Using Obsidian to manage goals, tasks, notes, and software dev knowledge base
joshwin.imprint.to
joshwin.imprint.to
I’d like to use a more powerful app like the one promoted here, but I feel desperately uncomfortable tying my notes [in] to a piece of proprietary software. I feel like that was a mistake I made already, and don’t want to repeat.
It’s almost like Obsidian et al are highly featureful filesystem browsers, but where the only files one can have are markdown files.
What about if I want to style a real PDF in Asciidoc? It’s really common for a note to evolve into a document. An idea becomes a lesson plan for a class, which morphs into a handout with tables, admonitions etc.
What about if I want to represent some idea with a quick spreadsheet? Or a sketchup? It would be fantastic if those were somehow all represented as first class documents in the filesystem, as markdown is inside the current wave of proprietary markdown editors.
Notes in Obsidian are just regular markdown files that sit in a directory on your hard-drive. The notes can be version controlled with Git, synced with Dropbox, etc.
I've found that if I write something down on paper, I'll physically remember where it is. It's really hard to get past the permanent aspect of pen and paper, but once you do, and you're willing to write down imperfect thoughts with zero organization, it's wonderful.
As someone that grew up without any electronics except a TV, how the world has changed ... now if you could pretty please get of my lawn :)
I find also pen and paper to be limited for programming notes.
It is a set of vscode plugins, which help me manage a workspace of markdown notes. Fair warning: it can be a little overwhelming to get started. At least for me it was because I wanted to do everything the "right" way. Changing stuff and refactoring is easy, so now i just write things down and worry about structure as it grows.
So far, i really like it and the dev is very responsive on discord to fix issues i had with windows. He even offers personal 30 minute sessions with him where he tries to learn more about the way you'd like to use it.
If they need to create a PDF I use pandoc and if I need to move a table to a spreadsheet I just copy the text and use the text to columns from delimiter feature. Not sure if this help you or not.
> but I feel desperately uncomfortable tying my notes to a piece of proprietary software.
Obsidian is proprietary, but Markdown isn’t. And there will always be Markdown personal wikis. I guarantee it.
> What about if I want to style a real PDF in Asciidoc?
Pandoc can export MD to PDF, HTML, .doc, anything. And you can style with CSS.
But once you add spreadsheets and sketches into the mix, there’s basically no way around proprietary software.
Heck, I work as a developer, spend a third of my day in the terminal, and I still wouldn't use it.
For me, it is undoubtedly the best task and knowledge management tool I’ve ever used. But, I accepted a long time ago that the Emacs requirement is going to be a deal breaker for a lot of people.
My experience was that the Org-mode in Emacs was great. Especially how you could capture notes with backlinks from any Emacs context, and the Agenda view it generates. I went all-in on the ecosystem and wrote my mail in mu4e, for instance, which can integrate with Org-mode for writing HTML mail. Being able to add inline LaTeX equations and syntax-highlighted code snippets to such emails was gold. Having easily executable script snippets embedded in sysadmin notes was useful as well.
As for other implementations of it — including vim-orgmode and every mobile app I tried — too many features were lacking or broken for it to be useful to me. In the end, I ended up going back to MarkDown when I left Emacs. The transition was quite painful; Pandoc helped a lot, but many things in my notes just broke during conversion, so I ended up just trashing most of my notes and spending an evening manually fixing the important ones.
While Org is free and open source software, and it’s often repeated that “it’s just plain text”... I’d like to caution others that if you really use the features it has to offer, you’ll be locked into it just like with most proprietary formats. If you’re not completely sure you’ll use Emacs for the rest of your days, MarkDown is a safer option for a knowledge base you wish to keep over time.
There is one minor feature from Workflowy that I miss though, which is is the ability to "zoom" into a headline. You can narrow subtrees, but its level won't be adjusted. I tried to use some horizontal scroll hackery but couldn't get it to play well when you're switching buffers. I'm not sure how to save those narrowed buffers across sessions either.
But in any case I definitely will stay with org-mode for quite a long time, it truly is a powerful piece of software.
For mobile viewing I use the app GitJournal.
Thanks for this
--a fellow git + markdown + cron note-taker
I'm a happy user of Boostnote.
(1) It’s not proprietary - it’s a browser +/- IDE for markdown files. I point it at the folder in my Dropbox directory where I already keep all my md files. It’s a new front-end, but doesn’t lock me into anything.
(2) As an extension of 1, I can continue using NotePlan as my calendar/todo on my phone (which builds everything into md files), and Kiwi to access all of the above on my phone as a personal wiki.
