adjust the related RfC "SystematicMetadata"

There is a long-standing RfC which basically describes the
same idea on a much wider, conceptual scope. Indeed I consider
this approach used here for solving the problem with GUI uptades
also as a proof of concept, to be expanded to a much wider scope
in case it works out well.

The new insight here is, that, by transferring a diff in pull mode,
we can circumvent the architectural problems with typing, which
showed up quite clearly in earlier design studies towards this
concept. The change from push to pull is by far not so fundamental
as it looks, since the sender still may initiate the exchange
by sending a message offering the diff iterator for the receiver
to pull. This way, we get a handshake and still sustain the
crucial part, which is to decouple the data representation
and give the receiver full control over the interpretation
of the exchanged data.
This commit is contained in:
Fischlurch 2015-03-22 01:37:16 +01:00
parent b051845835
commit f90e9c7bbd

View file

@ -33,11 +33,13 @@ to span the whole virtual model, and thus there never needs to be an universal
model data structure definition. Data structure becomes implementation detail. model data structure definition. Data structure becomes implementation detail.
Parts of the system talk to each other by _describing_ some subtree of metadata. Parts of the system talk to each other by _describing_ some subtree of metadata.
This description is _always pushed:_ the receiver implements an API allowing the This description is transferred _in the form of a tree diff:_ the receiver pulls
sender to navigate to some path-like scope and populate it with values, similar a sequence of diff verbs from a diff iterator, and interpreting these verbs will
to populating a filesystem. It is up to the receiver to assemble these information walk him down and expand the tree in question. Sub-scopes are ``opened'' and
into a suitable representation. Some receiver might invoke an object factory, while populated, similar to populating a filesystem. It is up to the receiver to
another serialises data into an external textual or binary representation. assemble these information into a suitable representation. Some receiver
might invoke an object factory, while another serialises data into an
external textual or binary representation.
Abstract Metadata Model Abstract Metadata Model