Re: [rgui-dev] Idea for binding to the low level api

[Tom Sawyer]

On Fri, 2002-08-30 at 04:16, Massimiliano Mirra wrote:
> <interface>
>   <section level="1">
>     <title>Blah</title>
> 
>     <switch id="SomethingOnOff" title="Something"></switch>
>   </section>
> </interface>
> 
> The <switch> element would represent a widget that returns a boolean
> value.  This can be the classic check button as well as the graphic,
> animated depiction of a real switch, depending maybe on a style sheet.
> By the same token, sections of level 1 can be a tabs in a tabbed
> window as well as windows in a MDI application.
> 

this is interesting. so instead of defining specifically the type of
widget you want, you'd simply define the "class of widgetability", then
a stylesheet would provide the reamining specifics. is that about right?

that's very cool. just by changing the stylesheet you get a whole new
layout, but the functionality would remain in essence the same. that's
what we're talking about here: a seperation of form and function, quite
distincly divided from one another. this could prove a very powerful
approach.

we'd have to give some deep thought to the exact nature of this
seperation to see just how feasible it would be.

~tom

signature.asc
Description: This is a digitally signed message part