(2) was already my workflow and knowledge base; Obsidian just made for a nicer editing/browsing tool on PC.
The key is that, as what is essentially a wiki, both the files/notes and their interrelationships are conserved/non-proprietary, regardless of what happens to the chosen software in the future. No lock-in for any part of my calendar, workflow, or knowledge base, and all three can point to one another because it’s all just md in the same folder.
I would call requiring a paid commercial license if you plan on doing commercial use "proprietary." That was the deal breaker for me, as I'm more than happy to continue using Foam until a free Obsidian-like client comes along.
- https://www.youtube.com/watch?v=IS3i3DTUnAI
It's about connecting many graphs together, the killer app in this space is to figure out how to connect these graphs together such that teams and communities can grow connections on a massive scale
There will be many challenges in doing that particularly around selection, per edge permissions likely leveraging Clojure's namespaced keywords to get global attributes
This will mean a major re architecture of Roam Research, I predict most Roam clones are going to get absolutely blindsided by this
Roam today is ~just~ a datascript database on the client being kept up to date with firebase
For now if you want open source Roam keep an eye on https://github.com/athensresearch/athens
How often do you write something down and look it the next day and realize it's not relevant anymore because of reasons like:
• you memorized it
• it's not as important as it seemed yesterday
• it's a todo and having it there actually nags you and prevents you from completing it
• it creates some other kind of mental load: while skipping over it, it grabs your attention, which then prevents you from attending to other tasks on the list.
I have a theory that GTD helps only few, for most people it just makes them feel better because they are now spending time creating lists instead of doing what's on them.
I find it's actually the opposite. Writing things down means I can forget about whatever it is and focus on the task at hand.
> it's not as important as it seemed yesterday
This is a good thing in some situations. I generally write down ideas whenever they appear. Coming back to the ideas in days/weeks/months helps to provide some perspective. If it still looks like a good idea, it can be acted on.
Bad ideas can be discard or ignored.
It's a better outcome than spending a day getting side tracked only to wake up the next day to realise the idea was rubbish.
It doesn't hurt to have a backlog of things for those slow, inspiration free days either.
Agree with raytracer, that it works contrary to this. Writing things down in a system that you can reliably go back and check, tells your brain to calm down about this thought. If required, you can always reference it later.
You will need to classify points as either tasks or general information you want to store for later.
The cool part of tools like Roam, Obsidian, is that you can create bi-directional links. So if you are intentional about this and if you're referencing something related later, you will come across this point and it may be useful then.
> • you memorized it
This means I've got about a day or two at most while I still have it memorized - but my plans usually span longer than that.
> • it's not as important as it seemed yesterday
That would account for most of the irrelevant 5% of my notes, I suppose; but most of the time, it will still provide important context when I come back to it weeks later.
>• it's a todo and having it there actually nags you and prevents you from completing it
Never really understood these sorts of todo issues. If a task is sitting somewhere for longer than necessary, it's likely unimportant. If it is important, I'm more than glad to be "nagged".
>• it creates some other kind of mental load: while skipping over it, it grabs your attention, which then prevents you from attending to other tasks on the list.
Move it to the bottom, or to another list outright ("To do later").
>I have a theory that GTD helps only few, for most people it just makes them feel better because they are now spending time creating lists instead of doing what's on them.
I have a theory that people like to riff on organizational tools when they do not know how to utilize them properly - which is why most are drawn to utilize them in ineffective and inefficient ways, generalizing this as a property of the tool.
Besides, GTD/task management is a miniscule portion of why it makes sense to construct knowledge bases.
My breaking point: I have DropBox folders for 1770 math papers, that I periodically read on computers and tablets. The significance of most of these papers is lost to me. I want to be able to browse this collection in a "supermarket shelf" topology. Most of creatitivity is not planned; planning tends to stifle creativity. Rather, a frenetic mind is fortunate enough, after months of fruitless wanderings, to encounter the right set of ideas in close enough proximity to notice. For example, a PCB protoboard has the correct 0.1" spacing to guide a router jig for making a pasta guitar. Who knew?!
Ideally one traverses a private web site with links to every asset. The associations need to be both manual and automated, logical and accidental. For example, time of creation or visits is critical data for determining proximity. Looking at each PDF in plain text to attempt to infer proximity would be a great application of deep learning tools, leveraging existing proximities as training data.
