Accessibility: Focus order sequence issues and text scalability fixes

[AnaFerreira015]

Hi! I’m Ana Ferreira, a Master's student in Computer Science.

As part of my research on mobile accessibility, I am analyzing open-source Android applications to help identify and correct accessibility barriers so users can utilize assistive technologies (like TalkBack) more effectively.

I’ve been exploring Tridenta and identified several accessibility issues across the main screens, so I wanted to propose a fix.

1. The Issue

On the Main Screens, the navigation sequence for assistive technologies presents the following behavior:

• Current behavior: The Focus Order (the sequence in which elements are highlighted by TalkBack) does not follow a logical or intuitive path. It skips certain interactive elements or jumps between sections in a way that doesn't match the visual layout (violating WCAG 2.4.3 Focus Order).
• Impact: Users who rely on screen readers or external keyboards become disoriented, as they cannot predict where the focus will move next, making navigation inefficient and confusing.

2. Proposed Solution

To fix this, I suggest explicitly defining the traversal order in the layout XML using android:focusable="true" and android:nextFocusForward attributes where the default linear order fails to match the visual hierarchy.

Code snippet / XML suggestion:

<View
    ...
    android:focusable="true"
    android:nextFocusForward="@+id/next_logical_component" />

3. Additional fixes (Need your input)

I have identified 3 other accessibility improvements for Tridenta.

To avoid spamming your notification feed, how would you prefer I submit these?

• Option A: List them all here in this issue (I can update the comment below).
• Option B: Open separate issues for each specific problem.

Please let me know your preference, and I will proceed accordingly.

Best regards,

[Stypox]

Thank you for looking into this and sorry for the late reply! In my experience whenever there are accessibility issues it's often easier and faster to propose a fix directly rather than explain the problem. Would you be up to make a couple pull requests to fix the issues?