Hi Al. -
I guess I agree that a simple peer-to-peer interface (basically just a socket connection with some protocol on top) is a special case. However a hub and a bus differ mostly in terms of efficiency, not semantics. The more general messaging infrastructures support all the options with various degrees of performance.
So, a critical agreement is on whether having *only* peer-to-peer is sufficient. Even if this agreed (I hope not, as producer/consumer messaging is fundamental, and be implemented without some infrastructure), then the interface should be formally defined so that this can be layered upon a more general implementation.
On Sat, 17 Feb 2007, Alasdair Allan wrote:
> Doug Tody wrote:
>> If we try to think about this a bit more formally, what we need to
>> define is merely an "interface" (e.g., in the sense of a modern
>> language such as Java etc.) which provides certain well-defined
>> operations with well-defined semantics. An "implementation" like
>> the PLASTIC hub merely implements this interface. So long as some
>> software implements the interface in a compliant fashion, it doesn't
>> matter how it is implemented.
>
> Well no. Are we going to use a hierarchical, peer-to-peer or hub based
> system? Before we decide on the message format, protocol for exchanging the
> messages and a transport standards for exchanging them over, then we really
> have to decide on what sort of topology we're talking about. There are
> fundamental differences between building a standard around an entirely
> peer-to-peer architecture, than building one which uses a central hub!
>
> Al.
>
Received on 2007-02-17Z01:45:40