Musings on an Erlang GUI System.

Daniel Dudley daniel.dudley@REDACTED
Mon Feb 17 20:51:40 CET 2003


Vlad Dumitrescu wrote:
> From: "Daniel Dudley" <daniel.dudley@REDACTED>
> > Vlad Dumitrescu (EAW) wrote:
> > > The reason why it is difficult is that an automatic
> > > engine has to assess how "beautiful" a layout would look to
> > > a person - I don't think this is researched enough...
> >
> > On the contrary, both Apple and Microsoft have done
> > extensive research on this. Both have publish their own
> > recommendations on UI design, somewhere on the web.
> 
> So that a *program* could look at a GUI description and validate
> it as "beautiful" or "ugly"? I don't think so, but if I am wrong
> I would be the first to be glad about it.

My point was that it is possible to automate most of the
recommendations found in those publications. The program
would then have CI/AI capabilities, much more so than the
current wave of "wizards" found in many GUI applications.

> > > I mean, it makes sense for text, documents, and such,
> > > but UI? Do we mean different things by UI? >>
> >
> > I've been asking myself that, too. Maybe the arguments are
> > directed towards the web, *most* of that stuff is text and
> > document orientated, ain't it?
> 
> That I overlooked; seen like that it make a lot more sense. With
> GUI in the general sense I mean even one for a CAD program, or
> for a game. I don't see how all those can be fit into the same
> framework.
 
I agree, but Bill(ion) Gates probably wouldn't. ;-)

Daniel




More information about the erlang-questions mailing list