toad-devel
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

[Toad-devel] UI toolkit vs UI design


From: Ben Finney
Subject: [Toad-devel] UI toolkit vs UI design
Date: Fri, 4 Apr 2003 14:05:42 +1000
User-agent: Mutt/1.5.3i

On 03-Apr-2003, Janne Jalkanen wrote:
> > UI design is a prime concern for this project.  Choice of UI toolkit
> > based on "looks good" or "high performance (on some architectures)"
> > is not a concern at all.
> 
> Perhaps you're not understanding what I'm saying, or I am not
> understanding what you're saying - to me it seems that you just said
> two totally mutually exclusive sentences.
> 
> > Design of the UI has nothing to do with the toolkit beyond "can I
> > implement my good UI design with this toolkit?"
> 
> But "looks good" and "is fast" *are* important parts of a good UI
> design!  Not the *only* criteria, but *important* nonetheless.

Yes, all things being equal, naturally it's preferable to have
good-looking, fast-performing widgets than not.  However, I've seen too
many programs where the widget set -- chosen because it looks good and
whizzes fast -- has obviously dictated the design of the UI, to the
detriment of the whole program.

My point was to contrast the design of the UI -- which, as you
mentioned, can be done with tools as simple as pen and paper -- against
choosing the UI toolkit based on how nice the widgets look.  The UI
should be designed without any regard to which tooklit will implement
it.

Pretty widgets and poor UI design are depressingly common; likewise,
many programs with excellently-designed UIs have quite primitive-looking
widgets.

Good design on the UI is orthogonal to "the widgets look nice".  The
only time "what do the widgets look like?" is relevant to the UI design
is to ask "do the widgets implement the design?"

If the choice is between hacking the program to cope with the poor
programmability of the widget set, or throwing away the widget set and
choosing another, I would say the latter is the obvious choice, and
should be anticipated as a possibility from the outset.

Choosing the widget set *first*, based on prettiness of the widgets, is
putting the cart before the horse.

I would much prefer a widget set that does what is asked of it, over one
that looks great but is a monster to work with.  Therefore, UI design
comes before widget prettiness in my book.

> Please, I have done this for years =).

Your experience will be much appreciated.  Care to have a crack at it?

-- 
 \       "A man's only as old as the woman he feels."  -- Groucho Marx |
  `\                                                                   |
_o__)                                                                  |
address@hidden  F'print 9CFE12B0 791A4267 887F520C B7AC2E51 BD41714B




reply via email to

[Prev in Thread] Current Thread [Next in Thread]