Open

Desktop App – UX Consistency, Performance & Accessibility

First, I want to reiterate that I think Capacities has something genuinely unique. The object model, relationships, and structured approach to knowledge are excellent, and they solve a real gap in the market.

My overall feeling is that the foundations are now strong enough that the biggest opportunity isn't adding more functionality—it's refining what already exists. The product is becoming increasingly feature-rich, and I think polishing the experience across the board would deliver more value than another wave of new features.

Editor Reliability

The editor occasionally behaves unexpectedly.
Web links don't always retrieve metadata correctly.
Images and previews sometimes fail to load.
Some editing interactions feel inconsistent.

UX Consistency

One thing I've noticed throughout the application is inconsistency in interaction patterns.

Examples:

Some features open in modals.
Others open in full-screen pages.
Some modals can expand to full screen while others cannot.

These inconsistencies make the application feel less cohesive than it could. A more consistent UX across the product would make it feel significantly more polished.

Search & Discoverability

I couldn't find a search function within Settings.
As the application grows, finding configuration options becomes increasingly difficult.

Performance with Large Knowledge Bases

After importing 3,000+ notes:

Opening the collection became noticeably slower.
Initial loading takes considerably longer.
Infinite scrolling sometimes hangs indefinitely.
I occasionally have to navigate away and back before more items load.
Large collections could benefit from better virtualization and performance optimization.

Accessibility / Text-to-Speech

This is one of the biggest usability issues for me personally.

I consume much more content by listening than reading and regularly rely on the native macOS and iOS text-to-speech features.

macOS

When using the native Speak command:

It begins reading parts of the application's interface before reading the selected text.
It doesn't appear to expose the selected content correctly through accessibility APIs.
The result is that native text-to-speech is effectively unusable.

iOS

The situation is even more limiting.

Because the editor selects blocks instead of text, the native iOS text-selection menu never appears.
That means the built-in Speak feature cannot be used.
There is also no equivalent accessibility feature provided within the app.

Supporting native accessibility properly would benefit not only users with accessibility needs, but also people who prefer listening to long-form content.

Global Layout Preferences

I couldn't find a way to enable Wide Layout globally.

Ideally this should be configurable at:

Workspace level.
Collection level.
Object type level.

Enabling it manually for every note becomes impractical when you have thousands of objects.

Overall, I think Capacities is approaching the stage where user experience becomes the biggest differentiator. The vision is already compelling, and the feature set is already powerful. Now I'd love to see the team focus on making every interaction feel fast, consistent, polished, and effortless. That's what will turn an excellent product into one that people genuinely enjoy using every single day.

2 Comments

Omran·3 months ago

vision a Thanks! I actually like your rephrasing better than mine. What I was trying to express was:

"...try to keep UX/UI in line with growth of functionality... at some point make the basics primary..."

I think that's a more balanced way of putting it. The feature set is already compelling—that's why many of us are here. My hope is simply that the next phase puts equal emphasis on polish, consistency, performance, and the overall experience of using those features day to day.

Again, thanks for taking the time to read through the whole thing. I'm rooting for Capacities 👏

vision assemblies·3 months ago

Even though I like this post reiterating some points from old debates, I also think the prop "Capacities is approaching the stage where stage where user experience becomes the biggest differentiator" is quite sharp.

I think UX consistency is a lingering topic spread across different topics here. And if I rephrase "Editor Reliability" against "Editor flow" it also touches on a global topic otherwise "lost" in many singular topics/tickets here (e.g. on long form writing, block behavior etc.).

I also think "Layout Preferences" is a similar field identified correctly as something that deserves collective reflection, and gets adressed in multiple forms/tickets, often unconnected. (Width settings of text, global vs. space wide controls, minimal customization of interface, accessibility issues etc.)

So, while this can be read as a idiosyncratic list, I would support it and read it as a renewed write up of something along lines often heard over time: try to keep UX/UI in line with growth of functionality, improve coherency where a) it wasn´t perfect from start, b) even more where it is challenged by function growth, c) at some point make the basics primary, and put them before (head-spinning) innovation… bec things otherwise loose their 'joints'.

So for me this is a very worthwhile "collection ticket" talking about things beyond individual functions, or even this or that usage scenario. So: supported!

– The one thing I would add from my evangelion: priortize things that bend Capacities away from attaining too much of a database feeling – and strengthen things that are geared to other modes of working than recording, auto-ordering, searching, retrieving, listing etc.