From: Beman Dawes (beman_at_[hidden])
Date: 1999-12-11 10:06:32
At 06:56 AM 12/11/99 -0500, Dave Abrahams wrote:
>My company is considering releasing a general-purpose,
>windowing toolkit to boost (open source, of course). It should serve
>basis on top of which a set of higher-level GUI tools could be
>its current state the code has Win32 and GGI (a low-level Linux
>library) implementations. All the basics are there for a lightweight
>1. Simple yet powerful cartesian coordinate support (Point,
>2. A coherent and easy-to-use imaging model (e.g. unlike Win32 GDI,
>no "selecting objects into DCs", resource management hidden behind
>scenes, or even any "state" in a drawing environment).
>3. Pixel maps and blitting with or without masks
>4. Minimal text support
>5. Rectangular windows with transparency (allows for any shape)
>6. Minimal color support
>7. Insulation from platform-specific header files (e.g. <windows.h>
>associated #defines stay out of the user naespace)
>8. An easy way to access platform-specific implementation details
>neccessary to support platform-specific idioms.
>9. Some preliminary event queueing support (needs work).
>The implementations are less interesting than the interfaces, which
>believe provide a powerful set of basics while remaining extensible
>committing too much in advance to ideas we are not sure may be
>supported on various platforms.
>There is still work to be done in many areas, however. For some of
>"know" what the answer is; for others I have no clue. In all areas I
>be open to input. Before we do anything else, though, I'd like to
>level of interest. Are there members of the boost community be
>developing and extending this work? We don't want to release the
>just to have it languish and die...
Are you talking about a full-fledged GUI toolkit, or a lower-level
windowing toolkit from which someone else could build a full-fledged
Sure, it would be wonderful to have a boost.org portable C++
full-fledged GUI toolkit. But lots of people have tried; see
www.geocities.com/SiliconValley/Vista/7184/guitool.html for a
listing. None have caught on. Why? My guess is that it is much
harder that it looks. And it looks hard enough at that. It would be
a very large project, and would roll on and on in time.
So while I am personally very interested in such a project, I also
think it needs to be approached very, very carefully.
Boost list run by bdawes at acm.org, gregod at cs.rpi.edu, cpdaniel at pacbell.net, john at johnmaddock.co.uk