Agreed, better PDF support would be great. The main feature I constantly need is search. But the 'open in external application' option gets me there pretty quickly, so it's not that much slower - but it would be nice to not have to switch apps.

What do you prefer Obsidian over Logseq for?

As an outliner, Logseq is too opinionated about how I am supposed to use it. Obsidian is less strict and allows me to follow whatever principle I want to which extent I want.

I can mix a bit of Zettelkasten here, some daily notes there, and some 'old-fashioned' folder structures for projects to my heart's content.

(not op) Logseq is open source and really great in the way it connects notes and highlights pdfs, but the sync part is too “asynchronous”. It works well if you use a single device, but as long as you add something else you have to manually reindex and refresh a lot.

Obsidian is less opinionated on the txt file format and folders too, so I consider it more future-proof.

The sync part I solved with syncthing.

I started with logseq and now obsidian doesn't work for me anymore. Tried to switch but I am into this small self containing bits now. Plus journal with timestamps

Syncthing is very good, but the best iOS implementation, Möbius Sync[1], is not comparable with the Android one (the OS limits the syncs, you need a regular notification and so on).

The great things about Logseq are his weakness for me. Everything is so interconnected (you can say: "here put the paragraph of this other note") that I sometimes lose confidence in the system. It becomes too complex. With Obsidian I know that a note is a file. Less convenient but simple and reliable.

Logseq really excels with his outlining mode, I miss it (but I don't like the way it saves states in the markdown file). It has some problems with the code blocks too.

[1] https://www.mobiussync.com/

Can you elaborate on the syncing part? Last I tried Logseq, syncing across devices via iCloud was very unreliable. So I gave up. Although I would have loved to replace Roam with Logseq back then. Now I stick to Emacs + orgmode and Obsidian for plain markdown files.

I have a folder in syncthing called "logseq". This folder holds the logseq files. I have logseq-sync on my phone so it syncs the folder. I also have a Cloudserver which is the spoke and all other devices (2 laptops, the phone) sync it.

Also not OP, but personally I have a few gripes with Logseq:

- No export to PDF. There's a community plugin, but it's not great. The workaround is to export to HTML and print to PDF, but there's no real iOS option there.

- Managing images and other attachments are a mess. Using the "upload an asset" method gives it a random filename that if the app fails to save the page correctly, you either dig through the folder structure to find the random file name to link manually or you re-"upload an asset" creating a duplicate with a new random name. This could be alleviated if it was more stable or with a file picker with thumbnails to find previously "uploaded" files

- Pages fail to save correctly more than I'd like. I have no idea what the cause is, but it happens frequently on every platform I've tried.

- Page title changes don't propagate correctly sometimes, causing orphaned pages where it's a coin flip whether the page with the older title holds the content or the new one, leaving the other empty.

- Each page needs a unique title. I like how Notion allows multiple pages with the same title and are organized based on which parent page they're embedded or created in. I imagine Notion randomizes the actual file name similar to how Logseq already does for "uploaded" assets, so this could be alleviated if Logseq did the same. It could potentially alleviate the previously mentioned issue and it seems to me like the most logical method of handling this particular type of non-directory organizational structure.

- Their E2EE sync service is not yet ready, so no real mobile sync outside of iCloud (I use DropBox).

- Their documentation is terrible. There's tons of undocumented features, like admonitions, and the existing documentation is horribly structured, which is ironic since the documentation uses Logseq itself and the whole point of the app is to structure content.

----

Side note, since we're on the topic of personal knowledge bases and note taking, my personal dream app is Obsidian with Asciidoc support instead of Markdown. A lot of the extra features they add to markdown are part of the Asciidoc standard, like admonitions and document-to-document cross-references, it would potentially make the backend easier and the content more portable with page attributes like specifying an attachment directory, and some features are simply more flexible/powerful like tables.

I still use AsciiDoc to create PDF documents that require more flexibility, like table spans and nested ordered lists (Obsidian's markdown uses just 1,2,... instead of changing to e.g., a,b,... for a nested level). My current workflow is typing it up in VS Code, converting to DocBook with asciidoctor, then converting that to a LaTeX PDF using pandoc. The result is a professional, academic-like PDF, but the workflow is a bit of a hassle and I'd prefer to do all of my document typing in Obsidian since it's so nice to use.

If I had more free time outside of my CS master's program and thesis work, I'd learn JS/TS to attempt to create a community extension that added AsciiDoc support to it and support for exporting to HTML and DocBook (and basic PDF since I'm pretty sure Obsidian uses an HTML-based PDF export anyway because CSS themes affect the look of the export), even if I still needed to use pandoc to convert to a more professional LaTeX PDF. I'm sure the VS Code AsciiDoc extension as reference and asciidoctor.js could get one pretty far.

Sorry for the rant. I've just been itching for a AsciiDoc-based note-taking/PKB for a long time.

Fellow AsciiDoc supporter. I want an AsciiDoc note taking app so badly too. So many awesome features in AsciiDoc that blow markdown out of the water. I use asciidoc-pdf for doc/report generation and the workflow is so smooth.

- Includes/embeds (reference your source code by line range(s)

- Complex table support + the ability to embed CSV (automatic headers)

- Frontmatter as a first class citizen

- Macros (Variables) that can be referenced across documents

- Numerous Diagram parsing libraries (embed pretty much any diagram-as-code language)

I've had the same thoughts on building a Dendron type extension for AsciiDoc (AsciiDoc vscode plugin is fairly robust). Really would just need to hammer out some front matter parsing to get basic functionality.

Thanks for the reminder to check out asciidoctor-pdf! I forgot about it because I was using macOS's default, out-of-date Ruby version for the longest time and the gem required a newer version. I've since figured out how to install and change the default Ruby installation though, so I need to check it out. It's certainly a pain getting my current workflow set up on a new machine (e.g. installing TexLive takes forever for the latex-pdf support in pandoc).

ReStructuredText seems like it would be a better contender because it's the standard markup language for the most popular programming language.

It's also natively supported in VSCode (unlike AsciiDoc).

I honestly haven't played with ReStructuredText. I'll take a peek.

AsciiDoc includes are very powerful though, being able to populate your tables/code blocks from external data sources (filtering for certain lines/ranges). Also Tables can have complex structures (merged cells etc).

I feel like AsciiDoc was designed around writing technical papers than code documentation. It's essentially an opinionated wrapper around LaTex but less complexity. reST looks like it can produce great code documentation. AsciiDoc lets me cover both code documentation and customer facing documentation/reports with one code base. But I'll definitely continue to take a peek at reST. Although I think it's use case is more for embedding documentation directly in code. It's probably just my "I know the AsciiDoc toolchain well" bias though.

I was also kind of sad that Logseq didn't really have a vim-mode.