[rabbitmq-discuss] Impact of AMQP 1.0

[Jason J. W. Williams]

Sent via iPhone

Is your e-mail Premiere?

On Dec 14, 2010, at 0:40, Davorin Rusevljan <davorin.rusevljan at gmail.com> wrote:

> So in another words, risk as I see them is trying to cover too
> much ground, and in the end not covering any part of it well enough to
> produce solid _and_ truly interoperable standard for practical use
> cases.

I agree with this sentiment on 1.0. On a wire level I see where 1.0s changes will make gateways and interoperability mechanisms easier by making everyone a node. But the more important part I believe is keeping the current high level semantics the same...queues/exchanges/bindings etc.  It's the simplicity of these semantics that have made AMQP so successful and readily usable. I'd argue there are very few that should have to care or deeply understand the wire level. If you have to in order to relearn the high level semantics, I think that's a loss not a gain. 

-J