On a larger scale I have repeatedly urged the American Mathemetical Society to open up its MathSciNet cash cow to be a premier playground for machine learning. Unlike computer science, mathematics does have a detailed index of all published research, through the efforts of many volunteer reviewers. However, math is crippled by many archaic beliefs: That ideas are organized by fields such as number theory, and there is a single global worldview specifying this organization. That the value of MathSciNet is its hand-curated organization. I believe that we're crippling a generation of mathematical progress by not allowing mathematics to be at the forefront of mind mapping efforts, with many competing voices/tools describing its organization and facilitating idea browsing. Autocomplete is still at the "pager" stage of development, with "smart phones" yet to come. We could all benefit from the cloud autocompleting each of our mind-maps. Mathematics is a relatively small, constrained domain where shared mind-mapping and autocomplete could first flourish.
I've separately used a text file as scratch paper many days for the last fifteen years, saving every file. I can retrieve much useful information through full text search. I am reminded however of Don Knuth's adage "Premature optimization is the root of all evil," and the smaller optimization problem of how to best organize physical receipts. Excessive organization is wasted effort and a form of neurosis. I scan lots of documents but not every receipt; they all go in boxes dated by quarter, vacuum packed after a year. When I really need a receipt I can find it, but the effort to file has been properly balanced with the effort to retrieve. Same with text files; full text search works. It would be a mistake for me to put more effort into the structure of these files. Rather, through scripts and AI tools, I can evolve an indexing exoskeleton around these existing unstructured raw materials.
I need mind-mapping to manage assets, not to create a private web site of my thoughts. And there cannot be a user/programmer dichotomy; we each need to script additional associations into our mind maps. That belongs in exoskeleton, not cluttering up primary text.
The overhead that these systems need is my worst nightmare.
I tried obsidian, because I saw how people use notion, roam and obsidian ro manage tasks and as knowledge base... but I just can't imagine me doing this.
I programmed a Bot for todoist and at one point I developed a github > Todoist sync, to manage my work on the Bot... at on point it almost was comical how many additional tasks I ended up with, just to make it happen.
I tried obsidian and quickly tried to use Typoda for markdown, because even writing markdown comes with an overhead... additionally I always need to install 3rd party apps to even access my knowledge..
I'm still looking for the perfect system, but always to back to Todoist for tasks and a simple static file generator for my knowledge base(docsify).. I can edit and access it from everywhere and dont have much overhead.
What I also dont get is the lack of API of theses systems. Afaik notion still don't have one, and even roam doesn't.
For journaling and brainstorming, I build a Todoist function that creates and links a Dropbox Paper directly to the task in Todoist... this way I already have a task and it's out of sight when I'm done... and it only requires a browser.
I envy everyone who can use these apps though. It does seem to be a nice way of offload stuff of the brain.
Usually what I want is something that loads very quickly and has a minimal UI that allows me to focus on writing, but also allows me to search all files quickly. iA Writer ticks all those boxes, and also has support for simple #hashtags so I can organise the notes.
Where things become more difficult is if the app adds a lot of custom functionality that is either specific to that app or adds lots of custom metadata to the files. Then you start to lose some of the above benefits, particularly cross-platform.
I looked at Obsidian but found the UI to be heavy and slow, and lots of extra functionality I wasn't interested in. I've tried a lot of apps[1] but always come back to iA Writer.
[1] https://davidmytton.blog/the-best-note-taking-apps-for-mac-m...
I then tried various 'outliner' applications (Dynalist, Roam, Workflowy). With these kinds of apps, there is hardly any friction between document and content granularity levels; everything can be just one big tree / directed graph.
I finally settled with Dynalist [1] -- it recently introduced backlinks and is the far more mature, sleek and feature-rich option compared to Roam.
Roam is evolving quickly and has (what appears to be) little features that make it much more useful to me. More like a living, organic thing.
Eg: I really enjoy [[doing this]] to a concept and turning it into a page in less than a second...
Not sure why you’d need a WebGUI since the app appears to be cross-platform; personally, I tend to view web interfaces more as a last resort when there is no native app available. I’m not sure a WebGUI makes sense either if you’re storing your data in a local folder.
As far as I know, they don’t have a companion mobile app though, and that’s a deal-breaker for me too.
> Honestly thinking of going back to a notebook
Tried doing plain files again (via Notable because people said so, but that also doesn't really offer anything an editor with markdown support (cmd + click to open other file) doesn't offer itself), but by then I didn't really need to keep notes anymore (I passed beyond the "research" phase of my current project).
Markup, searchabilitiy, simple.
I prefer to link my sites organically inside of text in the page. E.g. on a page about Linux I'd write
GNU/Linux is an [operating system](../os) based on the Linux [Kernel](../kernel).
Tada, two backlinks without having to list them. If there's no link on the page it's not important.
Do you see problems with that approach?
