Status

Hi. Having bothered a lot of times with my speculations regarding
specifications and applications, this time I'll try to recap some status
and, by the way, to wish everyone a happy holidays...

Being just a bare Java / Web developer, mildly seasoned into the habitual
frameworks and stacks for such platforms, I can no more than say that I'm
obsessed with Semantic Web and, specifically, with its application in the
fields of integration and declarative applications development.

Sadly, for an aged, high school degree, bipolar disease diagnosed patient,
I currently cannot "make my point" in the documentation I do regarding my
work. I usually transcribe a couple of pages from my drafts weekly,
thinking I've put in clear text some "important concepts". Then, later, not
even myself can understand what I wrote. This "iterations" comes a long
long way ago (aprox. since 2004). Apologizes for that.

This "project" of mine, and all my attempts of specifying something may
seem delirious, including for myself. But I'd really be glad if the
opportunity comes and I'll be able to work on this kind of stuff seriously.
Let me say that I didn't find any local job position regarding this field.

That said, let me describe the sort of "solution" I do envision could be
built. Let me say I lack of the previous knowledge necessary for building
what my "intuition" thinks could be done (math, category theory, type
theory, etc.). So I really gonna need some help for validation (or
auditing) of whatever I've have done or to see if I'm in the correct path:

Semantic / Knowledge enabled ESB (Enterprise Service Bus) solution bus
addressing:

Plugin of diverse applications, services and data sources for the setting
of a syndication and virtualization layer via augmenting and matching
plugged contexts with each other's data, information and knowledge by means
of ontologies and semantics techniques.

Render a SOA (Service Oriented Architecture) aggregating data, schema and
behavior of plugged / connected sources in a declarative and discoverable
manner: REST HATEOAS[1], HAL[2].

Interactions involving more than one data source / backend (ERPs, CRMs,
SCMs, BI / BigData / ML Pipelines or in-house applications) data or updates
are view as a single operation over the corresponding consolidated entities
and the services architecture expose a composite of interrelated use cases
of the underlying aggregated back ends. Domains alignments may involve
check preconditions or performing actions involving a (distributed)
transaction.

Solution Semantic features:

Ontology Matching:
For the interaction and syndication of different systems is necessary to be
able to distinguish whether references to some data, schema or behavior
refers to the same "entity".

Alignment / Inference:
Once ontology matching aligns equivalent entities there left to regard
where, with some built "metamodel", which entities left to be "augmented"
via ontologies, semantics, aggregation and Machine Learning.

Activation / Dataflow:
Implied behaviors. Given some "metamodel", reactive behavior is encoded and
materialized according some sort of functional protocol interactions:
streams, request / response Dataflow[3].

So, for example, interactions in the (declaratively stated use cases)
exposed and consumed functionally, are driven by a series of "descriptors"
(as in [1] and [2]) maybe following a declarative approach of actors and
roles like in DCI (Data, Context and Interactions)[4]. Thus, source models
reflects this "facade" operations.

References:
[1]: https://en.m.wikipedia.org/wiki/HATEOAS
[2]: https://en.m.wikipedia.org/wiki/Hypertext_Application_Language
[3]: https://en.m.wikipedia.org/wiki/Dataflow
[4]: https://en.m.wikipedia.org/wiki/Data,_context_and_interaction

https://snxama.blogspot.com
https://github.com/snxama/scrapbook

Best Regards and Happy Holidays!
Sebastián.