[cap-talk] How desirable / feasible is a persistent OCAP language?

Jonathan S. Shapiro shap at eros-os.com
Fri Jul 25 15:17:10 CDT 2008

On Fri, 2008-07-25 at 21:44 +0200, Valerio Bellizzomi wrote:
> Experience with the windows registry shows that a system registry can grow
> up to many many megabytes, which affects directly the startup time and
> other registry operations.
> A database whould be a better option.

The registry *is* a database. It's just a very badly *implemented*

In short: a typical MS first try: get it into the field and deal with
performance only when you cannot stand the pain any more.

Which actually isn't a bad philosophy for production software if cost
management is a significant concern.


More information about the cap-talk mailing list