Has anyone implemented CORBA?

Andrew C. Greenberg werdna at gate.net
Thu Mar 9 15:05:16 UTC 2000


At 8:56 AM -0500 3/9/2000, Stephan B. Wessels wrote:
>Because Squeak needs to interact and CORBA is a very good normalizer, in my
>opinion.

I don't disagree with any of Stephan's remarks concerning the utility 
of Corba.  Nor do I disagree that Corba would be a useful project 
(for myself included).  I was quibbling with the use of the word 
"necessary," and asking why we should concern ourselves with the 
scope of adoption of Squeak in prioritizing issues for its 
development.

In short, I'm 100% behind these projects, but doubt that any one of 
them is "necessary" for adoption of Squeak, or that "adoption of 
Squeak" should necessarily drive our determinations upon which 
projects to work.  Let us focus on quality and depth of the things we 
do well, and build projects that matter to us.  In time, we will 
build a product that may or may not be adopted by others, but will be 
exceptionally useful to us.

The problem is not with the substance of the particular project, but 
rather with ALL arguments of the form "this is NECESSARY for 
ADOPTION."  Indeed all or most desired features might become 
priorities with this simple mantra -- I suggest that it is also a 
danger that leads to creeping featuritis.  In short, if an ORB is 
necessary for you, that reason enough to build an ORB for Squeak, and 
not because it is "necessary" to lead to wider "adoption."

If you would find an ORB useful, or are looking for a project to get 
involved, by all means work on one.  If not, there are many useful 
tasks still to be accomplished.





More information about the Squeak-dev mailing list