> > > > > > 2: In the final paragraph of [1], a statement is made: "However, as > > Although I admit this is quite a complex sentence, so perhaps we > > just drop it all in favour of the earlier sentence "Implementations > > The problem with the older version is that it's obviously wrong at > the surface level of abstraction, even if it's correct at another. > If I have 2 STOP actions, they *are* semantically equivalent at a > high level - that's exactly what it means when they both say > type=STOP (so that's where the old stmt is wrong). Where it's right > is that they're *only* "as equivalent" as those types say they are, > and no more. Since types rarely talk about side effect, blah blah, > they're not semantically *identical* even if at some level of > abstraction they are semantically equivalent.
Trying "(However, as already stated, implementations **MUST NOT** assume that actions with exactly the same set of `rdf:type` values are identical - they are only as equivalent as those types define.)" > > > > > 11. In the "what are actions?" section [13] it now says "without > > OK so we flog this one a bit more. Maybe bullet list(s) or simple > table with "n2k/don't n2k" colunmns. Until we settle on something, > maybe a bit of red "we're still working on the wording here"? > There's enough threads flowing now that with everyone else back from > vacation I worry about losing things more. Heaven knows I have a > harder time concentrating. Taken it out into a separate section with 2 bullet lists: By using this explicit, loosely-coupled description of actions that are available on a given resource R, consumers _do not_ need to understand: * the `rdf:type` of R, and the domain that defines it * the native lifecycle of R * the properties of R and the meaning of their values instead, the consumer only needs to know: * the URI of R * how to find the actions and their bindings that exist on R (as defined by this specification) * how to execute those bindings (as defined by this specification) Therefore this specification frees the execution of these actions from domain specifics, instead providing a common point of interoperability between multiple domain specifications. Unless stated otherwise above: IBM United Kingdom Limited - Registered in England and Wales with number 741598. Registered office: PO Box 41, North Harbour, Portsmouth, Hampshire PO6 3AU
