An architectural comparison of Obsidian Canvas and Heptabase | Hacker News Reader

An architectural comparison of Obsidian Canvas and Heptabase

fluxlinks.com

36 pointsezrasu15 comments

Bit surprised to hear that it doesn’t link notes in Obsidian. It seems strange to go to the trouble of implementing a visual graph later on top of a vault but then not even offer an optional setting to have those links create e.g. wikilinks.

The site is incorrect. This was added in Obsidian 1.12:

> Added detection for backlinks in Canvas files. They now appear in the Backlinks view, and count as links in the Graph view.

https://obsidian.md/changelog/2026-02-27-desktop-v1.12.4/

Excellent, thank you - that makes much more sense.

The classical A vs B, but recommend C at the end post.

If Obsidian on the desktop has plugins installed, then opening the same vault with the iOS app becomes absurd performance of "wait a minute", "load another vault". And it takes so long I forget what I wanted Obsidian for. I don’t even know at this point if it loads on iOS at all. Because this "loading" is taking forever. And I am using a few top popular plugins that everyone is using. Obsidian's way of working with plugins on iOS defeat the purpose of using Obsidian. My guess is plugin architecture was decided long before the iOS app was developed. But still, absolutely nothing can be done?

Seems to be an AI-generated ad, and not particularly well researched. Obsidian has supported links in Canvas since 1.12

> Added detection for backlinks in Canvas files. They now appear in the Backlinks view, and count as links in the Graph view.

https://obsidian.md/changelog/2026-02-27-desktop-v1.12.4/

Tinderbox has been providing this kind of card-centric structure for a very long time now.

At a fundamental level, they are indeed the same; much like Notion, Obsidian, or Roam Research—when you strip all these note-taking apps down to their core, they are essentially no different from an ER diagram.

Is this spam?

I’m not sure—I just posted a blog post that I’d spent two or three days writing. Could you tell me why [there’s an issue]? I am very sorry.

If you say you wrote it, I'll read it. I commit to this.

We can all throw questions like this at our LLMs, and the objections you'll see here revolve around having our time wasted by folks who believe their LLM output is an accomplishment; we're getting sick of it.

—

I'm curious why would you think is spam.

It’s not and seems very human-written. Thanks to OP, I was actually dealing with this topic at work recently, so it was very helpful to read.

Glad it helped! Really appreciate you chiming in.

Reply on news.ycombinator.com