[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [List Home]
Subject: Re: [xdi] Downsides of the notation shift
With “notation shift” we would be packaging the operation and the statement subject into one address. So the only way to separate the message operation and the statement which is its argument, would be to use a serialization which separates address parts, which currently only the unused PARSE and TREE serializations do. PARSE currently explodes addresses all the way down to context symbols and identifiers, but we could modify or fork it to only break up an address when it is a common prefix for more than one suffix within the message.If we simply special-case statements to serialize messages with the message’s statement as a unit, this is no different in practice from not doing the shift. What we have been calling “short notation” used to be called “statement root”. It was the XDI way of packaging a statement for reuse in larger structures. If we treat statements as a unit, we are inventing and using something equivalent. Another solution is to generalize from triples to quads or n-tuples, which creates slots for the message operation. This seems most straightforward to me. Joseph On Jun 17, 2014, at 1:17 AM, Markus Sabadello <markus.sabadello@xdi.org> wrote:
|
[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [List Home]