Comment 5 for bug 950730

Revision history for this message
Rik Shaw (rik-shaw) wrote :

In my opinion, re-work of the Applications lens is required... and this is certainly *not* low priority. Many new computer users that I have set up with Ubuntu (3rd world context) have greatly struggled with this. Only because of the 3rd party "stop gap" hack "classicmenu-indicator" have I been able to help people more easily browse and discover what applications are already installed.

Here are some additonal comments on design-rework help from my perspective:

1. Needs to be by category by default (I don't think that browsing a list of 129 Applications sorted alphabetically is useful to anyone).

2. Need to have quick keyboard navigation, or,

3. Just "hover" over a category name to see the apps under the category.

4. Currently have to manually "unclick" filters if want to see a different category. So, if click "Games" need to then unclick "Games" before clicking "Office" or else you get "Games + Office". This concept is just not efficient or intuitive for beginning users, and beginning users are the ones that desperately need a usable Application Menu.

5. In this case, Filters would be best located on the left side, so you hover and then the results dynamically show on the right. This is "standard main menu" style design, but obviously different than the dash "filter concept". Yet I argue that an Application menu / lens is a bit different / unique and it is essential!

Here are some examples:

AxeMenu is a good one (Gnome-Shell: https://extensions.gnome.org/extension/327/axe-menu/ but not their default of showing "favorites" but rather "Applications": also note when hovering over an App its description is showed in the bottom corner: again helpful as someone may not have a clue about what "Pidgin" does)

MintMenu (Linux Mint) is very similar to Axe Menu (maybe they share code??).

Lastly, Cardapio (https://launchpad.net/cardapio) is a good one to look at.

In summary, discoverability, especially for limited skill computer users, is essential and needs significant re-working over the current paradigm.