Bug 434904 - Implement a triangle menu filter for Kickoff's categories sidebar
Summary: Implement a triangle menu filter for Kickoff's categories sidebar
Status: RESOLVED FIXED
Alias: None
Product: plasmashell
Classification: Plasma
Component: Application Launcher (Kickoff) (show other bugs)
Version: 5.21.0
Platform: Other Linux
: NOR wishlist
Target Milestone: 1.0
Assignee: David Edmundson
URL:
Keywords:
: 434902 441502 (view as bug list)
Depends on:
Blocks: 433819
  Show dependency tree / graph
 
Reported: 2021-03-24 20:25 UTC by Nate Graham
Modified: 2021-09-29 12:42 UTC (History)
10 users (show)

See Also:
Latest Commit: https://invent.kde.org/plasma/plasma-workspace/-/commit/36ca390f56115ae9cbb1423cd00976e8dffec4d9
Version Fixed In: 5.22
Sentry Crash Report:

Attachments
Add an attachment

Note You need to log in before you can comment on or make changes to this bug.
Description Nate Graham 2021-03-24 20:25:15 UTC 
Right now the category sidebar used a delay timer to prevent unintentionally activating a new category when moving the mouse to something to click on, but it is not sufficient as it causes

1. The delay is annoying; see Bug 433819.

1. People request different sidebar modes to work around it (Bug 434902); see https://community.kde.org/Get_Involved/Design/Lessons_Learned#Constant_redesign_proposals_is_a_warning_sign

My sense is that if we implement a triangle menu filter; all of these complaints and requests will cease as if by magic, because the filter will end the root cause problem of accidental category switching.
Comment 1 Nate Graham 2021-03-24 20:25:24 UTC 
*** Bug 434902 has been marked as a duplicate of this bug. ***
Comment 2 Nate Graham 2021-03-24 20:26:28 UTC 
David, am I mis-remembering or did you volunteer to write one for this?
Comment 3 David Edmundson 2021-03-25 18:02:28 UTC 
POC: https://invent.kde.org/davidedmundson/trianglefilter
Comment 4 Bug Janitor Service 2021-03-25 23:47:08 UTC 
A possibly relevant merge request was started @ https://invent.kde.org/plasma/plasma-desktop/-/merge_requests/396
Comment 5 David Edmundson 2021-04-11 21:31:13 UTC 
*** Bug 435630 has been marked as a duplicate of this bug. ***
Comment 6 MK 2021-04-21 12:53:15 UTC 
As a professional UX designer I see two ways to solve the hover problem:

1. Give users the choice as to whether they want to open categories on hover or on click

2. Implement an option where an expanded alphabetical list is the default view in the left panel and favourites are kept in the right panel so no accidental hovering can happen
Comment 7 Nate Graham 2021-04-28 14:34:36 UTC 
Fixed by David Edmundson in Plasma 5.22 by implementing a Triangle menu filter for the categories list with https://invent.kde.org/plasma/plasma-workspace/-/commit/36ca390f56115ae9cbb1423cd00976e8dffec4d9 and removing the hover delay in https://invent.kde.org/plasma/plasma-desktop/-/commit/d3c1366ff57775c4ee5a6dc9e105da118a0d74a0!
Comment 8 Nate Graham 2021-06-08 16:25:41 UTC 
*** Bug 438234 has been marked as a duplicate of this bug. ***
Comment 9 Paul Worrall 2021-08-24 21:28:39 UTC 
*** Bug 441502 has been marked as a duplicate of this bug. ***
Comment 10 Nate Graham 2021-08-30 16:11:26 UTC 
*** Bug 441746 has been marked as a duplicate of this bug. ***
Comment 11 Nate Graham 2021-09-08 21:09:04 UTC 
*** Bug 442108 has been marked as a duplicate of this bug. ***
Comment 12 Nate Graham 2021-09-28 20:31:26 UTC 
*** Bug 443082 has been marked as a duplicate of this bug. ***
Comment 13 flan_suse 2021-09-28 23:20:27 UTC 
(In reply to MK from comment #6)
> As a professional UX designer I see two ways to solve the hover problem:
> 
> 1. Give users the choice as to whether they want to open categories on hover
> or on click

^ I agreed with this, and/or just place the most-frequently clicked targets (e.g, favorite apps) in a single pane above the cursor. If they insist on a split view, then have the most-frequently clicked targets positioned in the pane *directly above* the cursor; prioritize *THAT* pane over the less-often-used categories and sub-menus.


I demonstrated this in two videos, while using Plasma 5.22.5, which uses the well-intended "TriangleMouseFilter" attempted "fix".

VIDEO #1: Legacy method, cleaner, fewer mistakes, easy targets

https://imgur.com/a/LN7uXUz

---

VIDEO #2: New method, more cluttered, less intuitive, more mistakes, requires conscious effort

https://imgur.com/a/iNH9yAF

---

This new launcher requires more conscious effort, whereas the legacy launcher was more straight-forward with larger, breathable commonly accessed targets.
Comment 14 Tobias Leupold 2021-09-29 05:39:33 UTC 
Video #2 demonstrates EXACTLY what happens to me when using the new launcher! And it makes it frustrating to use! Thanks for visualizing this!

Gladly, you can still install "Legacy Kickoff" to keep the old (way more usable) menu ...
Comment 15 flan_suse 2021-09-29 12:42:24 UTC 
(In reply to Tobias Leupold from comment #14)
> Video #2 demonstrates EXACTLY what happens to me when using the new
> launcher! And it makes it frustrating to use! Thanks for visualizing this!
> 
> Gladly, you can still install "Legacy Kickoff" to keep the old (way more
> usable) menu ...

There was nothing to "improve" from the legacy design.

* Larger, easier targets (the whole point of a quick-access menu launcher)

* Vertical mouse cursor movement without risk of collision nor interruption.

* Bottom panels that display *desired quick-access* content... which are reached with a simple upwards vertical mouse movement.

* Less reliance on sub-menus.

* Most-often used favorites/shortcuts were easy to click (yet if you need to dig further you still can navigate by categories and sub-menus)

---

The paradigm shift with the new menu design comes off as a "change for the sake of change".

* Forces arbitrary "diagonal" or zig-zag movements just to launch a favorite app.

* Targets are not located directly above the mouse.

* Requires overly complex "fixes" such as the "TriangleMouseFilter" that is only needed *because* of the inherent design flaw. (Clean designs don't require predictive filters to try and decipher the user's intended mouse movements).

* The split view (with the desired targets on the opposite side) requires user-consciousness. You can see from the videos that in the Legacy Launcher I didn't even have to worry or think about it... just wham, wham, wham, over and over, clean and easy to use, simple and fast. With the new launcher you have to be more aware *as* you are moving your cursor "Do I want to use the categories or launch a favorite app?" The reason is because your *initial mouse movements* will determine whether or not the categories get activated or if they ignore your cursor.

* The above is also because it's too crammed. In the Legacy Launcher your intention is clear before doing anything: tabs on the bottom will present you with LARGE menus of: Favorites, Applications, Places, Etc. In the new launcher it's trying to fit too much into a small area.

To illustrate this last point in the extreme:

Imagine if the launcher split into THREE panels: Categories, Favorites, Places

Not three tabs on the bottom, but three split *panels*. Imagine trying to quickly use the launcher regularly in your day-to-day. But don't worry, they'll come out with another "fix" that measures the acceleration and deceleration of your mouse cursor to predict which of these three panels you intend to click on.

I know it sounds silly, but that's the point. The Legacy Launcher *design* was clean, and required no fancy fixes nor conscious awareness to use it intuitively.