Part of the problem is Obisidian treats anything with a tab indentation as a source block, there’s a fix you can make in logseq to change how indentation behaves by swapping it from tabs to number of spaces. Drop this into your config.edn, and then try creating a new document with indented bullet-points and you should see them render correctly in Obsidian.
I think the possibility of collapsing all the blocks with a single click or a keyboard stroke would be a good solution, the document would become an outline of itself.
Chanced upon this FR. I’ve built a TOC plugin in the marketplace that can be opened in the right sidebar to facilitate easy navigation. Can take a look!
Maybe I am missing something. But in a large page, when you click outside the content and the CMD-UP a few time all blocks are folded till the main level, effectively the H1 level.
When I CMD-DOWN 1 time it opens the H1/H2 level.
And sometimes I even go further down but most of the time from this level I enter a Block by hitting the bullet and I can edit (also the sub-bullets for that H3 level).
And using the breadcrumbs at the top of the page I can quickly see where I am and navigate to a higher level … to deep dive in another place in the document.
So I guess an outline overview is already in the document.
@Miro — the Zettlekasten method recommends zettles remain small. So one way to avoid the massive pages requiring a table of contents is to avoid massive pages. And one way to achieve some division in large bodies is to use the hierarchies feature. You could for example, write a book where every chapter existed as part of a hierarchy.
I’m not disagreeing with the feature request, I’m just offering suggestions for effectively dealing with it in its absence. When I find a page that’s oversized, I generally feel it needs split apart.
I’d like to use Obsidian to do more longform writing inside the same system as my existing outline-based notes, but have a similar worry that my structure will be flattened when I return to my outline in Logseq.
Would a community plugin that told Logseq to parse Markdown heading levels (#, ## etc) as levels of an outliner hierarchy solve this problem? I guess you’d also have to work out how to tell it to deal with plain text blocks in between hierarchy levels (maybe always as children of the most recent header?).
Does a plugin like that exist, and if not how hard would it be to make one?