I won a lot and didn't lose anything. Obsidian plus a few plugins beats Joplin on every level. And I can edit all notes without redundancy by exporting from a Joplin database with nvim, iA Writer, find, grep, cat and millions of other tools.
> Obsidian plus a few plugins beats Joplin on every level.
Not really. Obsidian has its shares of problems too, and most of them originate from using Markdown. Markdown is a freeform text-format, and works very well for writing text, but it really sucks for data and structured content. Most plugins and features in that area are very brittle and overspecialized, working only well enough in their specific use case. And gosh, Obsidian has really a huge amount of plugins for data-handling. At some point, it was so bad that there were multiple competing task-plugins which broke each other just because they had different formatting for dates.
> And gosh, Obsidian has really a huge amount of plugins for data-handling. At some point, it was so bad that there were multiple competing task-plugins which broke each other just because they had different formatting for dates.
Depends on how you see it. I think it's great that they have a free market of open-sourced plug-ins that you can donate to. I'm actually impressed of how rarely they break each other.
> Not really. Obsidian has its shares of problems too, and most of them originate from using Markdown.
Aha. Which problems do you mean?
> Markdown is a freeform text-format, and works very well for writing text, but it really sucks for data and structured content.
Joplin is using md to. And if Joplin does a good job on "data" and "structured content" (whatever you mean by that) by separating that in their DB, it's a big NO for me since it's a closed silo.
> Most plugins and features in that area are very brittle and overspecialized, working only well enough in their specific use case.
Aha. I don't think so. Which authority says that? And even if It's like that, my markdown files would survive everything, since they are a) in git. https://github.com/denolehov/obsidian-git and b) easy to fix since it's a text file. Gosh!
> And gosh, Obsidian has really a huge amount of plugins for data-handling.
And gosh, this is a good thing!
> At some point, it was so bad that there were multiple competing task-plugins which broke each other just because they had different formatting for dates.
Installing multiple task plugins shows that something is "broke" on the user side. It's not the fault of Markdown or Obsidian.
The way it's handled can make the difference in control.
> by separating that in their DB, it's a big NO for me since it's a closed silo.
Joplin is using a popular open database with a healthy community and good tooling. It's as open as markdown. Maybe not for you, when you lack the knowledge, but markdown is similar closed for anyone not understanding filesystems and editors.
Good for you, but that is very low level in terms of data-handling. Dataview is really just an elaborated search, there is no good level of interaction. Datacore, the next project of the Dataview is supposed to bring this, but it's not even usable yet AFAIK. Coincidental, the Obsidian-devs are also working on that front, but nothing is finished yet.
That's useless when the app itself is not working. And even worse if you are not realizing the errors early.
> Aha. I don't think so. Which authority says that?
My own experience. I've tested enough plugins over the years to know their dark corners.
> And even if It's like that, my markdown files would survive everything
The thing is, technically you are not even having proper markdown, but a fork with some extensions of Obsidian. So some features of your parts might break when switching away from Obsidian. And the reason for all this is also because markdown is lacking definitions for what obsidian-people are doing with it. Coincidentally, this seems also one of the reasons why Joplin is using a database.
> And gosh, this is a good thing!
Not if they all suck.
> Installing multiple task plugins shows that something is "broke" on the user side.
Sure, because the plugins are lacking features, its the users fault... Maybe some users have just very different levels of requirements from you.
