> Ian, it is not difficult to do code statistics and see who contributed, > and how much. Check the svnstats project, or similar. > > I must say that, if You think people worked on FLTK 2 much less, You are > terribly wrong. No offense.
No offence taken - rather, I think you may have misunderstood the point I was trying to make. I do not question that a great quantity of work has been done on fltk2 - if we were to look at the stats we might well find that more change has occurred than in fltk1.1, and in a way that's actually the problem! Specifically, most of the change in fltk-1.1 has been a large number of small, localised changes that introduce fixes, with no regressions (I think there have been only a few minor regressions in the 1.1 tree in the last 5 years.) Most of the fltk-2 changes seem to appear as huge chunks, often with some associated API changes, and generally very many regressions. I have come to fear these changes, as I know that my code will all get broken. A few years back the whole fltk-2 API was bouncing back and forth as two of the key developers checked in their (differing) take on what it was supposed to be - my code was seriously broken for months due to that, and that is when I stopped working on the fltk-2 variants of may apps. Worse, on several occasions, I have seen bugs fixed, then the fix removed by some later set of changes. I can't work with that... > years to integrate patches into both branches. As stated above - I've seen some patches from fltk-1.1 worked into fltk-2, then removed by later fltk-2 updates - presumably because many of the fltk-2 updates appear as monolithic updates? > That is why we have > developers who have no idea that something advanced even exist in FLTK 2 I still keep a sandbox for playing with fltk-2, so I pretty much know what it can do. But I still use fltk-1.1 when I need to actually get the job done. -- Ian _______________________________________________ fltk mailing list [email protected] http://lists.easysw.com/mailman/listinfo/fltk
