[Evolution] Evolution 2.0 UI proposal

[Ettore Perazzoli <ettore ximian com>]

Hello!

Here at Ximian we have been brainstorming a bit about what happens
next in the Evolution world.  One of the ideas that has come up is a
substantial overhaul of Evolution's UI.

Since images speak better than words, here are the mockups for some
designs that Anna has developed: (this is just to give a very rough
idea of what it would be like; the icons and labels are not final, the
widgets are not the real ones etc.)

        http://primates.ximian.com/~anna/evo2/evo2_contacts.png
        http://primates.ximian.com/~anna/evo2/evo2_calendar.png
        http://primates.ximian.com/~anna/evo2/evo2_mail.png
        http://primates.ximian.com/~anna/evo2/evo2_tasks.png
        http://primates.ximian.com/~anna/evo2/evo2_navbar_shrunk.png

The most important changes are:

        * You no longer see all the types of folders at once.  You
          switch between calendar, mail, tasks and contacts by
          clicking on the buttons at the bottom.

        * The calendar allows you to see multiple calendar at once.
          Also you can subscribe to web calendars and see them in the
          pane on the left as well.

There are a few reasons for us to go with this design:

        * It kills the all-in-one tree view, which currently makes it
          difficult to reach for your calendar or contacts folders,
          since they are hiding between all the various mail folders.
          You no longer need to hunt for you calendar folder scrolling
          through the tree to see what your schedule is like, you just
          click on an easily accessible button marked "Calendar".
          Much better navigation.  (Please note that, although it's
          not obvious from the mockup, we would still have a mail
          folder tree, the same way we have it now.  Calendar, Tasks
          and Contacts, however, would be just flat lists.)

        * Killing the tree view also simplifies the architecture a
          lot.  Right now there is a lot of machinery in place to
          handle the tree, making sure that components don't step on
          each other's toes.  In particular, the handling of local
          folders is a maintenance nightmare, and also makes it very
          hard to provide the hooks that hackers need eg. to access
          Evolution's folders and do cool desktop integration hacks.

        * The shell's APIs would be drastically reduced to just
          a couple calls and it would become a lot simpler to 
          implement new components.

        * This design simplification would also allow components to be
          launched independently from each other.  We could
          potentially even launch the shell without certain components
          (e.g. launch only the mailer) if the user wants it that way.
          If we wanted to have separated apps a la OS X we could
          trivially do that too.

        * As I mentioned, it allows side-by-side calendar viewing,
          which increases the usability of the calendar manyfold.

On the other hand, if we go this way we are probably also going to
drop the following features:

        * The summary.  While the summary is neat, there is a general
          feeling (at least amongst the developers) that the mail and
          calendar summaries are not tremendously useful, and that
          weather and RDF and weather information is better suited for
          a specialized application.  Also we are trying to reduce the
          amount of code we have to maintain, and this seems like a
          good candidate for trimming.

        * The shortcut bar.  It's been shown that only a relatively
          small part of the Evolution user community actually uses it, 
          and we feel that it unnecessarily complicates the UI.  The 
          new design is much simpler to navigate anyways, and the 
          shortcut bar would add clutter and complexity, both in code
          and UI.  Also, it wouldn't be easy to implement in this 
          model without keeping some of the shell's complexity that
          we would like to get rid of.
          
Opinions?

-- Ettore