Decoupling is just a means to get a particular improvement, not an
improvement by itself - but it is added complexity and such adds possible
cause of failure. So what will *this* decoupling actually provide *here*,
despite the problems I already mentioned in my last posting?
BTW, with "Most users" I wanted to say developers applying JAX-RS
technolgoy, not end users -- Do you really think that developers applying
JAX-RS to their application really would run so few tests to not detect
severe failures of the caching facility (we are talking about several
thousands of developers running thousands of tests against one singleton
implementation -- that's the benefit of having *no* pluggability)? And if
they really wouldn't, then pluggability wouldn't bring any benefit, as they
wouldn't see that they need to change the plugged part.
So again, what *actual* benefit does plugability provide *here*?
JPA 1.0 also came with default caching and I did not hear anybody
complaining about it being so hard to implement, and JPA 2.0 even added
configurability using a default API. I am using JPA for years (even 1.0) and
never had the wish for plugability of the cache. There are also squid,
memcached, ISA, and the caching in the browsers ... I do not believe that
all those are so buggy that we must provide pluggability or need to fear
that the world will explode. If that would be true, then caching would not
exists. This is ridiculous. http caching is nothing but a spec that needs to
get implemented, nothing else. If you really think Oracle is unable to do
this correctly, then nobody would ever be able to implement it at all. So
for what was it invented then?
Regards
Markus
> -----Original Message-----
> From: Julian Reschke [mailto:julian.reschke_at_gmx.de]
> Sent: Samstag, 7. Mai 2011 15:19
> To: jsr339-experts_at_jax-rs-spec.java.net
> Cc: Markus KARG
> Subject: [jsr339-experts] Re: [jax-rs-spec users] Re: Re: Proposal to
> downgrade [JAX_RS_SPEC-39] Client Cache Support to MINOR
>
> On 07.05.2011 15:02, Markus KARG wrote:
> > Julian,
> >
> > thanks for chiming in.
> >
> > What improvement would pluggability provide regarding the risks you
> > mentioned?
>
> Decoupling?
>
> > If it is non-pluggable and there is a bug in the cache
> implementation, most
> > users would notice it and so chances are good to get it reported and
> get it
> > fixed some day.
>
> "most users would notice it"? Not convinced :-)
>
> > ...
> > If it would be so hard to provide an implementation for the http
> Cache
> > paradigm, then there would be no squid and no ISA server.
> > ...
>
> I'm pretty sure these have bugs too, but what hasn't? :-)