> Speaking of future-proofing, Markdown is as close as it gets to having interoperability with the notes. Obsidian is at its core just a Markdown file editor, which means your notes are stored as plaintext and easy to export. There is a bit of Obsidian-flavored syntax (e.g. bi-directional links [[...]]), but these are becoming standard in note-taking. Many note-taking apps claim export functionality, but at the end of the day they're not incentivized to give you your data in a format that will work with other editors.
This is actually the reason I don't use Obsidian. It's built on markdown, but it is not markdown. Things may be different now, but when I gave up on Obsidian, there was no general purpose exporter - you either use their editor or you lose much of the useful information in your notes. This would not by itself be a big problem except for the fact that Obsidian is a standard piece of proprietary software. It genuinely solves nothing in regard to future proofing.
The notes are very much markdown files. There is a very small number of extensions the Obsidian developers created (e.g., a syntax for linking to a specific paragraph within a note) that are not markdown, but it's very easy to not use them.
I'm using the exact same folder of markdown notes in parallel with Obsidian, The Archive, 1Writer, Calca, TableFlip, Python scripts, and Keyboard Maestro macros I have written and everything works flawlessly together.
To me, the killer feature of markdown notes is not the future-proofing, but this kind of seamless interoperability.
I don't much want to get into a debate over semantics, but markdown plus extensions with useful information (like links) does not qualify in my book as markdown. Sure, if you want to limit yourself to the parts that are interoperable go ahead, but then you're not using some of the best parts of Obsidian.
In that case, GitHub doesn't support Markdown, then, considering it shits the bed on one of the most crucial parts of the Markdown design: line-breaking behavior to preserve the readability of the "raw" form.
(Watch now as everyone rushes in to try to say that this behavior is an exception, as if you just brought up that their favorite uncle has some unsavory qualities.)
Agreed. And what is exactly expected to be done, if sticking to strictly "Markdown spec"? You can't do anything meaningful with such a small subset of language features, where as Obsidian and Obsidian Plugins are all about extending Markdown to provide additional features ontop of the language, but stored in plain markdown.
This would only be non-Markdown if it fundamentally broke something in Markdown. Ie lists no longer worked, or *bold* was used to link documents, etc.
I'm not arguing against anything you've written. None of it changes the fact that you're dumping your notes into a format that only works properly with a single proprietary app.
Your Obsidian vault is a folder of Markdown files that use [[links like this]]. You can load your Obsidian vault folder in the open-source version of VSCode with either the "Foam"[1] extensions or Markdown Memo[2] VSCode extension. [[This style of link]] works great with either. I think I also saw a new feature (or plugin?) for VimWiki that allows [[links]] to work even when the target is in a different subfolder, but I haven't tried it myself so don't quote me on that. There are also other programs that use this style of linking.
You can alternatively set up Obsidian to use [regular markdown links](regular.md) , it's just not the default setting.
What's the definition of properly though? It opens and works fine.
If you're saying "Features that only exist within Obsidian won't exist in other apps!" then.. yea, that's true. However features that _do_ exist in other apps, like the loosely spec'd `[[link]]` will work in many apps. Same with latex, github extensions, etc.
> markdown plus extensions with useful information (like links) does not qualify in my book as markdown
Almost every useful markdown system extends markdown arbitrarily. Markdown is standardized much like SQL is — a standard exists, but it standardized very little; mostly just defining the look & feel of the extensions
I am really curious: Why does restructured text get so little mindshare? I used it once to make an ebook into a website, using sphinx. The format is easy to write and read and it has extensibility as a builtin.
For me, reST had the big issue of not being as quick to type as markdown. It's a better standard, that's true, and especially the table features are A LOT more useful than MD, but I've always felt a lot more mental resistance and less compatible tooling. I basically picked up MD on the side (on SO and GitHub), but reST I would have to learn.
If you wanted strictly markdown then you could just stick to markdown, but the set of popular features which extend markdown are developing consensus across the ecosystem, such that you can get most of Obsidian's functionality with less than a handful of popular VSC extensions.
In that sense Obsidian is just a fancy viewer for documents which can be edited in VSC.
And with regards to vanilla markdown (which would exclude the likes of Github or Gitlab, or the most popular extensions on VSC), personally, I wouldn't be satisfied with the exclusion of Latex which is already in widespread support across markdown supporting apps.
Community markdown is a moving target because John Gruber's initial vision is frozen in time whilst the demand for innovation is ballooning. It's unfortunate because the trademark for markdown is in some ways in the same ballpark of value as the community buy-in for the technology.
The devs have been pretty busy with the mobile apps lately and the desktop version hasn't hit 1.0 yet. But they say they're now currently "working on" the markdown export feature (previously the mobile apps were in that column)
On the open source front you get zettlr.com, a pure markdown WYSIWYM editor with image preview, pandoc integration for exporting groups of pages into whole documents, support for Zettlekasten workflows, and academic references management.
It's a one-man effort by a guy who created it for working on his PhD, but it's quite robust and usable despite a few small flaws (who should be corrected in version 2.0, likely coming during the year).
it's close enough that you can write your own parser to some other format with a little bit of hacking, in fact that's what I did to port my notes from obsidian to logseq a while ago.
Of course it's a little bit of trouble but I imagine most people won't want to switch their notetaking apps that frequently.
This is actually the reason I don't use Obsidian. It's built on markdown, but it is not markdown. Things may be different now, but when I gave up on Obsidian, there was no general purpose exporter - you either use their editor or you lose much of the useful information in your notes. This would not by itself be a big problem except for the fact that Obsidian is a standard piece of proprietary software. It genuinely solves nothing in regard to future proofing.