[UI] Well, shall we do something then?

[Bill Schwab]

Gary,

IMHO, you hit it.  Presenters and event handling are the most important
features to get right.  Views are an active part of the framework, but
they are not something that should have to be created or changed in
ordinary development.  Others will probably disagree, but given text,
list (one and multiple column) and image presenters and buttons/menus, I
can do just about anything at the low-level.  Then one would want the
ability to create composites, including layout managers.

In ordinary Dolphin MVP work, the only time one messes with a view is to
create something that cannot be created with the existing views. 
Suppose one wants to produce a Gantt chart: I say create it as an image,
and present it with ImagePresenter.  It needs to be interactive?  Fine,
process events, and change the image, and invalidate the relevant part
of the view.  I have in past subclassed View for such things, and later
realized that I was missing the obvious.

So, while one could try to emulate the view protocol, I think it is not
strictly necessary, and probably not even a good idea.

Bill




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

>>> gazzaguru2 at btinternet.com 09/14/07 3:32 PM >>>
For Dolphin, without platform widget (specifically Win32) support, we'd
have
to emulate the api it does use too... Or are we just talking about
taking
its event/composition model?

> -----Original Message-----
> From: ui-bounces at lists.squeakfoundation.org 
> [mailto:ui-bounces at lists.squeakfoundation.org] On Behalf Of 
> Jason Johnson
> Sent: 14 September 2007 4:46 pm
> To: Squeak's User Interface
> Subject: Re: [UI] Well, shall we do something then?
> 
> 
> On 9/14/07, Bill Schwab <BSchwab at anest.ufl.edu> wrote:
> > Jason, Gary,
> >
> > Sorry for the delay in response.  I am a little nervous 
> about mining 
> > Dolphin, because it would be easy to trick ourselves into 
> thinking we 
> > have a cleanroom implementation.
> 
> Hadn't thought about that.
> 
> > The other Dolphin-related topic is the buyout effort.  If 
> we succeed 
> > with it, then Dolphin would be open, and one could simply port its 
> > framework to Squeak.  I dread putting the brakes on 
> enthusiasm, but in 
> > this case, it might be appropriate to do other things while 
> the buyout 
> > works its way forward.
> 
> Does this buyout realistically have a chance?  I mean don't 
> they want like half a million USD or so?
> 
> > With that said, how do you envision a rewrite working?  How would I 
> > ensure that I am not simply copying OA's work?  Typically, 
> I have seen 
> > this kind of thing done by two people, one or more 
> consultant/guides 
> > familiar with (aka tainted by knowledge of) the reference 
> system, and 
> > developers sworn to never look at the reference system.  
> The "blinded" 
> > developers write the actual code.  In that model, I would be a very 
> > good guide, but I know too much about Dolphin to write the 
> framework 
> > myself.
> >
> > Fair?
> >
> > Bill
> 
> Well, I hadn't thought about it honestly, but there are 
> several papers on the subject matter.  One from Dolphin at at 
> least one other one that predates the Dolphin effort.  I 
> think we could work from there, no? 
> _______________________________________________
> UI mailing list
> UI at lists.squeakfoundation.org 
> http://lists.squeakfoundation.org/mailman/listinfo/ui
> 

_______________________________________________
UI mailing list
UI at lists.squeakfoundation.org
http://lists.squeakfoundation.org/mailman/listinfo/ui