> -----Original Message-----
> From: [EMAIL PROTECTED]
> [mailto:[EMAIL PROTECTED]]On Behalf Of marc
> fleury
> Sent: Wednesday, June 13, 2001 2:23 PM
> To: [EMAIL PROTECTED]
> Subject: RE: [JBoss-dev] Should CacheKey copy its id on creation?
>
>
> |BTW, it's only a code bug in our application because we didn't know about
> |the Optimize flag. The spec states that all calls on home and remote
> |interfaces must copy their parameters. The Optimize flag,
> out-of-the-box,
> |is True, isn't it? Maybe "true" is not the correct
> out-of-the-box default
> |setting, but that's just my humble opinion.
>
> no it is right, it optimizes in VM comm and that is what people bench from
> the get go.
>
> We have been there before. It is a deviation from the spec but
> the spec is
> broken in many places, believe me, optimize false is the spec compliant
> version.
>
Off topic,
I'm very glad you take this kind of approach with the EJB spec. We tried to
add a fix to Jasper(Tomcat JSP compiler/engine) that was non-spec compliant
and were flatly denied, not because the fix wasn't a good(and right and
sensible) one, but because the fix was non-compliant. In other words, I'm
glad to see that you view JBoss as a Commercial Product, and not an
academic, proof-of-spec, one.
Bill
_______________________________________________
Jboss-development mailing list
[EMAIL PROTECTED]
http://lists.sourceforge.net/lists/listinfo/jboss-development
