I don't know if we have the resources available to make every client/broker
implementation 0-9 compliant as part of M2.  I think we are far enough down
the process that we should stick to 0-8 for this release.

However, following the discussions that we've been having on this and the
rabbit list (we really should have some sort of amqp-users list to
cross-post this to)... I think we *really* need to implement a version of
our clients/brokers that speak 0-8/0-9(non-WIP)/Qpid-enhanced-0-[89].

Given the small magnitude of the changes involved - I don't think this is a
huge piece of work (from the Java side)... I can't really speak for the
other languages.

Cheers,
Rob

On 29/08/2007, Martin Ritchie <[EMAIL PROTECTED]> wrote:
>
> On 29/08/2007, Robert Greig <[EMAIL PROTECTED]> wrote:
> > On 29/08/2007, Martin Ritchie <[EMAIL PROTECTED]> wrote:
> >
> > > It ignores the bigger picture and this is where I think we need to
> > > take a moment. Interoperability was one of the highest goals for M2,
> > > but is interoperability with ourselves really enough? Given that our
> > > 0_8 spec is a hybrid with some 0_9 and 0_10 features can we really
> > > claim that this is an AMQP product?
> >
> > One question I have is this: how many of the changes we made in 0-8
> > went into 0-9? Why are we not 0-9 (non-WIP) for M2?
> >
> > RG
>
> Given that M2 has taken so long I can understand the desire to release
> it, but now that you state the obvious.. that we are not using 0-9 I
> have to wonder if we should re-scope M2.
>
> This would IMO work out nicely:
>
> M1 = AMQP 0_8
> M2 = AMQP 0_9
> M3 = AMQP 0_10
>
> The more we talk about this the less convinced I am that we should
> release M2 in the current state.
> --
> Martin Ritchie
>

Reply via email to