On 5/18/05, Alex Lau <[EMAIL PROTECTED]> wrote: > Simon Poole wrote: > > > Denis Oliver Kropp wrote: > > > >> I'd like to use LiTE as a basis for the new AWT, so future > >> support for peer components is there, but Swing isn't using > >> them anyways, or does it now? > > AWT itself is bascially dead but its a huge amount of work to implement and the design has some intrinsic problems. Most J2SE apps use swing. ME is midp.
Swing itself is written on "primitive" awt bascially frames and graphics sort of like gdk. But J2SE allows swing and native components to be mixed which greatly increses the complexity of the implementation. My sugesstion for LiTE is to focus on a good java binding thats easy to use for the LiTE api and don't worry about the standards. If people then want to use the java api to implement awt go ahead but if you do a good job then LiTE can be used for AWT or SWT. There already exists a AWT on SWT binding so if you use your bindings to implement SWT you get AWT. But I've found that all the current public java gui api's AWT/Swing/SWThave major problems and are increadibly poor toolkits. So if you do a good job on your own api then I think it can easily become popular on its own merits. I've personally implemented AWT three times Swing twice MIDP twice and done serious work on SWT internals so trust me they aint the answer. Come up with something better. If its good enough the Linux distro's will include it so don't worry about not being bundled they would have to do the same thing if your providing the native layer for the current gui api's anyway. Don't think about pure java or wrapping native widgets or any of the other stupid ideas that shackled the current toolkits build a good one for programmers to use if you need native code for some of it write it in C if java works use java but above all build a good toolkit. Sorry for the rant but I've been hoping that the opensource community would wake up and deliver a good java gui toolkit. I see a chance here. > > > > Is no-one using SWT[1]? Supposed to be a big improvement over AWT, > > especially in terms of porting[2]. Allegedly used in embedded systems > > because of small size and efficiency, and can be used as the peer > > layer of an AWT implementation. Seems to me it would map extremely > > well onto DirectFB. > > > Yeah... I also get the similar idea... but the problem is SWT still not > in the standard library ...most applications > in the world ( that I know of ) are still using normal AWT ... Of couse > just by looking at eclipse we can see > how well SWT is ...but which way to choose... or may be let the directfb > extension work for the SWT project? > > > However, if no-one wants to use it then that's of no relevance! There > > are plenty of SWT vs AWT debates on the web[3], so best to go with > > what DirectFB developers would actually prefer to use I guess! > > > > [1] http://www.eclipse.org/swt/ > > [2] > > http://www.eclipse.org/articles/Article-SWT-Design-1/SWT-Design-1.html > > [3] http://www.google.com/search?q=swt+vs+awt&hl=en&lr=&start=10&sa=N > > > I think it is no need to debate .... just waste of time :) Pick one and > go :) IMHO > > -- > > Alex Lau (AvengerMoJo) > > ------------------------------------- > > Live or Die but Fight for It. > AvengerGear.com > > ------------------------------------- > > > _______________________________________________ > directfb-dev mailing list > [email protected] > http://mail.directfb.org/cgi-bin/mailman/listinfo/directfb-dev > _______________________________________________ directfb-dev mailing list [email protected] http://mail.directfb.org/cgi-bin/mailman/listinfo/directfb-dev
