainstorming and such via pencil and go back after to make a more coherent list.
For me, free form really needs a more direct analog input. Plus I can doodle at the same time.
My desire would be the same but with writing via stylus (Apple pencil specifically) instead of typing.
I take very free form notes during meetings and brainstorming and such via pencil and go back after to make a more coherent list.
For me, free form really needs a more direct analog input. Plus I can doodle at the same time.
Interesting you should make that connection, because I've always thought of Workflowly as org-mode lite (not meant at all as an insult, by the way, I think it's a great product). Though I haven't used Scrapple I can see how that would be a cool product. Though TheBrain is already pretty mature.
I've been using org-brain since a few weeks after its release when it was posted on Reddit. It was basically the tool I had been waiting for, combining the significant amount of knowledge I've built up in random files and org-mode outlines, but able to connect them in meaningful ways. If you jump back and forth between org-mode and brain there is some manual maintenance, but it's been worthwhile. Great tool.
I've been a Markdown guy for a long time now, and despite being a long-time Emacs user, I've never really taken the time to learn org-mode.
There was a post from 2014 that was on the homepage last week [0], and it got me _really_ interested in org-mode + ssh sessions.
I keep a daily log in Markdown, as well as a list of tasks that I'm tracking, and the ability to seamlessly describe _and_ execute code locally or remotely was incredibly enticing.
[0]: https://news.ycombinator.com/item?id=16559004
There was a post from 2014 that was on the homepage last week [0], and it got me _really_ interested in org-mode + ssh sessions.
I keep a daily log in Markdown, as well as a list of tasks that I'm tracking, and the ability to seamlessly describe _and_ execute code locally or remotely was incredibly enticing.
[0]: https://news.ycombinator.com/item?id=16559004
I've also been a long time Emacs+Org mode user, and while I was aware of babel and the ability to tangle literate programming documents, it simply never occurred to me that I could execute those snippets in-place and manipulate the results in-place like an ipython/jupyter/mathematica notebook.
I had the same reaction to that post, and I'm really excited to get more familiar with it, as it makes a great addition to some of the work I'm doing now.
I had the same reaction to that post, and I'm really excited to get more familiar with it, as it makes a great addition to some of the work I'm doing now.
I have made extensive use of org-mode for many years, and I keep trying to get into org-brain as a way to really have a "digital brain" -- I use org-mode to implement my own horrible version of GTD, to great success. I have tried to start using org-brain many times, but it never works.
I think there's something about having to externalize everything I know that confounds me. I wish org-brain could scan my brain and spit out something I could start working on.
I think there's something about having to externalize everything I know that confounds me. I wish org-brain could scan my brain and spit out something I could start working on.
Depending on how you use org mode (if you are creating IDs for everything), org-brain can scan it and spit out something. You can take an existing outline and turn it into a brain and start adding relationships.
Hello, I'm Harlan, the founder of TheBrain. Thanks for all the interesting feedback here.
We're pleased to give anyone who mentions Hacker News a 20% discount for the next week. To get the discount, just call us or use the chat system on TheBrain.com and ask for the Hacker News discount. (FYI, there is also a free version that is quite capable and never expires.)
We're pleased to give anyone who mentions Hacker News a 20% discount for the next week. To get the discount, just call us or use the chat system on TheBrain.com and ask for the Hacker News discount. (FYI, there is also a free version that is quite capable and never expires.)
Any relationship between TheBrain and org-brain, other than concept?
Curiously, I'd just run across Jerry Michalski's mind-mapping via Doug Rushkoff's Present Shock a couple of days ago.
Curiously, I'd just run across Jerry Michalski's mind-mapping via Doug Rushkoff's Present Shock a couple of days ago.
No relationship, I hadn't actually heard of it until just a few days ago. The author of org-brain states "It is heavily inspired by a piece of software called The Brain".
This is a useful package and one way in which to build a personal wiki with references to Org notes files and vice-versa. There can be a bit of a performance hit when the personal wiki grows to be quite large as the files are searched and indexed when performing visualization mode, but perhaps with additional popularity there would be people willing to tackle that enhancement.
thebrain.com is cool. Thank you for sharing!
Been around for a while too, TheBrain is an interesting program. Probably ahead of its time.
I hope it’s eventually open-sourced as I hope they’ve got loyal long-term enterprise customers to keep paying for support contracts and it would greatly expand the user-base.
I hope it’s eventually open-sourced as I hope they’ve got loyal long-term enterprise customers to keep paying for support contracts and it would greatly expand the user-base.
Funny, I just used their chat to ask them to open source. Might be cool to integrate with Google drive the way they have with colab.research.google.com
Surprised by how few good omnigraffle clones are out there. Markdown to tree in js as slides feels like a tractable space
An ontology has more structure than a mind map, and you can apply reasoners to them. It was useful for me, but it undermined a layer of obfuscating abstractions some of the technologists had created as a firewall between themselves and the rest of the org, so I lost traction for being politically tone deaf.
Mind maps are easy to dismiss because they are a unifying abstraction created by someone's point of view, which must compete with every other point of view in the org. Once you model it, it implies responsibility, accountability, and that can disrupt the relationships that people have invested in.
The act of documenting something well also adds newly competent stakeholders to decisions, which has the consequence that it dilutes authority in the org.
In this way, mind maps can get a reputation for being naive technology solution to a problem that isn't actually a problem, because the nebulous relationships they expose in fact already work for someone. The helpful guy with the mind map just doesn't 'get it'.
Personally, I look forward to ontologies becoming more common in orgs, as they are immensely powerful for getting everyone aligned on a body of knowledge - assuming that's what one is aiming for.