> What I meant was: > what we'll answer when a user will ask > "Which is the Harbour's GUI lib?" > > Also in Java there are Swing, SWT and many others but if you buy a > book about Java you'll find that Swing IS the Java GUI for desktop > apps. >
Just this i was meaning. None of existing GUI can be thinked as THE Harbour GUI. I think that THE Harbour GUI would be OOP, multiplatform, well documented, well supported and in an active development and, mainly, aligned with the performances of Harbour and with the wishes of the Harbour developers. I don't see any OpenSource project having this characteristics, but only a lot of GUIs covering too few of this wanted features to be preferred on others. And i think that a GUI proposed by Harbour group as "official" could be only under GPL. > > A choice is very hard and will result to disagree the > developers that > > don't use the elected one. > > For elected I don't mean that everyone has to use it. > Simply that after saying "Harbour can create Terminal and GUI apps" we > need to add how it can do it. For terminals we've GT for GUI we have > ... and saying "there are a lot of them choose the one you like" is > good for skilled Harbour users but not for a new user that is > evaluating it. > Ok. Now the function of the "standard" GUI is clearer... An idea could be to have a demo, showing the common relevant features wanted by a GUI, builded with the best current libraries and self including the code needed to reach the functionality (a simple button "Show the code" triggering the listing of the current section). A not skilled Harbour developer could analyze the performances of the GUIs and the style and cost of coding, getting info about the library better matching his needs. > > In HwGUI, i.e., this isn't true even if a great effort was > done to obtain > > an high degree of compatibility between WinAPI and GTK. > > Only a project giving strong headlines, including the type of > > implementation and clear rules for syntax and behaviour of the code, > > could interest the community of programmers currently developing GUI > > Libraries, otherwise each developer will continue to talk > his dialect. > > This is why I think that xHGtk has more chances to survive than other > solutions. > Following closely Gtk standards they somewhat inherited the doc, the > tools and the "already proven" from it. But I agree with Viktor: OpenSource is a must, IMHO. > A complex GUI lib without a clean architecture, a tutorial and > complete doc is hard to pick up unless there is an existing code base > that require it. This is because i wrote "...but i hope that the option 3 could have a future". Best regards. Maurizio _______________________________________________ Harbour mailing list Harbour@harbour-project.org http://lists.harbour-project.org/mailman/listinfo/harbour
