I'd like to know just how many of you who wish to pattern the Gnome object model after
COM/Active-X/OLE/whatever Microsoft wishes to call it today have ever actually used
the system rather than just reading about it in a book. I've used ActiveX and it is quite,
quite awful. No, the books do make OLE sound great, but using it is a different story. Try

Look guys. This is a serious mistake. COM is NOT what you are looking for. Get yourself a
of Visual C++, and create an extensive COM/OLE/Active-X program. You won't for quite a
Linux and Unix in general attract programmers because, although at the surface are more
to use, are much easier on what is required from the programmer. COM requires you to learn
an entire
jungle of information, which will turn developers away from Gnome. Microsoft may be able to
unprogrammable systems, but we can't afford to.

Chris Knight

Miguel de Icaza wrote:

> Robert Ellsworth said:
> > I hate to ask this.  And perhaps I don't know enough.  But in light of the discussion
> > and sniping in the last few hours regarding object models for GNOME, it might be
> > worth posting some information...
> >
> > As I understand things from sources such as Stefan Westerfeld:  "The GNU Component
> > Model will not support KDE/Qt (the license is not acceptable)...
> I posted to the list that this is a missunderstanding from Stephan's
> point of view.  The libraries that make up the component model will be
> released under the GNU LGPL, which will allow linking with restrictive
> libraries, such as Qt.  Even usint it in proprietary software.
> > Meanwhile, you're doing a solution based on "OLE", instead of
> > 'reworking' the KOM in some way.  Is this the Microsoft OLE?  (If
> > not, what then?)
> It is modeled after OLE2.  You can not really implement OLE2 without
> providing a subset of the Win32 API, that is why it is being *modeled*
> after OLE2 (ie, the interfaces, the messages, the interactions).
> > Has the Evil Empire, then, put the OLE standards and all the
> > documentation thereto in the public domain, under the terms of a
> > GNU-acceptable license, and agreed that all further development of
> > OLE shall be completely open in the public domain, again as required
> > (imho) by the GNU requirements?
> They have done a pretty good job at publishing all of the inner
> workings in various books published both by them and by third
> parties.
> But your question regarding the license does not apply as we are not
> using any Microsoft code at all.  I am reading published specs by
> third parties.
> > This isn't about OLE's technical superiority to something like KOM.
> > On the other hand, it would seem that a lot more developers in KDE
> > are concerned with workable free software than is the case at
> > Micro$haft.
> Can you point exactly what is wrong with COM/OLE2/Active-X or are you
> just speaking out from an anti-microsoft point of view?
> Miguel.
> --
>          To unsubscribe: mail with
>                        "unsubscribe" as the Subject.

[Date Prev][Date Next]   [Thread Prev][Thread Next]   [Thread Index] [Date Index] [Author Index]