[crossfire] skill window

[Kevin Bulgrien]

> Questions:
> 
> 1) Can the skills and experience pane now disappear?

Probably, but the answer may not be as simple as you are thinking.  Not only
that, but I think that even though there are a lot of people that like to
hide detail, there are some that like everything visible on the HUD, and
that people should not make sweeping changes that hide information into
separate windows for all layouts.  I for one, generally want a lot of info
on the screen all the time.  I don't consciously read it all, but its being
there lets me feel what's going on better because I rely on subconscious
feedback a lot.

The gtk-v1 series layouts (gtk-v1, v1-redux, and un-deux) are deliberately
based on the gtk-v1 UI format.  I should think it would be kind of odd to
remove the look of the client the layout was to emulate.  But, the gtk-v2
emulation of the gtk-v1 only went so far as gtk-v1 fonts were much different
than gtk-v2, so the look/feel is not 100%.  To preserve the intent behind
these layouts the underlying code would have to remain to support them.  I
think libglade is not offended when a .glade file doesn't have all the items
the code tries to connect.  It think it does the right thing at run-time.

Now, the original skills layout was pretty ugly and the underlying code did
not manage the panel very well.  There are apparently issues with some of
the code - going an old bug list on SF - buffer overruns or some such, so that
lends some credibility to removing the code support, but I won't be doing that
because I really liked the gtk-v1.  I did the libglade conversion because
people kept threatening to shutdown the gtk-v1 and I was not about to play
the original gtk-v2 layout as it just did not work on my small monitors.

I did intend not to, and continue to avoid messing with the gtk-v2 original
layout.  It was the author's design and I did not want to presume to muck it
up for him since he probably designed it that way because he liked it that
way.  What happens to gtk-v2 is not really not something I particular feel
concerned about except that I'd like to see people respect the author's
preferences.

I definitely support pulling it out of sixforty.glade and oroboros.glade
because the point of those layouts was to crunch screen space.  Note that
sixforty has some really good ideas, but its not that successful at making
640x480 work yet.  Maybe removing skills from it would help break that loose.

From there it starts to get a bit more personal.   All the designs except
gtk-v2, are "mine".  A few I don't care a lot about as there is a bit too
much similarity on some of them, so I would invite some diversity added, or
the field thinned if removal makes too many of them even more similar than
they already are, but I'd pose the reminder that the point of the libglade
rewrite was to give people the freedom to hack the UI to their tastes and
not have to be locked into someone else's choice.  Design your own glade
even if you base it on one of the existing designs.

These are in SVN and are community property on an open license, but still I
think some consideration to the original author is warranted - at least if
they are still around.  It would seem kind of rude to up an mod the author's
favorite out from under him.  That's kind of why the AUTHORS file in the
glade directory ties layouts to authors.  There's an awful lot of hours
been put into them.

meflin.glade is Meflin's design.  It should not be messed with unless he goes
for it.  I did it up custom for him to his specifications, down to the bit
sizing of some of the defaults.

Other layouts might be other player's favorites, so making sweeping changes
seems to take some extra thought about who it affects.

Ok, so I've gone off a bit here here...  I'll stop.  Overall, I am probably
fine with removal from the glades that are not gtk-v1 strains, but I would
like some say in changes to my favorites... chthonic very definitely, and
sixforty because it is WIP with a lot of new ideas, and to some extent
caelestis as it was my first glade (though I see it as needing some
rethinking especially after addition of the message control system).

> 2) If it goes, should character, protections and core statistics move
> into the same pane? (so one pane with 3 tabs rather than 2 with 2 each)
> 3) Is there time for either or both of 1 and 2 to occur before the
> release of 1.50? (and when is that taking place?)

There's not a single answer.  Various layouts have various design goals,
though there is an element of experimentation in all of them.

> *The following are things I would like to add probably during the
> next release cycle or the one after:
> 
> Columns showing the keyboard shortcuts for the skills (ie looking for
> shortcuts for use_skill or ready_skill) - not sure whether those should
> be settable there, since that treads on the task of the Keybindings
> window a little

No problem with setup being able to be done there as long as it doesn't
preclude use of the keybinding dialog - at least until something better
comes along for other keybinds.
 
> icons for all of the skills (implemented by sending a face number along
> with the reply_info skill_info response)

Its fine, but I'd note that graphics blows out the vertical space
requirements, so some consideration might be given to allowing them to
be turned off or made quite small so as not to expand every line.  The
thing I like about your dialog now is that it is pretty tight and
readable.  I'm not really sure that I see a lot of value in skill
graphics myself except as eye candy - which tends to aggravate things
if you happen to have a really small monitor, but can be nice if you
have screen real estate to throw away.

> The 'character' pane on the client to show the icon for the skill
> rather than the name, and clicking on the icon to trigger the callback
> for the skill window in the same way that the menu entry does.

Not sure I know what you mean by the 'character' pane.  If you mean the
core stats, I'm not in favor of the text version of the readied skill
going away on all the layouts... but maybe I'm not groking what you want
to do.  Whereas some people like pictures, I tend to like text for that.
I think graphics are ok for fairly static things like button bars, but
I would find the readied skill shown only graphically to be frustrating.