> Make every link two-way navigable in published sites
> Make Foam notes more portable to different apps and long-term storage
(via https://foambubble.github.io/foam/materialized-backlinks)
Though I may be getting confused between the backlink differences that Foam, Obsidian, and Roam offer. A list of forward links per document would actually be useful as well, which IIRC neither Foam or Obsidian offer at the moment. Obsidian also shows the user 'unlinked mentions' along with explicit backlinks, which is also a nice feature that Foam lacks.
1. Write in Markdown. Stay as less markup as possible and tend towards plain text.
2. Organize/Categorized into folders; such as -- parenting, entrepreneurship, startup, homelab, books, etc.
3. Choose a really simple publisher (Jekyll for now). I sprinkle the least front matter in the header of each file. So, I can publish the selected ones online for me, friends, and family.
4. Let Obsidian look over it (it drops just one folder .obsidian), write with it sometimes. I don't want any tool taking over, chewing it and making smart decisions for me.
My focus are those files inside the folders. I should be able to just replace Jekyll or Obsidian in future without losing the integrity of my content.
it supports all the same features and built on top of vscode. use it to manage my personal knowledge base of 20k md files
check out MindForger - it is open source and quite simple to review as it's mainly one maintainer.
Google Cache - https://webcache.googleusercontent.com/search?q=cache:https:...
Google Cache seems to be just a proxy thats loading resources from the main site.
The lack of linking between notes is definitely the missing feature of Simplenote.
I like the approach here of trying to sync markdown files with iCloud
I regularly use Flying Logic for this sort of thing - I make a huge dependency graph, and then I create a collapsible group that encompasses the stuff that is active. The part of the graph that is "below" that window is stuff that is already done, and the stuff that is "above" that window is stuff that can't be started until I do stuff in the active window.
("below" and "above" can be reoriented to "left" and "right" if you prefer)
It looks to be non-free software, with a $-free licence tier for personal use.
No task management, planning, etc. Just plain reading, thinking and writing.
The quest continues...
Does anyone know if you can write your own plugins for it?
1. open-source. 2. One command away from my editor(I only use desktop for taking notes), also synced to my dropbox in case. 3. Amazing extensibility with plugins and custom scripts. Eg: bugwarrior + pomodoro(using Timewarrior, I've used this a lot when freelancing, not so much now) + some custom scripts. 4. Journaling - custom generated template of what I need to records/daily tasks etc. 5. Viewports in taskwiki are amazing, eg: for any project related notes I can do write `# Task | project:Pro1 and +PENDING` to list pending tasks in Pro1, you can also configure to use tags like development, bug reports etc.
I couldn't find another tool that is as extensible. The downside is you need to dedicate a weekend to understanding and setup!
One can add a function in rc files to commit and push to version control, if the `.task` directory is a git repo:
function tupd() {
git -C ~/.task commit -a -m "Update tasks $(whoami)@$(hostname)"
git -C ~/.task push
}
I also have a transformer script that generates markdown from a specific project's tasks and annotations, with headings and lists, and commit to a project's wiki that is a git repository, and push the changes.
I had an alias named `note`:
alias note='vi +star +'\''normal Go'\'' +'\''r!date +\%F\ \%T'\'' ~/workspace/notes/notes.md +'\''normal Go'\'' +'\''normal Go'\'
It's been a long time I haven't updated it. Not because I'm perfect, but because I'm not.
[1]: https://karl-voit.at/2020/07/22/org-super-links [2]: https://www.orgroam.com/
anyone got a link to what this "obsidian" is?
RTF etc?
Great UI, 'Mind-mapping' thing that everybody wants, backlinking to other notes, markdown, offline (so private notes).
Other alternatives - roam research, notion, etc
Using Notion.io because of that, but i would change to Obsidian when they have it.
There is a workaround of storing the md files in iCloud and using a markdown editor like 1Writer to open and edit them. But you loose some functionality with that.
You just have to sync a single folder. Google Drive/ Dropbox/ Syncthing, etc. Any solution works.
Use any markdown app on iOS - https://duckduckgo.com/?t=ffab&q=markdown+ios
This style of notetaking software help 'resurface' those information you even forgot you jot down long time ago and make the process of 'discover' new idea much easier, something like that.
Well, I mean it behave like our mind, just more tangible, in software form. Also, people forgot much more nowadays because of information overload...maybe that's why some people need to transcribe it into something and relieve their brain's load.
What I am saying is, that often we need an access trigger first even if the memory is technically there. Mindmapping is a way to rewalk the thoughts you once had and reenter a specific state of memory and mind.