[prev in list] [next in list] [prev in thread] [next in thread] 

List:       kde-core-devel
Subject:    Re: the MICO/CORBA issue.
From:       Simon Hausmann <shaus () uermel ! Med ! Uni-Magdeburg ! DE>
Date:       1999-09-20 7:45:19
[Download RAW message or body]



On Mon, 20 Sep 1999, David Faure wrote:

> On Sun, Sep 19, 1999 at 01:46:26PM -0700, Kurt Granroth wrote:
> > Some things to note: IF Orbit had the features that we wanted AND it
> > was robust in those areas, then who cares if it is not fully 2.1
> > complient.  We aren't using 1/10 of all CORBA offers anyway.
> > 
> > I seem to remember only two "true" knocks against Orbit in the past --
> > it had no C++ binding and it did no error checking while marshalling
> > data.  I was under the assumption that both are now either fixed or
> > are in the process of being fixed.
> > 
> > Is there any other issue which shows a "broken" implementation or
> > displays a lack of robustness for what we need?
> > 
> > For the record, I don't think we should always dismiss Orbit.  From
> > what I can tell, it is incredibly efficient.  The fabled "tinyMico"
> > may or may not be as effecient... we don't know as it doesn't yet
> > exist!
> We're working on it ! It already exists !
> 
> No other ORB we can chose can be as easy-to-use as 'cute', the one
> we're creating right now (see kdelibs/corba/cuteidl), because none
> has direct marshalling code for Qt types.

Yes, cute is cute :-)

What makes me wonder about ORBit are the c++ bindings. How advanced and
complete are they? How efficient are they?

And most important: How difficult is it to integrate new marshallers into
ORBit and it's c++ mapping?

With MICO it's wonderful, I have to say that. The SII is really nice,
meaning it's very easy to extend it (assuming that you have access to the
idl compiler at the same time :-)

Ciao,
 Simon

[prev in list] [next in list] [prev in thread] [next in thread]