Learn how easy it is to sync an existing GitHub or Google Code repo to a SourceForge project! See Demo

Close

if SW and UPP get married

UIT
2007-05-14
2013-04-08
  • UIT
    UIT
    2007-05-14

    if smartwin++ and Ultimate++ get married,using SW'S policy based and UPP'S Multi-Platform,change something of UPP:macro and using XCB instead XLib

    that would be nice!

     
    • andrew7
      andrew7
      2007-06-19

      Bjarne Stroustrup in a recent paper..

      Given those general concerns about libraries, the answer to “Why didn’t we provide a GUI?” is obvious: The committee couldn’t do it. Even before tackling the GUI-specific design issues, the committee would have had to tackle concurrency and settle on a container design. In addition, many of the members were simply blindsided. GUI was seen as just another large and complex library that people — given dynamic_cast (§5.1.2) — could write themselves (in particular, that was my view). They did. The problem today is not that there is no C++ GUI library, but that there are on the order of 25 such libraries in use (e.g., Gtkmm [161], FLTK [156], SmartWin++ [159], MFC [158], WTL [160], vxWidgets (formerly wxWindows) [162], Qt [10]). The committee could have worked on a GUI library, worked on library facilities that could be used as a basis for GUI libraries, or worked on standard library interfaces to common GUI functionality. The latter two approaches might have yielded important results, but those paths weren’t taken and I don’t think that the committee then had the talents necessary for success in that direction. Instead, the committee worked hard on more elaborate interfaces to stream I/O. That was probably a dead end because the facilities for multiple character sets and locale dependencies were not primarily useful in the context of traditional data streams.