Created attachment 1116466 [details] screenshot of keyboard selection dialog under Japanese install Description of problem: It is a common use-case to have a US keyboard ("global keyboard") outside the US or in general a different keyboard to the country default, specially for power users and users in international companies, etc. Currently it is too hard to find the most common keyboard layouts from the long list of keyboard offered in Anaconda. For example when searching for the "US" keyboard, just filtering the list with "US" still lists many layouts. It is also not obvious that filtering with "english" works. The quickest way is the filter with "(us)" but it can be hard to find the parens keys on a non-US keyboard if one doesn't remember where they are... Also users should not need to know that they to search for "(us)"... I would like to suggest listing the the most common keyboard layouts first (before any filtering). This will make keyboard setup much easier for international users installing Fedora. Steps to Reproduce: 1. Install Fedora in Japan with US keyboard (or non-native keyboard anywhere) Actual results: 1. Defaults to JP keyboard and hard to find US keyboard from the long list of all layouts without careful filtering. Expected results: 1. Should be very easy to setup a US or other common keyboard types since they are frequently used.
Note all the obscure keyboards listed initially in the dialog screenshot.
Same question from bug 1158370: how do you propose that we determine what is and is not a common layout? gnome-initial-setup uses a hard-coded list of "common" languages to determine which layouts to present to the user first. anaconda is not as able as gnome to make the determination of should or should not be considered a common or support language, and this solution is not appropriate.
*** This bug has been marked as a duplicate of bug 1158370 ***
(In reply to David Shea from comment #2) > anaconda is not as able as gnome to make the determination of should or > should not be considered a common or support language, and this solution is > not appropriate. Why is it not appropriate? The i18n can help with the determination I think. I think it would be an iterative process - we have a reasonable idea of some of the most commonly used languages/regions and which layouts they use.