Brad,

I'm sure we did have this conversation.  If anyone is in an infinite
loop, it's me :)  My only disagreement is that I find feel much harder
to address that look, or at least look _should_ be relatively trivial. 
Ideally, it should be pluggable, and one simply draws a little more to
obtain a richer look.  Feel involves routing of messages, managing
focus, etc.    IMHO, that is a lot more trouble, is it not??

Regardless of relative difficulties, I would be happy to help with an
effort to improve the situation.

Bill






==============
Brad Fuller wrote:

I'm recalling in my old mind that we had similar discussions a while
back. We
seem to have these conversations every once in a while.

Sorry if I repeated myself!


On Sat August 11 2007 4:27 pm, Brad Fuller wrote:
> Bill,
>
> I generally agree with you that out-of-the-box, Squeak's feel (and
look)
> are different than what users are used to. And this is a concern for
> deploying applications quickly. We have the unique opportunity to
change
> most of these "feel" issues to our heart's content. The 'look' is a
bit
> harder, but can be done, too.
>
> I don't worry too much about the 'feel' because it's easy to change.
>
> Having said that, I agree that it would be nice to have a consistent
feel
> out-of-the-box for Squeak that travels along all of the releases and
bug
> fixes.
>
> Maybe we should have a UI team that addresses that issue. That would
be one
> area that I could help on and would surely join the team. There is
> a "Morphic" team, but I don't think they are too active. I wouldn't
think
> the morphic team addresses the overall concept of Squeak's UI. It may
be a
> matter of setting some guidelines that can be followed that address we
we
> collectively agree is the "standard" feel of Squeak. Those guidelines
can
> be violated, of course, and should be when one wants to go beyond or
even
> invent a better method (which could then be incorporated into the
> guidelines.) But, if developers follow the guidelines, it would keep
Squeak
> UI consistent.
>
> Do guidelines, or some such list, exist today?
>
> brad
>
> On Sat August 11 2007 3:30 pm, Bill Schwab wrote:
> > Matthew,
> >
> > Fair enough. I am mostly thinking of my users. Suppose I display a
> > menu, they click in one of the magic spots, and instead of doing
> > something or nothing (which would cause a reflex to click again), it
> > gets attached to the hand. I can hear it now: the menu is "stuck to
the
> > mouse". The file/directory picking "dialogs" are not really that at
> > all. They are inconsistent, do not provide a clear way to show/hide
> > hidden files (at least on Linux). As long as Squeak has been in use,
it
> > seems that there should be a lot more polish in the interaction with
the
> > user.
> >
> > The behavior of input focus is a lot better than it once was, but it
is
> > still not consistent. I hate to think about putting a clerk in front
of
> > Squeak-based form. If they have to touch the mouse, the software is
> > broken. Why care what clerks think? They enter data that can be
turned
> > into serious money, but one has to make life easy for them, or they
find
> > ways not to cooperate. It can be hard enough when it is easy. Squeak
> > is starting to show some respect for tabbing, but it is again not
> > consistent. It might be far enough along that one could build
something
> > robust for end users. For example, in a deployed app, one would not
use
> > a system window; the main window would be app's shell (MDI fans will
see
> > it differently of course), and an alignment morph would likely cover
the
> > entire world, with the widgets living inside it. I have yet to
actually
> > do this, but I can imagine that it would hide many of the IDE's
> > annoyances.
> >
> > One of my favorites is the method finder. Especially with an optical
> > mouse (the kind that moves the cursor even when still), one has to
> > "balance" the cursor in the selector field, lest the focus fly off
to
> > some other widget.
> >
> > Workspace menus: the browse-it command should be on the first menu,
near
> > inspect, debug-it, and friends. Many other ergonomic annoyances have
> > been posted recently.
> >
> > The Linux vm will shutdown w/o warning. It could do a better job
with
> > virtual keys. Some of that is Linux culture, but I notice that other
> > apps respond as expected to keypad vk messages.
> >
> > Again, it is mostly feel: how it reacts to keyboard and mouse input.
I
> > am of the opinion that Microsoft is losing their collective grip on
> > reality, but they did some really good usability testing - what, 20
> > years ago?? Scary. Much of what they learned watching "idiots"
> > interact with computers has become widely adopted. If I am giving
them
> > too much credit, I apologize, but there is a mechanical vocabulary
of
> > interaction with computers, with a fine line between being
responsive
> > and fragile. I argue that Squeak as packaged is in the latter camp.
> >
> > Please note that I am trying to remove a barrier to use of
Smalltalk. I
> > believe there is nothing one can do to make the masses leave the
> > sharp-infested waters(TM) for the power and elegance of Smalltalk.
> > However, we can help those who "get it" make their would-be users'
life
> > as easy as possible, making it just that much easier to apply
Smalltalk.
> >
> > Bill
> >
> >
> >
> >
> > Matthew Fulmer wrote:
> >
> > On Sat, Aug 11, 2007 at 04:20:21PM -0400, Bill Schwab wrote:
> > > At the risk of becoming a broken record: my complaints about the
> >
> > Squeak
> >
> > > GUI are not about look, they are about feel. I can sell funny
looking,
> > > but I cannot sell clumsy.
> >
> > Squeak is easy to get used to, so we usually forget what makes
> > it clumsy. The only things I can think of is using the Alt
> > (rather than Ctrl) key for modifiers (on Linux and Windows), and
> > the lack of support for one-click copy/paste (under X11). What
> > else bothers you about it? We are not conspiring to make a
> > clumsy user interface. I got used to the interface after 1 week
> > and never saw it as clumsy.
> >
> > I want to know. Really. What don't you like?
> >
> >
> > Wilhelm K. Schwab, Ph.D.
> > University of Florida
> > Department of Anesthesiology
> > PO Box 100254
> > Gainesville, FL 32610-0254
> >
> > Email: bschwab at ...
> > Tel: (352) 846-1285
> > FAX: (352) 392-7029

Wilhelm K. Schwab, Ph.D.
University of Florida
Department of Anesthesiology
PO Box 100254
Gainesville, FL 32610-0254

Email: bschwab at anest.ufl.edu
Tel: (352) 846-1285
FAX: (352) 392-7029