[wp-hackers] [GSoC 2014] Idea: Better Customizer

[Nick Halsey]

Hello again,

I'd like to outline another idea I have for GSoC and get some feedback. The
Theme Customizer has slowly been gaining traction among developers and
users since its introduction two years ago, and has received sizable
improvements in 3.9 with the addition of Widgets and improved support for
header images. There is still a lot more to do, so I've identified a number
of items that could be implemented as a GSoC project:

   - Adding menu management to the customizer (that is, editing menus
   directly, not just setting theme locations)
   - Adding theme-switching to the customizer (would require re-loading it
   entirely, but would be very useful)
   - Re-do the background image feature to work better with the new headers
   UI and incorporate the media library
   - Eliminate the existing header/background pages in favor of the
   customizer, as the customizer provides a better solution for these (only
   when customizer is available; wouldn't make sense to do this for widgets,
   menus, or themes)
   - Add more built-in controls; for example, a select-type control that
   allows users to select a value based on an image (useful for colorschemes,
   layouts, etc), a range/slider input type, basic number input type, etc.
   - Better documentation: fill-out the codex/handbook pages, perhaps
   create an example plugin with example usage of different levels of
   difficulty
   - Address the problem of losing changes when leaving the customizer
   (without implementing autosave/revisions for theme_mods, which would be an
   entire project in itself)
   - Better showcase the customizer in bundled themes: development of
   features for Twenty Fifteen (near end of project if timelines allow),
   improvements to older bundled themes
   - Improve access to the customizer from the front-end: perhaps make it
   more like a toggle in/out of customize mode? Remove or deep-link the
   appearance menu in the front-end toolbar.

I've created trac tickets related to some of these ideas previously. From
the above list I'd like to select a few items that together would be about
the right scope for a GSoC project; for example, either menus or themes
could take about half of the time to implement in a core way. If anyone has
ideas as to what should be prioritized (or if any are invalid), that would
be very helpful for creating a specific/final proposal.

Parts of this would be developed in plugins designed for eventual merging,
other parts would be best implemented as core patches. Any feedback you
have is much appreciated!

Thanks!

Nick Halsey
IRC / WordPress.org: celloexpressions
http://celloexpressions.com/