matthiasm wrote: > FLTK 1.1 is ... final ... 1.1.8 ... wonderful ... incredibly stable ... > very fast ... still very fast. ... > > So what is next? ... > The bigger question for me right now is, where is FLTK heading?
Hi all, let me give an outside view from what i can see here. FLTK1 is obviously mature and very usable. There is a club of good and ambitious developers maintaining huge code bases af code using FLTK1. binary compatibility is a must here. On the other side is an evolved playground called FLTK2 that exists for years without any real background. I can notice a lot of ppl interested in FLTK2, mainly for UTF8, but also for a more powerfull and cleaner API. No killer-app yet. Up to now no one really uses FLTK, its labeled "dont use it", better use FLTK1. > Very few OpenSource projects survive ten years! Its of vital interest to attract new developers, who probably are more intersted in cutting edge technologie. Why should anyone start programming using GTK1? I would like to see a well designed, flexible and fast core library, implementing all the know how and experience of good programmers. Why not pushing 1.1.8 into maintainance mode (it does what it should) and concentrate on a really good redesign. This could mostly decided by all those people who did all the work for FLTK1. Thats where all the know how and experiance is! IMHO there has to be a stable core FLTK2 as Widget Toolkit allowing addons for Printing (does this affect the core implentation?), IO (pens / GL /..) and further optional components ( network/ wrappers for other libs like sound/video ...) Implementing library quality code is a much greater challenge than writing ordinary code. Designing a class library needs very much experince and doing this for multiple platforms is a really high target. I cant decide f.e why menues do call their callbacks or if it is better to leave this for the user, why the widget base class should manage childrens or if it is better to use a special group for this, what decicion has whitch advantage and drawbacks. Main interest should be here "fast and light" (i think all agree) So what is wrong with FLTK2 which is going to die by starvation (like dillo). As long as nobody is really interested and no one uses it, it will always be a toy. But if someone starts using some tools he also learns to live with all the quirks (like windows users have learned to get around common problems) > Splitting developers between 2.0 and 1.1 is a drag! Many GTK1 apps have been ported to GTK2. Qt breaks the API every new major release (if i know right) and i think the same is neccesary for FLTK to survive the next ten years. Enuff said Greetings _______________________________________________ fltk mailing list [email protected] http://lists.easysw.com/mailman/listinfo/fltk
