RE: ILE vs CLI. was V5R3

["jt" <jt@xxxxxx>]

| -----Original Message-----
| [mailto:midrange-l-bounces@xxxxxxxxxxxx]On Behalf Of Hans Boldt


| Well, I'm the first to admit that I'm no expert on CLI. But from
| what I've read, I have some gut feelings on the subject. And here's
| what my gut tells me: If you're really trying to suggest that ILE
| should be more like CLI, I believe you understand neither CLI nor ILE.
|
| Apples and oranges. One is an environment for static binding. The
| other is yet another object integration framework.

Sometimes contention serves to keep one's thinking sharp, and helps develop
a common understanding...  I'm not even a novice at CLI but heard-a the
concepts, know about Java bytecode and have some fair bit-a familiarity in
ILE.

I dunno if I have all that much understanding of these issues, but I recall
Jon Paris' post a while back that the inter-operability provided by CLI was
NEVER a design goal of ILE.  (Don't have time to link to it, was about a
couple weeks ago, iirc.)  I believe that's what Hans means by "apples and
oranges".  But I can't help but notice, like Steve and perhaps others, that
ILE has been enhanced over the years (and probably much more solid than CLI,
because of that).

And ILE has come to provide the fundamental functionality of
inter-language-operability, regardless of whether that was a design goal or
not.  As far as the benefit of providing an exact duplicate of the
*proposed* "standard" of CLI, I dunno...  I'm hearing a lotta skeptics say
that this is just yet another attempt by MicroSoft to "dictate the
standards" and get developers "locked in the M$ trunk"...  I, personally,
have no opinions in this regard, but know it would necessarily divert a
lotta resources for ToroLabs to fully implement CLI...  And dunno that they
don't have a better starting point to implement this functionality with what
codebase they already have (but dunno they do either, 'course).