Hello Dean,

On Oct 11 05:10 94, Dean Roddey of 1:141/209 wrote:

 DR> will never work. If I have built a large system that uses, say, a 
 DR> standard mechanism for throwing errors as exceptions (an incredibly 
 DR> powerful system when used with forethought), it makes no sense for me to 
 DR> then try to use a bunch of modules that have no such concepts.

but you can use exception inside the implementation. And - how do you throw
an C++ exception from one machine to another one? My class implementation
runs under OS/2 on machine 1, but my clients (users) run on other machines
in the network. (even on other HW / OS platforms. eg. RS/6000 with AIX).

 DR> For large, multi-programmer system, the standardization of such 
 DR> mechanisms and styles is paramount for consistent, stable software 
 DR> development (particularly in our neck of the software woods which is 
 DR> clinical information systems where being wrong can mean someone being 
 DR> dead.)

we have similar requirements. My customers are organizations which built
eg. the controlling systems for power platforms. and doing something wrong
in an atom power plant means a lot of people being dead.

 DR> I see SOM as a great way to implement the little productivity and desktop 
 DR> goodies for OS/2 (as it is now), 

huh? whats the relation of SOM2/DSOM with WPS? ok, WPS is a SOM applikation
(under OS/2 2.x with SOM1, under 3.x with SOM2), but I'm using DSOM as an
industry standard CORBA ORB. (so i don't have to think about DCE, little
endian/big endian etc. problems).

 DR> but it will be a long time before serious system can be implemented under 
 DR> such a system.

mh. i would say controlling an power plant is serious.

 DR> Taligent
 DR> which DOES define many of the needed 
 DR> standards (though maybe not enough of them.)

talgient will not define ORB/persistent/... standards. these are defined by
OMG (CORBA ORB), ODBMG (OODB ODBG93), OSF (DCE). Taligent will have
implmentation of these standards, but so do SOM2.
Mario

---